Kortlandt Beekes Armeniaca

$UPHQLDFD�

&RPSDUDWLYH�1RWHV�

by Frederik Kortlandt ��

with an appendix on the historical phonology of Classical Armenian

by Robert S. P. Beekes

CARAVAN BOOKS ANN ARBOR

7R�WKH�PHPRU\�RI�+ROJHU�3HGHUVHQ�

v

Contents Introduction� .....…………………………………………………………….. vii Notes on Armenian historical phonology I .……….……….……… ……… 1 A note on the Armenian palatalization …………………..…………………. 10 Albanian and Armenian ……………..…………………..………………… 13 Notes on Armenian historical phonology II: The second consonant shift .... 20 On the relative chronology of Armenian sound changes ..……..…………... 26 On the Armenian personal endings …………...……..…………………….. 34 Notes on Armenian historical phonology III: K- .…………………………... 39 Proto-Armenian case endings ………………………………...…………… 45 Demonstrative pronouns in [Balto-Slavic,] Armenian [and Tocharian] .,.... 52 PIE. *+- in Armenian …………...………………….. ………………….. 54 Notes on Armenian historical phonology IV ………………….........…….. 57 Arm. DUWDZVU ‘tear’ ………………….. ………………….. ……………….... 60 The syncretism of nominative and accusative singular in Armenian .……... 63 Armenian and Albanian ………………….....…………………..………..... 68 Notes on Armenian historical phonology V ……………………………...... 75 Sigmatic or root aorist? …………………..…………………..…………....... 79 The Thraco-Armenian consonant shift ……………..……………………... 83 The making of a puzzle …………………..…………………...……………. 88 Arm. FDQDZWµ ‘known’ ……………………………………......................... 96 Proto-Armenian numerals ……………..…………………..…………….... 98 Intervocalic *-Z- in Armenian …………………..…………………..……. 102 Palatalization of dentals in Armenian …..…………………..……………... 104 The sigmatic forms of the Armenian verb …………..………………...….. 107 The Proto-Armenian verbal system …………………..…………………… 110 Arm. FDáU ‘laughter’ …………………..…………………..……………… 117 Arm. Q U ‘sister-in-law’ ……………..…………………..………………. 120 The development of *\- in Armenian …………………..………………... 122 Arm. JRP ‘am’ …………………..…………………..…………………... 125 Armenian glottalization revisited ……..………………..……………….. 126 The Armenian causative …………………..………………..……………. 129 Arm. DULZQ�‘blood’ ………………..…………………..…………………. 131

vi &RQWHQWV

Appendix: Historical phonology of Classical Armenian by R.S.P. BEEKES … … … . 133 References .… … … … … … … ..… … … … … … … ..… … … … … … … … ..... 212 Indices ..… … … … … … … ..… … … … … … … ..… … … … … … … ..… … … . 226

,QWURGXFWLRQ vii

Introduction The main purpose of my work on Armenian is to bridge the gap in time and space between this language and its closest relatives. The emphasis is therefore on problems of relative chronology and the reconstruction of intermediate stages of development. While methodological issues are addressed on various occasions in the following pages, the general approach is strongly data-oriented. I have always been of the opinion that the challenge of counter-evidence is the driving force behind the growth of knowledge. A theoretical framework can often be helpful as a heuristic device, but the negative potential of aprioristic considerations must not be underestimated because theory can easily embody the reflection of rationalized prejudice. It is therefore important to reconsider time and again the primary data on which a reconstruction is based. The foundation for the work presented here was laid in the classic studies by Meillet (1936) and Pedersen (1905a, 1906). Following their example, I have limited the number of references to the minimum required by the context of the discussion. Since I started writing on Armenian, several introductions (e.g. Godel 1975, Schmitt 1981) and detailed investigations (e.g. Klingenschmitt 1982, Clackson 1994) have been published, which enables me to refer the interested reader to their bibliographies. Especially Clackson’ s list of references (almost 600 titles) is most helpful. The following studies are reprinted here in the order in which they were written. I have tried to eliminate inconsistencies by adding comments on those points where I have changed my opinion in the course of time. The resulting picture is one of a central Indo-European language with close ties to the Balkans. Most of the studies deal with phonological problems, especially segmental reordering of distinctive features. A development which is particularly characteristic of Armenian is the monophonemicization of features belonging to successive segments (p. 1ff.). The alleged development of *GZ > UN must be rejected (p. 88ff.). Other instances of segmental reordering of features are the palatalization of velars by a following front vowel (p. 10ff.) and the palatalization of dentals by a following *\ (p. 104ff.). A different type of reordering is involved in the rise of aspiration in voiced stops, the devoicing of voiced stops, and the voicing of voiceless stops in the Armenian dialects (p. 20ff.). It follows from the analysis that we must reconstruct glottalized stops for Proto-Armenian (p. 126ff.). An attempt at integrating the findings of earlier scholars into a coherent whole yields a relative chronology of Armenian sound changes from Proto-Indo-European times up to the classical language (p. 26ff.). This enables us to date

viii &RQWHQWV

the rise of K� from *V�, *S�, *+� and *\� in relation to other developments (p. 39ff.). I claim that *+ � � and *+ � � were preserved as K� before original *H but not before *R (p. 54ff.). The threefold color of initial laryngeals is preserved in their vocalic reflexes (p. 75ff.). Medial laryngeals were vocalized to �D� after syllabic resonants (p. 96ff.) and before consonant clusters, but not before single consonants (p. 120ff.). There is evidence for oralization of laryngeals after *X and depalatalization of palatovelars before *U (p. 27ff.). Intervocalic *Z yields �Z in absolute final position and �J� elsewhere (p. 102ff.). Initial *\� was lost in the classical language but is dialectally reflected as voiced aspiration (p. 122ff.). The establishment of a relative chronology enables us to derive Proto-Armenian from an Indo-European dialect which can be situated between Balto-Slavic in the north and Greek in the south. The monophonemicization of obstruents with a following *Z is similar to but not identical with the corresponding development in Albanian (p. 13ff.) and the same holds for the development of initial laryngeals (p. 68ff.). Perhaps the closest relative of Proto-Armenian was Thracian, which appears to have shared its earliest developments (p. 83ff.). It can therefore be suggested that the speakers of Proto-Armenian traveled east along the southern shore of the Black Sea. The relative chronology of sound changes also enables us to reconstruct the pre-apocope case endings (p. 45ff.). A question of particular interest is the merger of the nominative and accusative singular forms (p. 63ff.). The X�stems provide evidence for some highly archaic formations, as in the case of the words for ‘tear’ (p. 60ff.) and ‘laughter’ (p. 117ff.). The Armenian threefold demonstrative and anaphoric pronouns can be derived from a system with three uninflected deictic particles which can be reconstructed for Balto-Slavic (p. 52f.). The Proto-Armenian system of numerals was characterized by the existence of three different ablaut grades in the root, where the full grade was only preserved in the lower cardinals and largely replaced by the vocalism of the ordinals in the higher numerals (p. 98ff.). Like the pre-apocope case endings, the personal endings of the verb can be reconstructed against the background of the proposed relative chronology: the primary athematic and thematic endings in the present tense and the aorist subjunctive, the secondary active endings in the active aorist indicative, and the original middle endings in the middle aorist and imperfect (p. 34ff.). Sigmatic aorists were particularly frequent in Proto-Armenian (p. 79ff.). I argue that the expected marker of the sigmatic aorist in Armenian is �Fµ� if the language developed along similar lines as the other branches of the Indo-European family (p. 107ff.). More generally, I claim that the reconstruction of the morphological system should be largely independent of the establishment of etymologies and sound laws (p. 110ff.). The Armenian causative represents a nasal present derived from a sigmatic aorist stem (p. 129f.). The verb JRP ‘am’ , like XWHP ‘eat’ , must be derived from a prefixed formation (p. 125). The following studies cover most of the topics where I felt dissatisfied

,QWURGXFWLRQ ix

with earlier comparative treatments of the Armenian language. I am indebted to Professors Rob Beekes, Sasha Lubotsky, Kees Ruijgh and Jos Weitenberg for comments at various stages of my work. I thank Leonid Kulikov and Ilona Manevskaia for editing the present volume. It goes without saying that they cannot be held responsible for the views put forward below.

1

Notes on Armenian historical phonology I*

It has long been recognized that an exhaustive diachronic description of a language must include a statement of the historical order in which various developments took place. Such a statement is of particular importance when the attested daughter language differs considerably from the reconstructed mother language, as is the case with Armenian. Indeed, quite a few valuable indications about the relative chronology of sound laws can be found in the classical studies on Armenian historical phonology by Meillet and Pedersen. Not all of these statements are compatible, however. The first attempt, to my knowledge, to formulate a consistent series of successive developments was presented in Zabrocki 1951. In a recent study DåDXNMDQ� VWDWHV� WZHOYH� UXOHV� RI� UHODWLYH�chronology (1967: 318f.), not all of which can be maintained. In the following “Notes” I intend to make a contribution to the solution of the chronological problem.1

1. *NK�>�[ If the connection of [DFDQHP, FµD[, V[DOLP with Skt.Û LI EBUJ Û LI � TLIBMBUJ, Lith. NiQGX,�ãDNj is correct, there are two possibilities. Either the development *NK > [ preceded the rise of Nµ from *N as a result of the Armenian consonant shift (cf. Zabrocki 1951: 146), or *NK was a cluster which remained distinct from Nµ after the latter development (cf. Pedersen 1906: 334 [= Kl.S., 112]). The choice between these two possibilities will be made below.

2. *WK�>�Wµ If the connection of RUWµ with Skt. Q UIVLB , Gr. %²'-�+ is correct, the final

* Reprinted from 6WXGLD�&DXFDVLFD 3 (1976): 91-100. 1. The transcription of Armenian used here differs from the one in Meillet’s (VTXLVVH�in one respect: I write Fµ and þµ instead of the corresponding dotted letters. This is not only preferable for typographical reasons and for the sake of consistency, but especially because dotted letters are used to denote glottalization in Georgian, velarization in Arabic, retroflexion in Sanskrit, and various articulatory peculiarities in other languages.

[92]

2 )UHGHULN�.RUWODQGW�

[93]

consonant has escaped the regular voicing after a resonant, e.g. in DUG� Skt. UÓ � Gr. w'-E+ (cf. Meillet 1936: 36). Consequently, *WK must have remained distinct from *W up to a period posterior to the time when the voicing rule operated in the latter word, which is in turn posterior to the Armenian consonant shift (cf. Pedersen 1905a: 202 [= Kl.S., 64]). If *WK goes back to a sequence of *W and laryngeal, as was first suggested by de Saussure in 1891 in connection with Skt. Q UIÓ �(1892a: cxviii [= 1922: 603]), it follows that the laryngeal was preserved up to that time in the position under consideration. Both of the possibilities mentioned in the preceding paragraph are compatible with this result.

3. *SK�>�Sµ

If the connection of SµD\O with Skt. QIBMHÓ , Latv. VSXOJRW, of TQµJ Nµ��SµDUDW with Skt. TQIVSºUJ, Lith. VSuUWL, and of SµXO with Lith. S�OWL,�S~ROX, OHG. IDOODQ is correct, we have to assume that *SK was a cluster which remained unaffected by the radical transformation of *S into K�(initially) or Z (medially) as a result of the consonant shift, fricativization, and loss of consonantal articulation. There is no reason to separate the developments *WK�>�Wµ and *SK�>�Sµ chronologically.

4. *V�>�K The development *V� >� K belongs to the earliest stages in the history of Armenian. It is anterior to the development * > V� which in turn preceded the Armenian consonant shift and the sonorization of occlusives and affricates after UHVRQDQWV��FI��'åDXNMDQ��,W�ZDV�DOVR�DQWHULRU�WR�WKH�ORVV�RI� Q before V in XV, DPLV, Skt. º TB ��Lat. N OTJT� and to the loss of *S before V in VXW, HUHV, Gr. 5Jº #+, %'z%7, where the sibilant has been preserved (cf. Meillet 1936: 39). The only development which can be demonstrated to have preceded the loss of buccal articulation in *V is the assimilation of the initial sound in VNHVXU, Skt. Wº VSB , Lith. u uVSBT (cf. Meillet 1922: 88). This word will be discussed

below.

5. *Z�>�J The development *Z� >� J offers more chronological problems than any other change. First, it is hard to date this development either before or after the FRQVRQDQW� VKLIW�� 'åDXNMDQ� KROGV� WKH� ODWWHU� YLHZ� EHFDXVH� RI� WKH� YRLFHOHVV�occlusive in NF L, which goes back to a cluster *GZ through an intermediate stage *WZ (1967: 319). Winter also assumes that after the shift *G� >� *W,� *Z became *J and was assimilated to the preceding consonant in this word (1962:

1RWHV�RQ�$UPHQLDQ�KLVWRULFDO�SKRQRORJ\�,� 3

262). The argument is not cogent because it is equally possible to assume an intermediate stage *GJ with subsequent unvoicing as a result of the shift. Meillet regards the U in HUNX,� HUNDU as the remnant of an initial *G and concludes that “l'altération du groupe GZ est donc antérieure à la mutation consonantique arménienne” (1936: 51). If this is correct, the aspiration of Nµ from *VZ�in NµXQ��NµR\U can be attributed to the consonant shift after the unvoicing of *J from *Z in the position after *K from *V��It cannot be correct, however, because in that case one would expect *N instead of J from *Z in other positions. If the development *Z�>� J�preceded the consonant shift, it remains unclear why the latter did not affect the outcome of the former. It follows that the reflexes of *GZ�and *VZ must be explained differently. Second, the development *Z�>�J must be dated in relation to the loss of final syllabes. Meillet states that intervocalic *Z was not affected by the change and adduces the words WLZ and NRY in support of his view (1936: 50). These examples presuppose that the change was anterior to the loss of final syllabes. Indeed, the development *Z > J�in NRJL, Skt. HºWZB , requires this chronology if Meillet's statement about intervocalic *Z is correct. In WD\JU, Skt. EFW ,�Meillet attributes the development *Z�!�J to the following U� but this requires the inverse chronology. In ORJDQDP, Gr. �#E7, one cannot attribute the development to the following Q without discussing the chronology of the suffix. Meillet admits that “le détail des conditions ne se laisse pas déterminer”, but this is certainly because he posited the wrong development for intervocalic *Z� Pedersen pointed out that intervocalic *Z is reflected as J unless it became word-final as a result of the loss of final syllables (1905a: 196 [= Kl.S., 58]).2 This is by far the most straightforward explanation of the contrast between NRY, DUHZ and NRJL,�DUHJDNQ� Grammont correctly infers that “ce changement de Z en J est postérieur à la chute des voyelles de syllabe final” (1918: 225).3 Third, we have to date the development *Z� >� J in relation to the palatalization before front vowels and to the metathesis of clusters containing a resonant as their second member. Meillet remarks that “Arm. JH� dans les mots originaux ne peut représenter que *ZH�, puisque l'ancien *JKH� aboutit à H-” (1936: 50). It follows that the palatalization was anterior to the rise of J from

2��=DEURFNL�LQFRUUHFWO\�FRQFOXGHV�WKDW�³3HGHUVHQ�VIRUPXáRZDO�SURFHV�SU]HFKRG]HQLD�Z w J� Z� WHQ� VSRVyE�� L � XVLOQLHQLX� XOHJD� ND GH� Z� ]� Z\M WNLHP� LQWHUZRN>D@OLF]QHJR�� NWyUH�Sy QLHM� GRVWDáR� VL � GR� Z\JáRVX´� �� )LUVWO\�� 3HGHUVHQ� H[SOicitly denies the “ usilnienie” of *Z after obstruents (1905a: 197 [= Kl.S., 59]). Secondly, he does not give D�FKURQRORJLFDO�VWDWHPHQW��7KH�ZRUG�³Sy QLHM´�LV�RI�FRXUVH�RXW�RI�SODFH�KHUH� 3. Winter states in connection with the word LQQ, Gr. y��z�, that “ *-Z9��>��J9� must be more recent than the change of *�DQ in final position to �Q” (1966: 203). In fact, there is no reason to assume that the word ended in **�DQ at any stage of its development in Armenian.

[94]


[95]

*Z.4 Winter states that “ the change of antevocalic *Z to J had already taken place at the time of the metathesis” because the latter phenomenon is absent in the plural of the word FXQU, Gr. �²�/ (1962: 262). This relative chronology presupposes that the palatalization preceded the metathesis. Indeed, Pedersen comes to the conclusion that the palatalization “ ist älter als die entwickelung eines J aus idg. Y, eines Nµ aus idg. VY oder WY, eines N aus idg. GY� eines �UN� aus idg. �JU�; denn ein solches sekundäres J, Nµ, N, UN ist niemals palatalisirt worden” (1906: 394 [= Kl.S., 172]). D aukjan holds the opposite view and states that the palatalization is posterior to the metathesis because it affected the cluster in B BNV ,�with � from *�JKO\� (1967: 171). This word can hardly be separated from Gr. w3� +, OPr. DJOR (cf. Solta 1960: 281), in spite of Adjarian's objections. If this is correct, the palatalizing effect of a following \�must have remained operative during a longer period than that of a following front vowel. Fourth, the development *Z�>�J must be dated in relation to the rise of intervocalic Z from labial occlusives. Meillet's contention that Z is the regular reflex of intervocalic *Z is based on his assumption that both the first and the second element of the expression JHU�L�YHUR\�correspond etymologically to Skt. VQºSJ, Gr. Á%z', with elimination of the initial vowel. This equation requires that the development of intervocalic *S�� *EK into Z was anterior to the development *Z�>�J, and that the latter preceded the loss of final syllabes. As we have seen above, however, the material is better accounted for if we assume that the development *Z�>�J affected both initial and intervocalic *Z�and was posterior to the loss of final syllables. In that case it must have preceded the rise of Z from intervocalic *S��*EK and the etymology of JHU must be abandoned. Indeed, Solta mentions only L�YHU, L�YHUR\, L�YHUD\ in connection with Skt. VQºSJ and Gr. Á%z' (1960: 274). At this point there arises a chronological difficulty. The rise of \ from intervocalic *W in KD\U� Gr. %�-=', must have been posterior to the loss of intervocalic *\ in HUHNµ, Skt. USºZB , but it must have preceded the loss of final syllables because of the genitive KDZU� Gr. %�-'²+. It is here that the need to separate the development *Z� >� J into two distinct chronological layers arises. It is probable that the rise of buccal features in this sound and the transformation of the resonant into a resonantal feature date from the earliest stages of the Armenian language, whereas the transformation of the sound into an occlusive and the loss of the labialization feature are relatively recent. This state of affairs is perfectly compatible with the hypothesis that Georgian �YLQR was borrowed from Armenian. Thus, the above statements on relative chronology must be adjusted in the sense that the first phase of the development *Z�>�J was anterior and the second phase posterior to the changes in relation to which an ordering could be formulated. 4. The inverse chronology, which is suggested by DåDXNMDQ� �� FDQQRW� EH�accepted.


It follows from the above rules that the lenition of intervocalic and pre-consonantal occlusives was anterior to the metathesis. This is confirmed by the word DUDZU, Gr. x'#-'#�, not **DUDUG��FI��'åDXNMDQ��0RUHRYHU��WKH�word JD\O, Skt. W LB , shows that the rise of DO from syllabic *O must be dated between these two developments (cf. Winter 1962: 261). There is no reason to connect this word with Georgian PJHOL, which does not explain the vocalism.

6. *VZ�>�Nµ

The development *VZ�>�Nµ is well established, e.g. NµR\U��NµXQ��NµLUWQ��Skt. TWºT ,�TWºQOB ÛTWÂEB � Lith. VHVX},�VmSQDV, Latv. VYLHGUL� Scholars disagree about the intermediate stages, however. Pedersen remarks that the development *Z�>�J did not take place in the position after an obstruent and regards the velar element in Nµ as the continuation of the laryngeal feature evolving from the first component of the cluster *VZ, which passed “ durch die entwickelungsreihe KY� NµY� Nµ� ” (1905a: 197 [= Kl.S., 59]). Grammont considers the velar element the continuation of the resonant, assuming that the language “ a interverti *KZ en *ZK assez tôt pour que le *Z devenu initial de syllabe pût encore passer à *J� ce *J�placé devant l'aspiration sourde K est devenu sourd lui-même” (1918: 251). Meillet accepts the development *Z� >� J after an obstruent and ascribes the unvoicing and aspiration to the latter: “ ainsi *VZ�, devenu *KZ�, donne, avec assourdissement du Z par K, arm. Nµ ” (1936: 50). Zabrocki states that the cluster *VZ “ przeszáD�QDMSU]yG�Z�KJY��1DVW SQLH�Z�SRVWDü�KJ�>�KN�>�NK�>�Nµ ” (1951: 156). I think that we can avoid both Meillet's assumption that the change *Z�>�J�took place after an obstruent and Pedersen's hypothesis that the velar element continues the first component of the cluster *VZ and still need not have recourse to the metathesis proposed by Grammont if only the development is considered in terms of features. The need for a metathesis disappears if we assume that the cluster *VZ became monophonemic before the rise of the velar articulation. If this is correct, there was an intermediary stage between the developments *V�>�K�and *Z�>�J, characterized by an opposition comparable to English ZKLFK, ZKLOH vs. ZLWFK, ZLOH. The subsequent evolution of this voiceless aspirated labialized sound was wholly analogous to the development *Z > J.5 5. Winter’s laryngealist explanation of the initial occlusive in NµVDQ (1965: 106) is very attractive because the sound cannot be explained otherwise. It raises a problem in connection with JR\, Skt. YiVDWL� however. There is no need to assume a laryngeal in JRJ, Gr. J¾3#µ�� (cf. Beekes 1969: 90).

[96]


[97]

7. * Z�>�ã The development * Z�> ã is established in ãXQ, ãXQþµ,� ã� Skt�� W � WºTJUJ Ûº WB , Lith. ãX}, ãYDxNãWL, Lat. HTXXV (cf. Pedersen 1905a: 197f. [= Kl.S., 59]). Meillet regards VN as the normal reflex of * Z�on the basis of VNXQG�and VNHVXU (1936: 50f.). The former word is better compared with Russ. ãþHQyN and the latter does not contain an original cluster * Z, as Gr. x�KUR and OHG. VZHKXU show. The corresponding cluster * IX� lost its labial element in MD\Q, Russ. ]YRQ. Three problems remain: the change in the place of articulation resulting in V� the absence of the same change in its voiced counterpart, and the origin of the initial cluster in VNHVXU. Pedersen’s hypothesis of a prothetic V in the latter word cannot be accepted. The comparison with Skt. Wº VSB � and Lith. ã ãXUDV makes it probable that in Armenian, too, the initial fricative was assimilated to the reflex of medial * .6 This assimilation must have preceded the development *V�>�K, but was probably posterior to the assibilation of the intervocalic palatal consonant. Since the initial cluster in VNHVXU differs from the reflex of * Z, I wonder if the assimilation can be dated in a period when the latter had already become monophonemic, i.e. at a time when *Z had already become a feature of the preceding occlusive in the words which show ã. In that case we have to assume *F � Q, *HF � RV next to * XF VS at the stage immediately preceding the development *V�>�K. After the latter change, which entailed the rise of *K �

from original *VZ (cf. the preceding section) and which makes the diacritic of * superfluous, *F �

was assibilated to *V � and *M �

lost its labialization. At this time voiced fricatives did not yet exist. When the consonant shift turned the aspirated occlusives into fricatives, *K �

and *Z became *[ � and *�

�, respectively, and the

reflex of *Z in *VZHVXUD became the archiphoneme of these two sounds. This relationship was maintained until after the shortening of the fricatives to occlusives and the loss of labialization. The latter development yielded ã from *V �

because of the retracted position of the tongue which the labialized fricative requires.7 If this suggestion is correct, the opposition between VNHVXU and V[DOLP leads us to decide on the choice left open in section 1 above in favour of the second alternative, viz. that the cluster *NK was preserved up to a period after the rise of Nµ.

6. Solta incorrectly states that the assimilation characterizes Armenian and Indo-Iranian in contrast to the European languages (1960: 57f.). In fact, Avestan shows the reflex of initial *V, whereas Lithuanian took part in the change. It cannot be determined to what extent Slavic shared the development. 7. There is no evidence for Grammont's view that “ *FZ est devenu ã-, par l1intermédiaire de *F\; la sonante *Z s'est assimilée à la prépalatale *F en ce qui concerne le point d'articulation, et par suite est devenue la sonante prépalatale \” (1918: 252).

[97]


8. *WZ�>�Nµ The development *WZ > Nµ� is attested in NµH], Skt. UW N. Here again, scholars disagree about the intermediate stages. The sequence *WZ�> *WµZ�>�*NµZ�>�Nµ is supported by Pedersen (1905a: 197f. [= Kl.S., 59]) and Grammont (1918: 251f.), whereas Zabrocki assumes a sequence *WZ >*WµJZ� >� *WµJ� >� *WµN� >� *WNµ� >� Nµ (1951: 157). The latter view is based on Meillet’s statement that the development *Z >� J affects post-consonantal *Z as well and that “ certaines consonnes précédentes perdent leur point d'articulation propre, mais toutes conservent leur caractère de sourde ou de sonore, d'aspirée ou de non aspirée qui est attribué à la gutturale” (1936: 50). Here again we can assume that *WZ became monophonemic at an early stage, and that the resulting *W �

coincided with *N �

before the loss of the labialization. It is probable that *W � had already been

eliminated at the time of the consonant shift. The expected development after *V is found in RVNU, Gr. ±)-z#�.

9. *GZ�>�N

The development *GZ� >� N� is attested in NF L� Skt. N EÓ Û N EW . It is astonishing to see how Meillet's supposition that initial *GZ�yielded UN� which is based on HUNX, Skt. EW and HUNQþµLP, Gr. z#+, is repeated over and over again more than half a century after Pedersen's fully convincing demonstration that this identification cannot be upheld (1906: 398f. [= Kl.S., 176f.]). The initial part of HUNX must have been taken from HUHNµ because it is lacking in other words derived from the same root. The word HUNQþµLP is best compared with OHG. IXUKWHQ� It is probable that the words HUN, HUNQ, HUNDU also belong to this root (cf. Pedersen 1906: 479 [= Kl.S., 257] on the suffix -DU).8 If the development of *GZ proceeded along the same lines as the changes discussed in the preceding sections, we can assume that the cluster became monophonemic at an early stage and that the resulting *G �

merged with *J � before the loss of

labialization, and probably before the fricativization which started the consonant shift.

10. *LZ > X Godel writes (1970: 146): “ In view of such regular paradigms as JLU� ‘letter, writing’, G. JUR\;� PDUPLQ ‘body’, G. PDUPQR\� etc., one would expect the 8. The chronological difficulty inherent in Meillet’s assumption that U continues the voicedness of the initial *G was indicated above. Being aware of the problem, Grammont suggests that U continues the following *Z and that the resulting cluster *NU� “ a subi l'interversion habituelle en �UN�´ (1918: 252).

[98]


genitive forms of WµLZ ‘number, figure’ , DQLZ ‘wheel’ to be *WµZR\, *DQZR\� Instead, we find WµXR\, DQXR\� Not that Z clusters were avoided; such clusters actually occur in the inflection of polysyllabic nouns ending in -L (< *�L\R-):�RUGL ‘son’ , G. RUGZR\�� Wµ]HQL ‘olive-tree’ [sic], G. Wµ]HQZR\� etc. Since the latter genitive forms have admittedly developed from *�LZR\� (*RUGLR\�>�*RUGLZR\�>�RUGZR\), the contrast with DQXR\��*DQLZR\ is amazing.” It is evident that the endings �ZR\ and �XR\ cannot both go back to *-LZR\ because in that case the phonemic split would remain unexplained. Indeed, it is only the latter ending for which the reconstruction can be made plausible. In the inherited formative *�L\R� the *\ was non-phonemic. As a result of the palatalization in P , Skt. NºEIZB , the opposition between *L and *\ after an obstruent ceased to exist, so that after this change the *L in *RUGLR\ was phonemically neither a vowel nor a consonant (cf. Kortlandt 1972: 143f.). This situation remained unchanged when the assimilation to the following vowel yielded the form RUGZR\� The reconstruction **RUGLZR\ is incorrect because the *L and the Z never coexisted at the same time. On the other hand, the weakening of the prefinal syllable in *DQLZR\ resulted in the loss of the palatal element and subsequently in the merger of the vocalic feature of the *L with the resonantal feature of the *Z into a single phonemic unit, yielding the form DQXR\� This merger introduced the opposition between X and Z after a consonant. The developments *RUGLR\�>� RUGZR\ and *DQLZR\�>�DQXR\ are, in a sense, variants of the same phenomenon and need not be separated chronologically.

GLOSSARY ARMENIAN B BNV darkness, DPLV month, DQLZ wheel, DUDZU plough, DUG shape, DUHJDNQ sun, DUHZ sun, JD\O wolf, JHU�L�YHUR\ above all, JLU writing, JRJ I say, JR\� being, HUHV face, HUHNµ three, HUN labour, HUNDU long, HUNQ pains, HUNQþµLP I fear, HUNX� two, ã donkey, Wµ]HQL fig-tree, WµLZ number, LQQ nine, ORJDQDP I wash, [DFDQHP I bite, FXQU knee, NRJL butter, NRY cow, KD\U� KDZU father (nom. gen.), MD\Q voice, PDUPLQ body, NF L soft, N Ûmiddle, ãXQ dog, ãXQþµ breath, RVNU bone, RUGL son, RUWµ calf, XV shoulder, V[DOLP I fail, VNHVXU mother-in-law, VNXQG little dog, VXW false, VSµL Nµ dispersion, YHU� above, WD\JU brother-in-law, WLZ day, FµD[ branch, SµD\O lustre, SµDUDW scattered, SµXO�fall, NµH] thou (acc.), NµLUWQ sweat, NµR\U sister, NµXQ sleep, NµVDQ twenty. SANSKRIT �B TB shoulder, B WB horse, XSDUL above, UV season, LI EBUJÛchews, HBWZB bovine, USBZB three, UW N thou (acc.), EFW brother-in-law, EW Ûtwo, Q UIVLB �young animal, Q UIV broad, QIBMHV reddish, NBEIZB middle, N EV � N EW soft (masc. fem.), YDVDWL dwells, W LB wolf, LI branch,

[99]


[100]

WB VSB father-in-law, WBTJUJ blows, W dog, VNKDODWL stumbles, VSKXUDWL quivers, TWBQOB sleep, TWBT sister, TWFEB �sweat. GREEK x'#-'#� plough, w'-E+ arrangement, w3�S+ mist, �²�/ knee, z#+ fear, x�/'' mother-in-law, y��z� nine, J¾3#µ�� I pray, �#E7 I wash, ±)-z#� bone, %�-=' %�-'²+ father (nom. gen.), %²'-�+ calf, %'z%7 I appear, Á%z' over, 5Jº #+ lie. LATIN HTXXV horse, PHQVLV month. OLD HIGH GERMAN �IDOODQ to fall, IXUKWHQ to fear, VZHKXU father-in-law. LITHUANIAN �NDQGX I bite, SXOWL to fall, SXROX I fall, VDSQDV dream, VHVXR sister, VSLUWL�to kick, ãDND branch, ã ãXUDV father-in-law, ãXR dog, ãYDQNãWL to wheeze. LATVIAN �VSXOJRW to glitter, VYLHGUL sweat. OLD PRUSSIAN DJOR�rain. RUSSIAN ]YRQ ringing, ãþHQRN puppy. GEORGIAN PJHOL wolf, �YLQR wine. [For an elaborate relative chronology of sound changes see Kortlandt 1980b [this vol., 26ff.]. For *GZ > N, not UN, see Kortlandt 1989 [this vol., 88ff.]. For the development of intervocalic *Z see Kortlandt 1993 [this vol., 102f.]. For the palatalization of velars see Kortlandt 1975a [this vol., 10ff.].]

10

A note on the Armenian palatalization*

1. At the beginning of our century Pedersen wrote: “ Die gesetze fur die armenische palatalisation der velare sind übrigens sehr dunkel” (1906: 392 [= Kl.S., 170]). The problem has not received a final solution up to the present time. Though most scholars (e.g., Meillet, Grammont, Pisani, Solta) try to connect the presence or absence of palatalization with the laryngeal articulation of the velar occlusive, I think that Pedersen is right when he states that all velars were susceptible to palatalization to the same extent. On the other hand, the suggestion that “ Die palatalisation vor H wäre also nur in der betonten silbe eingetreten” (Pedersen 1906: 396 [= Kl.S., 174]) is not convincing. Since NLQ ‘wife’ shows H-vocalism, it cannot be identical with Gr. �/�=. The connection of NLQ and JHáMNµ ‘glands’ with Russ. HQi and HOH]i contradicts Pedersen's rule because the final accentuation of the Slavic words is due to Dybo's law (cf. Kortlandt 1975b: 14 and passim). I think that in the large majority of cases the absence of palatalization results from the restoration of the velar consonant. Thus, the apparent irregularity is similar to the problem of IE *V after L�X�U�N in Lithuanian �FI��HVSHFLDOO\�.DUDOL QDV��II�� 2. The palatalization is regular in the following cases: (1) HUP ‘warm’ , HU, H QXP, HUPQ, Gr. KJ'µ²+, Kz'#+, Alb. ]MDUP, Skt.

IºSB ). The original velar was not restored because H-vocalism is the only apophonic degree of this root which is attested in Armenian.

(2) LO ‘sinew’ , Lá, Lith. JêVOD, Russ. åtOD. This root also lacks a model for restoration of the velar consonant.

(3) LQ ‘stick’ , QHP, Q HP, Skt. KiQWL, Gr. KJ?�7, Lith. JHQ�, JHQL�, next to JDQHP ‘strike’ .

(4) þµRUNµ ‘four’ , Gr. (Dor.) -z-#'J+, next to NµD DVXQ ‘forty’ with Nµ- from *�N�WZ-.

(5) ãHá ‘wry’ , Gr. )�z�#+, OHG. VFsODK. (6) þLXá ‘branch, finger’ , Gr. �z�#+ (Pedersen 1906: 393 [= Kl.S., 171]). (7) þLP ‘bridle’ , þPOHP, Russ. åPX. (8) Lå ‘snake’ , Skt. ºIJ , Gr. |3�+.

* Reprinted from =HLWVFKULIW��I�U��YHUJOHLFKHQGH��6SUDFKIRUVFKXQJ (.=) 89/1 (1975): 43-45.

$�QRWH�RQ�WKH�$UPHQLDQ�SDODWDOL]DWLRQ� 11

[45]

[44] (9) DþµNµ ‘eyes’ , which is identical with Lith. DNu, except for the additional plural marker -Nµ.

(10) Rþµ ‘not’ , Gr. #½�(�), though the anlaut presents a problem (cf. Hübschmann 1897: 481).

(11) DþHP ‘grow’ , Lith. iXJX, Lat. DXJH , with the same aulaut problem. (12) JRþµHP ‘cry’ , etc., if -þµ- does not represent *-NM- in these words. 3. The velar occlusive has been restored in the following cases: (1) JHáMNµ ‘glands’ , Russ. åHOH]i, with dissimilation of the initial consonant

�FI��'åDXNMDQ��ZLWK�UHI��DV�LQ�&]HFK�KOt]D. (2) NµHUHP ‘scratch’ , Gr. �J?'7, OHG. VFsUDQ. The initial velar was taken

from the synonym NµRUHP, where it is phonetically regular. The same development must be assumed for NµHUWµHP, Lith. NHUW�, unless the vocalism of this word is secondary, cf. Skt. L OUºUJ.

(3) NHU ‘food’ , NHUD\ ‘ate’ , Skt. JLUiWL, Lith. JHUL�. As Meillet points out, the noun belongs to the type Gr. �²�#+, Q²'#+, “ mais avec le vocalisme du verbe [...], les alternances vocaliques de l'indo-européen étant éliminées de l'arménien” (1936: 73). The initial velar was preserved in the noun and extended to the verb.

(4) NLQ ‘wife’ , Gr. �/�=, Russ. HQi, where the initial velar was taken from the plural NDQD\Nµ or from the oblique cases of the singular, cf. OIr. EHQ, gen.sg. PQi.

(5) DUJHO ‘hindrance’ , DUJHOXP ‘keep off’ , Gr. w'�z7, Lat. DUFH . If the Greek verb is derived from a noun x'�#+ ‘protection’ (Solta 1960: 289), the velar consonant in DUJHO must have been taken from the cognate noun in Armenian. Pedersen compares DUJHOXP with Lat. F QVXO next to F QVH (1906: 355 [= Kl.S., 133]).

(6) HOLNµ ‘left’ , etc., where a thematic front vowel is found in Gr. |��%J, Skt. DULFDW. The velar was restored on the basis of the other verb forms.

4. The velar consonant cannot have been restored in KLQJ ‘five’ , KQJHWDVDQ ‘fifteen’ , Gr. %z�-J, Skt. SixFD. I think that the preceding nasal prevented the palatalization. If QNHU ‘companion’ is an old word, it corroborates the rule. It is possible, however, that this word took its front vowel from NHU after the introduction of H-vocalism in the latter, cf. above. 5. The velar consonant in NHDP cannot be due to restoration either, cf. Skt. M YDWL, Russ. åLY~. Elsewhere I have put forward the thesis that both the Balto-Slavic and the Italo-Celtic evidence point to an IE root *J � +L- (Kortlandt 1975b: 80f.). Since the loss of the IE laryngeals in Armenian was posterior to the last phase of the consonant shift (cf. my remarks in Kortlandt 1976), and the

12 )UHGHULN�.RUWODQGW� latter was posterior to the palatalization, it is reasonable to suppose that the velar occlusive in NHDP was not affected by the following front vowel because of the intervening laryngeal. This rule is supported by the preservation of the initial velar in [LQG ‘joy’ . The same rule can be observed in Sanskrit, where it does not hold for K WBUJ because of the laryngeal metathesis in this language. There is no reason to assume a “ tenuis aspirata” in ãHUW ‘chip’ , which may be connected with Gr. )�J 9�� µ� (cf. Pedersen 1905a: 205 [= Kl.S., 67]). [The reason why the original velars could easily be restored was the rise of new velars before front vowels from *WZ, *GZ, and later from the metathesis and from *VZ, *Z (see Kortlandt 1980b [this vol., 28f.]). I now think that the final velar of KLQJ ‘five’ was taken from the ordinal (see Kortlandt 1994b [this vol., 90]). For NHDP ‘live’ I now reconstruct *J � ZL+ � � with metathesis from *J � + � LX� (cf. Kortlandt 1992a: 237, fn.4).]

13

[244]

Albanian and Armenian*

0. It has long been recognized that there must have existed a particular relationship between the pre-Albanian and the pre-Armenian dialects of Indo-European (cf. Pedersen 1900b). A specification of this relationship requires the clarification of certain theoretical, factual, and chronological aspects. In the following I intend to make a contribution to this problem. 1. From a theoretical point of view, one can distinguish between three different types of linguistic development, which may, by way of convention, be labelled ‘genetic’ , ‘contactual’ , and ‘structural’ , and connected with Schleicher's 6WDPPEDXPWKHRULH� Schmidt's :HOOHQWKHRULH� and Martinet's eFRQRPLH� GHV�FKDQJHPHQWV� SKRQpWLTXHV�� respectively. The first type of development presupposes a chronologically identical series of identical innovations. A perfect example is the origin of the Romance languages, for which no dialectal features can be established for the period from Indo-European up to Vulgar Latin. The second type of development is exemplified in a chronologically not identical series of identical innovations, or in a series of historically connected but not identical innovations. Thus, both Czech EUi]GD and Upper Sorabian EUy]GD show metathesis and lengthening in the Common Slavic CorC-group, but not with the same chronology. The lengthening preceded the rise of new timbre distinctions in Czech, whereas it was posterior in Sorabian. The lengthening affected the initial diphthong in Slovak ODNHW,� but not in Czech ORNHW or in Slovak URELW, while all of these words underwent the metathesis. The third type of development links up a set of identical but historically unconnected innovations, such as the rise of [x] in Spanish GLMH and Old Bulgarian U [ . All of the possibilities mentioned in the preceding paragraph represent some kind of generalization beyond the immediately observed facts. It should be clear that a single observed phenomenon cannot be classified without taking into account the framework in which it plays its part. Consequently, there remains quite a bit of freedom for the investigator to interpret the facts one way or another. On the one hand, one may even deny the reality of a period of common Indo-Iranian development and explain the facts in terms of IE dialectal innovations (Makaev 1971). On the other hand, one can attribute such diverse phenomena as the labialization of the labiovelar in Rumanian SDWUX and the * Reprinted from =HLWVFKULIW�I�U�YHUJOHLFKHQGH�6SUDFKIRUVFKXQJ 94 (1980): 243-251.

14 )UHGHULN�.RUWODQGW��

[245]

comparable development in Welsh SLPS to one and the same ‘wave’ (Solta 1965). A final solution to these problems requires at least some agreement about the criteria on the basis of which the choice between various possibilities is made, and this is only partly a question of terminology. In the formulation presented above a 6WDPPEDXP is characterized as a series of common innovations with a common chronology. This does not imply that the language is in all respects homogeneous during the period of common development. The metathesis of liquids does not mark the end of the Common Slavic period, even if it affected different areas to a varying degree. The essential point is that there is a series of common innovations with a common chronology which are posterior to the metathesis. Thus, the presence of dialectal differences does not break up the linguistic unity. In this conception, there can hardly be any doubt about the reality of an Indo-Iranian or Balto-Slavic period. The distinction between ‘structural’ and other developments is only meaningful if the historical connectedness referred to in the above definitions is understood in the narrow sense of the word. Though the assimilation of the initial fricative in Skt. Wº VSB and Lith. ã ãXUDV is the same, it can hardly go back to a common development. If the apparent identitity originates from independent innovations in Indic and Baltic, these are historically unconnected in the sense advocated here, and the development is ‘structural’ rather than ‘genetic’ or ‘contactual’ . Some of those phenomena which at first sight seem to be explicable in terms of a wave theory, turn out to involve real 6WDPPEDXP developments on closer inspection. The well-known retraction of IE *V after L�X�U�N left fewer traces in Baltic than in Indo-Iranian. Yet isolated instances like Lith. PuUãWX show that the development fully affected the Baltic area and that the absence of retraction in the majority of the cases where it should be expected is due to posterior analogical levellings. A close examination of the chronological relations is imperative in such a case. The chronology of sound changes must be understood as the temporal order of their phonemicization. Words like Lith. NXOW Uj�do not imply that de Saussure's law was posterior to their being borrowed into the language. It is by no means correct to infer from Bulg. SOiGQH (= Russ. SyOGHQ) that the metathesis of liquids was posterior to the loss of the MHUV in this language. The establishment of a chronology must be based on less marginal words. Subphonemic changes cannot, as a rule, be reconstructed. 2. One of the most characteristic facts about Albanian and Armenian from the Indo-European point of view is that both languages have led a number of authors to suppose that they have retained the distinction between the three series of velars which are generally reconstructed for the proto-language. As far as I can see, the idea that three phonemically distinct series of velars existed

$OEDQLDQ�DQG�$UPHQLDQ 15

[246]

simultaneously in Proto-Indo-European is a fallacy. There are two decisive arguments against such a reconstruction which anybody who sticks to Bezzenberger’s three series should be ready to answer. On the one hand, the ‘pure velars’ and the other series are largely in complementary distribution, as Steensland has recently shown in his important monograph on the subject (1973). One hardly finds a single environment where all series were opposed to each other. Even if one may not always agree with Steensland's comparisons, the large majority of which were taken from Pokorny, one cannot but subscribe to his conclusion that the proto-language had two series of velar phonemes only. On the other hand, the cases of so-called “ Gutturalwechsel” are far too numerous to be disposed of as incidental irregularities. In a recent article (1974), ýHNPDQ�OLVWV��LQVWDQFHV�RI�YHODU�LQWHUFKDQJH�LQ�%DOWLF�DQG�6ODYLF��QRW�FRXQWLQJ�the onomatopoeic cases. Though it does not seem possible to relate all instances to a single origin, I think that most examples can be accounted for if we assume that the palatovelars lost their palatal character before a resonant in Balto-Slavic. Thus, we find Lith. u uVSBT, Skt. Wº VSB , next to Slavic VYHNU\, Skt. WB S . The regular development of the palatovelar is attested in Lith. DNPX} ‘stone’ , Latv. DNPHQV, whereas the palatal feature was restored in Lith. DãPX} ‘edge (of a knife)’ , Latv. DVPHQV, on the basis of DãWU�V ‘sharp’ . The coexistence of Russ. JXV and Lith. å VuV would seem to indicate an earlier alternation in the root of this old consonant stem. Cf. also Russ. åsOW\M,� åHUG next to ]yORWR,� ]RUyG, and Prussian EDOJQDQ ‘Sattel’ next to EDOVLQLV ‘Kissen’ . The development of a svarabhakti vowel before the syllabic resonant is a relatively late Balto-Slavic innovation because it is posterior to Hirt's law (cf. Kortlandt 1975b: 52). In pre-Slavic the palatal feature was lost before *ZR but not before *ZH, cf. Polish NZLDW,�JZLD]GD next to ZL W\,�]ZLHU]� The palatal feature was restored in ZLDW,�G]ZRQ on the basis of ZLW,�G ZL N��The connection of Latv. NX D with Lith. ãX} HWF��LV�LQFRUUHFW��FI��% JD�� The thesis that the three IE velar series remained distinct in Albanian was first put forward by Holger Pedersen (1900a: 305ff.). According to this author, the labiovelars were palatalized by a following front vowel, whereas the ‘pure velars’ remained intact. Pedersen's most convincing examples of this palatalization are the following: (1) SHVs ‘fünf’ , Skt. SixFD, Gr. %z�-J. (2) VL ‘wie’ < *N � HL etc. (3) V\ ‘Auge’ , Lith. DNuV, Arm. pl. DþµNµ. (4) ]MDUP ‘Hitze’ , Skt. HIBSNº , Arm. HUP, Gr. KJ'µ²+. (5) V ‘nicht’ , Arm. �R�þµ. (6) VMHOO ‘bringe’ , Gr. y�--z��7, Lat. FRO .


[247]

To these can be added ]RUUs ‘Darm’ < *H � SO (Jokl 1937: 141). The main examples of ‘pure velars’ which were not palatalized by a following front vowel are the following: (1) JMsQM ‘finde’ , Gr. 3�� 9�7, Lat. SUH�KHQG � (2) GHUJMHP ‘bin krank’ , Lith. VHUJ�, OHG. VRUJD� (3) HUJML] ‘kleine Laus’ , Lith. pUN . (4) KHOT ‘ziehe’ , Gr. {��7. (5) NRKs ‘Zeit’ , Slavic þDV � (6) THOO ‘bringe’ , Gr. �z�#µ��. (7) THWK ‘schere’ , Lith. NHUW�� Gr. �J?'7.

This material is not convincing. As Hermann pointed out in his discussion of the problem (1907: 47f.), the velars may have been restored analogically in these words. This is certainly the case with THOO� which cannot be separated from VMHOO (cf. Jokl 1963: 126). The word HUJML], the closest cognate of which is Arm. RU LO, offers various problems in several IE languages.

Though the overall view is particularly reminiscent of Armenian, there is a considerable difference in the details. Elsewhere I have pointed out that in Armenian, where we find the same palatalization before a front vowel, the original velar was restored wherever there was a model for its restoration (1975a). It is remarkable that Alb. THWK agrees with Arm. NµHUWµHP (next to NµHUHP and NµRUHP), just as Alb. ]MDUP agrees with Arm. HUP,� HU� There is no agreement between Alb. SHVs and Arm. KLQJ, and between Alb. HUJML] and Arm. RU LO� In the abovementioned publication I suggested that the preceding nasal prevented the palatalization in Arm. KLQJ ‘five’ , where the velar cannot have been restored because there was no model for such an analogical development. Similarly, one could ascribe the absence of palatalization in Alb. GHUJMHP,�HUJML],� KHOT to the preceding resonant. The restoration of the velar in THOO and THWK�must have taken place on the analogy of other apophonic degrees of the root. The word JMsQM does not really fit into the above list because there is no evidence against a palatovelar in this root outside Albanian. It has been suggested that palatovelars lost their palatal feature before a resonant (cf. Hamp 1960). This may have been a common development of Albanian and Balto-Slavic. In the present case the velar may have been taken from those forms which had zero grade in the root, cf. Gr. |3� #�. Finally, I have to discuss the word NRKs� In his etymological dictionary, Gustav Meyer wrote: “ Ist vielleicht mit asl. þDV ‘Zeit, Stunde’ verwandt”


[248]

[249]

(1891: 194), and in his /DXWOHKUH he characterized this comparison as “ Zweifelhaft” (1892: 86). Pedersen had less doubts on this point (“ läßt sich kaum bezweifeln” ), but adds: “ Allerdings ist dies das einzige Beispiel für inlautendes K aus idg. V; sonst herrscht ã” (1900a: 279). There is counter-evidence against intervocalic *V > K in R-stems like YHVK ‘Ohr’ and in the abl. pl. ending �VK� Jokl regarded the comparison as “ eine evidente Wortgleichung” and rejected the connection of the Slavic word with Gr. -�'z7 because the latter shows a labiovelar (1937: 159n.). In fact, the reconstructed pre-Albanian form *N V is not attested in Balto-Slavic, and the comparison must possibly be abandoned. Jokl's connection of the Albanian word with Gr. �?7 etc. may indeed be correct and leads us to another analogy with the Armenian devel-opment. Elsewhere I have suggested that the absence of palatalization in Arm. NHDP ‘I live’ must be attributed to the presense of an intervening laryngeal (1975a). The same factor may have been responsible for the absence of palatalization in Alb. TLV,�TRM ‘wecke’ , Gr. �D�z7, Lat. FLH . 3. The non-identity of the palatalization in Albanian and Armenian seems to be corroborated by the internal chronological evidence of the latter language. As I have pointed out elsewhere (1976), the oldest stage of prehistoric Armenian is characterized by the rise of new labialized consonants. I think that in this language the palatalization must be dated between the loss of the labialized affricates resulting from IE * Z, * KZ, which was posterior to the rise of K from IE *V, and the loss of the labialized dentals resulting from IE *WZ, *GZ� It was probably posterior to the former development because the palatalized labiovelars did not merge with the labialized palatovelars. It was certainly anterior to the latter development because the velar was not palatalized in NµH], Skt. UW N��Since there is agreement between the internal and the comparative evidence, we must reject the possibility of genetic identity between the Albanian and the Armenian palatalization. On the other hand, the facts are too much alike in the two languages not to suggest some kind of historical connectedness. Two other developments which date from the earliest stages of the Armenian language are the rise of K from IE *V and the assimilation in VNHVXU, Gr. x�/' . Neither of them was shared by Albanian. Pedersen attributes the Albanian rise of K from IE *V to historical times (1900a: 340), which does not seem to be in accordance with his analysis of the word VKRK ‘sehe’ (1900a: 283). The retraction of IE * can hardly be that recent. Anyhow, the development of K�from IE *V in Albanian is limited to the position before an unstressed back vowel and has been proved for word-initial *V only. Since it was posterior to the split of IE *V into a voiced and a voiceless reflex, once again we find agreement between the internal and the comparative evidence to the non-identity of the development in Albanian and Armenian. The alleged assimilation in Alb. YMHKsUUs ‘Schwiegermutter’ has

18 )UHGHULN�.RUWODQGW��nothing to do with the one in Arm. VNHVXU� The influence would be progressive in the former language, while it is regressive in the latter. Moreover, the Albanian assimilation is improbable because of two reasons. First, the development of intervocalic *V into K is supported by the questionable etymology of NRKs only, whereas it is contradicted by R-stems like YHVK and the abl. pl. ending �VK. Second, there is a regressive dissimilation in WKL ‘Schwein’ , Lat. V V,� and in WKDQM ‘trockne’ , Lith. VD VDV� In the case of YMHKsUUs it seems correct to stick to Meyer's phonetic explanation (1891: 475), which was also accepted by Pedersen (1900a: 339). There are two developments which may indeed have been common to Albanian and Armenian, and which are anterior to the ones discussed so far according to the Armenian internal evidence (cf. Kortlandt 1976: 96 [this vol., 5]). Pedersen established for Albanian that IE *VZ yields G before a stressed vowel and adduced the following examples (1900a: 286ff.): (1) GLHOO ‘Sonne’ ; Skt. VYjU� (2) GHUJMHP ‘bin krank’ , Lith. VHUJ�� (3) GLUVs ‘Schweiß’ , Gr. ¬ 'I+, Arm. NµLUWQ� This remarkable development, the result of which differs considerably from the normal reflex JM from IE *V before a stressed vowel, e.g. in JMXPs ‘Schlaf, Gr. Â%�#+, suggests that *VZ became monophonemic at a relatively early stage. The difference between the loss of buccal features in the reflex of *VZ before an originally unstressed vowel, e.g. in YMHKsUUs, Gr. x�/' , and the loss of the labial element in the words listed above is structurally analogous to the difference in Armenian between the retracted articulation in ãXQ ‘dog’ , Gr. �E7�, and the early loss of labialization in MD\Q�‘voice’ , Russ. ]YRQ� The suggestion that IE *VZ became monophonemic in Albanian at an early stage is supported by the development of * Z and * KZ in this language. Here again we have to assume early monophonemicization. The only reliable examples are VRUUs ‘Krahe’ < * Z UQ (Jokl 1937:146), Bulg. VYUiND, Lith. ãiUND, and Tosk ]s, Geg ]k ‘Stimme’ , Russ. ]YRQ, Arm. MD\Q� The hypothesis that IE * before *X yielded the same reflex cannot be maintained (cf. Ölberg 1968: 113f.). Thus, the agreement with Armenian is perfect. It encompasses not only the conditions, but also the chronology of the labialization. As Jokl pointed out (1937: 161), the Rumanian borrowing FLRDU < Alb. VRUUs proves that the affricate was preserved up to historical times. It follows that the rise of new labialized consonants was anterior to the assibilation of IE * in clusters. The same chronology must be assumed for Armenian. In contrast with the latter language, the palatalized labiovelars merged with the labialized palatovelars in Albanian.


[250] Apart from the labialization, we thus find that the chronology of the assibilation of IE palatovelars coincides in the two languages. In my discussion of Arm. VNHVXU (1976: 97 [this vol., 6]) I demonstrated that the assimiliation in this word requires a stage where IE * had been assibilated while the reflex of IE * Z was an affricate. The Rumanian word FLRDU shows that this stage may have been Albano-Armenian. The preservation of the occlusive element in the palatovelars before *Z may have been common to a larger area, cf. Ukrainian G]YLQ, G]YLU next to ]YLU� These cases can hardly be attributed to “ affective affricatization” (Shevelov 1964: 145) in view of the correspondence with Polish G]ZRQ and Macedonian G]YRQHF��G]YHU. Postscript [1980] When I wrote this article in 1975, E. Çabej's important contribution (1972) had escaped my attention. I quote from his article (145): “ Zusammenfassend sieht man, daß der jetzige Palatal bei Substantiven wie HUJMs] und bei Verben wie GHUJMHP,� JMHM, THOO, THWK noch kein Beweis für das Bestehen von idg. Reinvelaren vor hellen Vokalen und eine besondere Behandlung derselben im Albanischen ist.” Ölberg's assertions “ Zusammenfassend läßt sich also sagen, daß wir nicht umhin kommen, drei Gutturalreihen im Albanischen vorauszusetzen” (1976: 567f.) and “ Vor allem aber unterschätzt er [= Steensland] die Beweismöglichkeiten für die drei Phonemreihen seitens der kleineren Sprachen wie Albanisch” (ibidem: 569) cannot be maintained: they are based on two instances of non-palatalization before *H (THWK and JMDOP), which allow other explanations. In my view, these two words contain an original palatovelar which was regularly depalatalized before a following resonant (cf. Kortlandt 1978b: 242). The depalatalized variant was subsequently introduced analogically into the apophonic alternants of the root. It is not correct to project every formal distinction which is found in the daughter languages back into the proto-language, especially when there is an obvious explanation in terms of analogic change. Thus, the so-called ‘pure velars’ arose partly from the delabialization of the labiovelars before rounded vowels in the western IE languages, and partly from the depalatalization of the palatovelars before resonants in the eastern IE languages. [See further Kortlandt 1986a [this vol., 68ff.].]

20

[10]

Notes on Armenian historical phonology II: The second consonant shift*

In the first part of these “ Notes” (1976) [this vol., 1ff.] I pointed out that several problems of Proto-Armenian phonology can be solved if two methodological points are taken into account. First, historical developments must be viewed in their chronological perspective. The establishment of a genetic relationship without taking the intermediate links into consideration may lead to hidden contradictions. Second, developments must primarily be viewed in terms of features. The formulation of correspondences in terms of segments has more than once given rise to problems which derive from the method, not from the material. Here I intend to show the consequences of these premises when they are applied to the reconstruction of the Common Armenian system of obstruents. 1. Shortly after the turn of the century Pedersen challenged the obtaining views on the Armenian consonant system with the hypothesis that the voiced stops of the classical language were aspirated (1906: 336-342 [= Kl.S., 114-120]). This view was accepted by Vogt (1938: 327f.), who discussed the matter in detail in a separate study (1958), and later by Allen (1951: 134f.) and Benveniste (1959). Garibjan went a step further and surmised that the voiced stops in the Western dialects which correspond to voiceless stops in the classical language constitute an archaism (1959). After Agajan's demonstration that this view cannot be upheld (1960) it might seem superfluous to return to the problem. Yet I think that it is worth while to do so because of the methodological points which are involved. Indeed, the very sources from which Vogt and Benveniste drew their inspiration (Adjarian 1909 and Allen 1950) allow of entirely different conclusions, which are apparently supported by the newly discovered Southern dialects (cf. Garibjan 1958). I shall not present a survey of opinions, which can be found in two recent monographs (Pisowicz 1976a: 20-27 and Job 1977: 94-99; for a concise survey of the relevant material I refer to Pisowicz 1976b). The following analysis will be based on three principles:

* Reprinted from 6WXGLD�&DXFDVLFD 4 (1978): 9-16.

1RWHV�RQ�$UPHQLDQ�KLVWRULFDO�SKRQRORJ\�,, 21

[11]

(1) A reconstruction of the Common Armenian consonant system on the basis of the modern dialects must logically anticipate a comparison with material from other Indo-European languages. (2) If the consonant systems of two related dialects differ in more than a single feature, the historical connection between them involves at least two distinct developments. (3) If a single uninterrupted central area differs from the peripheral areas with respect to a specific feature, it is probable that the central dialect has innovated. 2. In order to simplify the discussion I number the modern dialects in such a way that the first digit reflects the correspondence with classical W and the second digit the correspondence with classical G, both in word-initial position, and that a minimum difference between numbers reflects a minimum difference between dialects in terms of features. In the following list I give, next to the number of each dialect, the reflex of classical W, G, Wµ, the corresponding number in the classifications of Vogt (1958) and Garibjan (1959), and a typical representative. class W� � G�� Wµ V G ex. 11 W� =�� W�� Wµ 4 7 Van 12 W� � G� � Wµ 3 6 Agulis 13 W� � Gµ� � Wµ 1 2 Erevan 21 G� � W� � Wµ � 4 Sasun 22 G = G� � Wµ 5a 3 Trabzon 23 G� � Gµ� � Wµ 2 1 Sivas 20 G� � Wµ� � Wµ 5b 5 Malatia 3. The dialects 11, 22, 20 have apparently come into existence as a result of the neutralization of a phonological opposition. According to Vogt (1958:148), “ il s'agit évidemment d'une simplification secondaire des systèmes centraux” , i.e. 13 and 23. Should 11 indeed be derived from 13? Since these dialects differ in two features, we have to assume an intermediate stage. If the voicedness was lost earlier than the aspiration, the reflexes of classical G and Wµ must have merged, which is not the case in 11. If the aspiration was lost earlier than the voicedness, the intermediate stage was identical to the system of dialect 12. But there is no reason why 12 should be derived from 13, both 11 and 13 may actually have to be derived from 12. Geographically, the area which 11 and 12 occupy together forms a semi-circle around the central dialects: ArþHã�� Van � Xoy (11) � Agulis � Meghri (12) � Karabagh � Kanaker (11) � Lori � Tiflis � Artvin (12). This situation suggests that 13 must be derived from 12.


[12]

4. The dialects 22 and 20 may indeed be derived from 23, but either of them can also be derived from 21, a dialect which Vogt does not take into account because it is not covered in Adjarian’s monograph. Moreover, 22 may be derived from 12 in the same way as 23 can be derived from 13. A choice can only be made by taking into account the geographical distribution of the dialects. Since the position of Trabzon (22) with respect to the Artvin-Tiflis area (12) is the same as that of Little Armenia (23) with respect to Central Armenia (13), it is reasonable to assume that the historical relationship between the Trabzon dialect and its Eastern neighbour is the same as that between 23 and 13. The suggestion that 13 must be derived from 12 and the impossibility of deriving 12 from 22 then involve the consequence that the dialects of Trabzon and Little Armenia must be derived from 12 and 13, respectively. If this is correct, the semi-circle discussed above can be extended to Trabzon. Other parts of 22 may have different historical connections. Thus, the isolated dialect of Maraš (22) must probably be derived from the contiguous Hadjin-Zeytun (21). The Malatia-Urfa area (20) is situated within the semi-circle Sasun-Dersim-Hadjin-Beylan-Svedia (21) and must therefore be derived from the latter. The situation round Hamšen is complicated: here we find four different systems in Djanik (21), Ordu (20), Mala (23), and Trabzon (22). A transitional dialect between 21 and 20 has been described by Dumézil (1964). 5. Which distinctive features can be reconstructed for the oldest stage of the apparently archaic dialect 12? I think that the answer is provided by


[13]

Allen’s painstaking phonetic analysis of an Eastern Armenian dialect (1950). The unaspirated voiceless plosives are glottalic (“ ejective” ) in this dialect, whereas the voiced stops of the classical language are voiceless in initial position. Thus, it is a transitional dialect between 12 and 11, having lost the voicedness of G while retaining the opposition between G and W. Actually, the opposition between voiced and voiceless initial stops was restored by the introduction of the loan words EHJ and ER\ (Allen 1950: 202). The term “ potential voiced aspirates” which Allen applies to the reflexes of classical G etc. has given rise to misunderstanding on the part of Benveniste, who inferred the existence of voiced aspirates from the description (1959: 50). In fact, voicing and aspiration are mutually exclusive, the “ potential voiced aspirates” being voiced and. unaspirated if preceded by a nasal, and lightly aspirated and voiceless in final position after U. These are precisely the positions where almost all Armenian dialects show unaspirated voiced and aspirated voiceless plosives, respectively (cf. Pisowicz 1976a: 61f.). In initial position, the “ potential voiced aspirates” are voiceless. “ They are distinguished from the ejectives by having pulmonic as opposed to glottalic plosion, and from the aspirates by the absence of voiceless breath on release. The most notable feature differentiating them from the ejectives, however, is to be found in a following vowel, which is articulated with markedly stronger breath-force and on a lower pitch than is general in other but comparable contexts.” (Allen 1950: 200) The transference of the distinctive feature to the following vowel is carried through completely in a part of the dialects 11 (cf. Pisowicz 1976b: 215f.). In the original system, the glottalic articulation of the “ ejectives” was apparently distinctive. Indeed, 19th century loan words from Russian showed aspirated plosives in Armenian, e.g. SµHþµ, PDQHWµ from SHþ, PRQHWD (cf. Pisowicz 1976a: 18). Thus, I do not share the usual view that the glottalic articulation of the unaspirated voiceless stops in the Tiflis-Artvin area is due to a Caucasian substratum. It is more probable that the latter simply favoured the preservation of a feature which was already present. 6. The newly discovered dialects 21 seem to corroborate the antiquity of the glottalic articulation. According to Garibjan (1959: 85f.) and Pisowicz (1976a: 78), these dialects must be derived from the Western system 23. Since 21 and 23 differ in two features, we have to assume an intermediate stage. If the voicedness was lost earlier than the aspiration, the intermediate stage was identical to 20. If the aspiration was lost earlier than the voicedness, the intermediate stage was identical to 22. In either case two series would have merged, which is not the case in 21. I conclude that 21 must be derived from an Eastern dialect. Since the geographical position of the Southern dialects (21) with respect to the Van area (11) is the same as that of Little Armenia (23) with respect to Central Armenia (13), it is reasonable to assume that the historical


[14]

relationship between 21 and 11 is the same as that between 23 and 13. We come to the conclusion that the Southern dialects developed from the system of their Eastern neighbour at a time when the latter had not yet lost the distinction between the original (glottalic) unaspirated voiceless stops and the ones that originated from the devoicing of the classical voiced stops. Indeed, the dialect of Šatax (between Van and Sasun) distinguishes between two series of unaspirated voiceless stops (cf. Pisowicz 1976a: 66) and is in this respect transitional between 11 and 21. We can now connect the two semi-circles discussed above: together they constitute an uninterrupted line of dialects which are archaic with respect to the encircled areas. 7. We have established the following relative chronology: I Rise of aspiration in voiced stops (12 � 13). II Devoicing of unaspirated voiced stops (12 � 11). III Voicing of glottalic stops (11 � 21, 12 � 22, 13 � 23). IV Loss of unaspirated voiceless stops (21 � 22, 21 � 20). The absolute chronology can only be established on the basis of loan words. Since loans from Arabic are subject to shifting while loans from Turkic are not (Agajan 1960: 44), we have to date III between the 7th and the 10th century. It is possible that GHYHORSPHQW�,�ZDV�LQ�SURJUHVV�GXULQJ�WKH�FODVVLFDO�SHULRG��DV�'åDXNMDQ�VXJJHVWV�(1976: 76). 8. Now we return to the Hamšen area. Though definite conclusions can only be reached when further material will have become available, the geographical situation and the existence of transitional dialects suggest an explanation in terms of the established developments. The Trabzon dialect (22) arose from 12 as a result of III. The Mala dialect (23) cannot be separated from the neighbouring dialects of Little Armenia: it originated as a result of I and III. For the dialects of Ordu (20) and Djanik (21) we have to assume a local development, viz. the devoicing of voiced stops (II). The latter dialect was apparently not affected by I. Since the reflexes of classical W and G did not merge in Djanik as a result of the devoicing, we have to assume that here too the glottalic articulation of the original unaspirated voiceless stops was preserved until the Western Armenian voicing took place at stage III. 9. Thus far I have limited the discussion to word-initial plosives because it is the position of maximal contrast in the Armenian dialects. The same type of analysis can be applied to other positions. The generalization of voiced stops after nasals and voiceless aspirates after prefinal U are probably early developments because they have affected the large majority of dialects. Intervocalically, the following types are found:


[15]

class W� � G� � Wµ ex. 10a W� � Wµ = Wµ Karabagh, Lori 11a W = W� � Wµ Van 12a W� � G� � Wµ Agulis 20a G� � Wµ = Wµ Malatia, Trabzon, Erevan, Tiflis 21a G� � W� � Wµ Sasun 23a G� � Gµ� � Wµ Sivas The areas 12a and 23a are small islands within 10a and 20a, respectively, while 11a and 21a are considerably smaller than 11 and 21. A large part of the Eastern dialects have the Western (voiced) reflex of intervocalic W etc. The Northern dialects have all been subject to the aspiration of intervocalic G etc. except for Djanik, where aspiration was lost. Following the principles which have been put forward above, one arrives at the same reconstruction and the same relative chronology as have been established for initial plosives, plus one extra development: V Devoicing of voiced aspirates (13a � 10a, 23a � 20a). 10. The reconstructed Common Armenian obstruent system now appears as follows: aspirated plain glottalic voiced G voiceless Wµ� � � � � � W Elsewhere (1977) I have put forward the following reconstruction of the Proto-Indo-European obstruent system: aspirated plain glottalic lenis GK� � � � � � G fortis � �� W The glottalic articulation of PIE *G etc. is established primarily on the basis of the Balto-Slavic evidence. It has been preserved until today in Latvian S GV ‘footstep’ , QRJV ‘naked’ . The first Armenian consonant shift is now seen to consist in the loss of the aspiration of PIE *GK, the reinterpretation of lenis PIE *G as voiceless, and the weakening of the occlusion of PIE *W. The actual development was rather more complicated, as I intend to show on another occasion. [See also Kortlandt 1998c [this vol., 126ff.].]

26

[98]

On the relative chronology of Armenian sound changes*

The author presents a possible chronological ordering of the main develop-ments in the Proto-Armenian phonological system. It is an attempt at integrating the findings of earlier investigators into a coherent whole.

The comparative linguist who turns to Armenian is confronted with three characteristic problems. Firstly, the number of etymological correspondences between Armenian and its Indo-European sister-languages is limited. There is no reason to expect that the number of reliable etymologies will grow substantially in the future. Secondly, the phonetic changes which affected the Armenian language in the course of its prehistory are considerable. The correspondences between the attested forms and their Indo-European forbears are therefore less straightforward than in the case of most other branches. Thirdly, the time gap which separates the earliest Armenian texts from the reconstructed proto-language is much longer than the prehistory of Anatolian, Indo-Iranian, or the classical languages. In consequence, the course of development which led to the oldest observable linguistic system cannot be determined with the same degree of precision. One way to tackle these difficulties is the tentative reconstruction of prehistoric stages on the basis of relative chronology. This approach has never, as far as I am aware, been pursued in a systematic way, although many interesting statements on relative chronology can be found in the earlier literature, especially in the writings of Grammont (1918), Zabrocki (1951), and Djahukian (1967). Here I want to present a possible chronological ordering of the main developments in the Proto-Armenian phonological system from PIE times up to 400 A.D. The reconstruction attempted here is subject to three requirements, which correspond with the three types of difficulty which have just been mentioned. First, the proposed relative chronology should be compatible with as many etymologies as possible without leading to inconsistencies. Second, the transition of the linguistic system from any stage to

* Reprinted from 3URFHHGLQJV�RI�WKH��VW�LQWHUQDWLRQDO�FRQIHUHQFH�RQ�$UPHQLDQ�OLQJXLVWLFV (Delmar, N.Y.: Caravan Books, 1980): 97-106.

2Q�WKH�UHODWLYH�FKURQRORJ\�RI�$UPHQLDQ�VRXQG�FKDQJHV 27

[99]

the next one should be due to a natural type of development. Third, the proposal should yield a reasonable periodization of the linguistic prehistory. It must be emphasized that the relative chronology which is presented here is a tentative analysis and does not pretend to solve the riddles of Armenian historical phonology. It must rather be viewed as an attempt at integrating some of the findings of earlier investigators into a coherent whole which will have to be modified and refined in the light of new findings and alternative theories. PA 0 = Proto-Indo-European. Elsewhere I have argued that the unaspirated voiced stops of the proto-language were glottalic and that the glottalization was preserved in Armenian up to historical times (Kortlandt 1978a [this vol., 20ff.], 1979c). The opposition between the velar orders was neutralized after *X. The development of the laryngeals will not be discussed in this paper. PA 1. Loss of aspirated stops. I assume that the loss of aspiration was a dialectal Indo-European development which Armenian shared with Germanic and Balto-Slavic, and probably with Albanian, Iranian, and Tocharian. For this stage I assume the existence of injective (PIE glottalic), plain voiced (PIE aspirated), and voiceless stops. PA 2. Rise of new labialized stops. Elsewhere I have argued that the development of *VZ, * Z, *WZ, *GZ into Nµ, ã, Nµ, N is best understood if we assume that these clusters became single phonemes at an early stage (Kortlandt 1976 [this vol., 5-7]). Thus, I reconstruct *F � Q ‘dog’ , *HF � RV ‘horse’ , which were to develop into ãXQ, ã. To the same stage I date the following developments: a) *-XJ-, *-HXN- became *-X -, *-HZ -, e.g. OXF ‘yoke’ , OR\V ‘light’ , Skt. \XJiP, URNi . b) *DXJ-, *RXN- became *DJ �

-, *RN �- in DþHP ‘grow’ , Rþµ ‘not’ , cf. Lat. DXJH ,

Gr. #½�?. c) *DQJ �

-, *RQJ �- became *DZ - in DZM ‘snake’ , DZFDQHP ‘anoint’ , Lat. DQJXLV,

XQJX . The absence of this development in DQNLZQ ‘angle’ and DQNDQLP ‘fall’ betrays a different ablautstufe. PA 3. Assibilation of * to * and rise of * from *VN, e.g. WDVQ ‘ten’ , FµHOXP ‘split’ , Lith. GHãLPW, VNHOWL. This development was posterior to the rise of new labialized stops because * Z did not become * Z, as is clear from the result of the next development. PA 4. Assimilation in *VZH XUD ‘mother-in-law’ , which became * ZH XU at this stage and then developed into VNHVXU. It is reasonable to assume that the assimilation was posterior to the assibilation of medial * to * . PA 5. *V became *K unless it was preceded by a consonant or followed by an obstruent. This development was obviously posterior to the assimilation which yielded * ZH XU . Intervocalic *K was subsequently lost, while word-final

28 )UHGHULN�.RUWODQGW

[100]

[101]

*K was to develop into Nµ, e.g. in the nom.pl. ending. PA 6. Development of * (from PIE * ) and * (from PIE *VN) into V and *F, respectively. This change was posterior to the development of earlier *V into *K because the two sibilants did not merge. PA 7. Redistribution of labialization: a) *M �

became M, e.g. MD\Q ‘voice’ , which is best compared with Slavic ]Y N . b) *F �

became *V �, e.g. *V � XQ ‘dog’ , *HV � RK ‘horse’ .

c) *KZ became *K �, i.e. the voiceless counterpart of *Z, e.g. *K � HKXU ‘sister’ ,

*K � RSQRK ‘sleep’ , later NµR\U, NµXQ, Skt. TWºT ÛTWºQOB . The opposition between *K �

and *Z can be compared with that of English ZKLFK, ZKLOH vs. ZLWFK, ZLOH. PA 8. Palatalization. Elsewhere I have argued that all velars were palatalized by a following front vowel (Kortlandt 1975a [this vol., 10ff.], cf. Pedersen 1906: 392-396 [= Kl.S., 170-174]). The absence of palatalization in the historical material results in most cases from the restoration of the velar consonant on the basis of apophonic alternations. Thus, the initial velar of NµHUHP ‘scratch’ was taken from the forms which underlie the synonym NµRUHP, while there was no model for its restoration in þµRUNµ ‘four’ or HUP ‘warm’ , Gr. -z-#'J+, KJ'µ²+. The palatalization was posterior to the elimination of the labialized palatals because the latter did not merge with the palatalized labio-velars (which had not yet lost their labialization at this stage). PA 9. *W �

and *G � became *N �

and *J �, e.g. NµH] ‘thou (acc.)’ , PHáN

‘soft’ , Skt. UW N Û N EW .1 It follows from these examples that the shift of articulation was posterior to the palatalization.

PA 10. Lenition: *S, *W, *N, *N � became *Q, *K, *3, *3

�. This

development was posterior to the voicing of these stops after resonants, e.g. PDUG ‘man’ , KLQJ ‘five’ , Gr. �'#-²+, %z�-J, because there is no trace of spirantization in the latter environment. If my conjecture that the absence of palatalization in KLQJ can be attributed to the preceding nasal (Kortlandt 1975: 45) is correct, the voicing of the final stop may have been anterior to the palatalization (PA 8). It was posterior to the assibilation of PIE * (PA 3) because the latter did not share the voicing rule, e.g. KDUVQ ‘bride’ , cf. Lat. SURFXV. The devoicing of the glottalic stops was probably anterior to the voicing after resonants because the latter did not block the devoicing, e.g. VLUW ‘heart’ , JRUF ‘work’ , Gr. ��' ?�, |'�#�. The loss of the palatal articulation in DUW ‘field’ , Gr. w�'²+, must probably be dated between the assibilation of * and the devoicing of the glottalic stops because the change of implosive *M to *G before *U is more natural than other possible developments, especially because the palatal articulation was preserved in PHUM ‘near’ , Gr. µz3'�.

1. On the erroneous supposition that initial *GZ yielded UN see Pedersen 1906: 398f. [= Kl.S., 176].


[102]

Unlike Godel (1975: 101), I agree with Meillet (1936: 56 and 79-80) and Pedersen (1905a: 215-218 [= Kl.S., 77-80]) that word-final postvocalic nasals were lost in Proto-Armenian, except in monosyllables. Most probably, we have to assume the early rise of nasal vowels, which were subsequently denasalized. The syllabic nasals still remained in existence at this stage. The preservation of word-final nasals in monosyllables suggests that the rise of nasal vowels was posterior to the fixation of the stress on the penult syllable. Postvocalic nasals were also lost before fricatives, e.g. XV ‘shoulder’ , KDVDQHP ‘reach’ , Skt. º TB Û B OÌNJ. It follows from these words that the loss of the nasal consonant was posterior both to the assibilation of PIE * to * (PA 3) and to the development of PIE *V to *K (PA 5). It was also posterior to the raising of *R to X before a nasal consonant, as is clear from XV. On the other hand, it was probably anterior to the lenition, which gave rise to new fricatives. The loss of *K (from PIE *V) in various positions and its assimilation to a following *U can perhaps be dated to the same stage as the lenition. The loss of *K was certainly posterior to the raising of *H to L before a nasal consonant because the latter development did not affect HP ‘am’ , ]JHQXP ‘dress’ , Gr. J¨µ?, {��/µ��. The loss of intervocalic *K was probably posterior to the development of * into X, but anterior to the development of *HX into R\, cf. NµR\U ‘sister’ , Skt. TWºT . The loss of intervocalic *\ must be dated to an earlier stage: it was posterior to the development of * into L, but anterior to the loss of phonemic quantity because the long vowel was not preserved in HUHNµ ‘three’ , Skt. WUi\D . PA 11. Rise of new velar fricatives: a) *Z became *�

�, which was to develop later into J in most positions.

b) *K � �(from *VZ) became *3�, e.g. in the forms which were to develop into

NµR\U, NµXQ. c) Final *K (from *V) became *3, e.g. in the nom. pl. ending. It is probable that these changes were posterior to the rise of fricatives as a result of the lenition. It is also probable that they were posterior to the loss of intervocalic *K because the latter did not develop into *3. PA 12. Simplification of consonantal articulation: a) Initial *Q became *K, e.g. KDUFµDQHP ‘ask’ , Skt. Q DDI NJ. This *K did not merge with *K from PIE *V, which had apparently been lost in several positions already, cf. DUEL ‘drank’ , Lat. VRUEH . b) *Q, *K, *3 became *I before consonants, e.g. gen.sg. *KDIUR[ ‘father’ , later KDZU, Gr. %�-'²+. Initial *Q, *K, *3 were lost before a consonant, e.g. HUHZLP ‘appear’ , Gr. %'z%7. c) Loss of labialization before *R, *X, and nonsyllabic *U. This development was posterior to the rise of velar articulation in the reflex of PIE *Z (PA 11) because it affected the fricative which developed from the latter, e.g. QRU ‘new’ , VRU ‘cave’ , gen.sg. DáEHU ‘spring’ , DOHU ‘flour’ , as opposed to


[103]

nom.sg. DáELZU, DOLZU. In these examples, delabialized *� was lost at the next stage. Delabialized *V �

became ã, e.g. ãXQ ‘dog’ , ã ‘donkey’ . PA 13. Loss of intervocalic and preconsonantal fricatives: a) Intervocalic *K and *3 became *\, e.g. KD\U ‘father’ , þµRUN ‘four’ , Gr. %�-=', -z-#'J+, but were lost before a syllabic *U, e.g. inst.sg. IBSC ‘father’ , earlier *IB CJ. The development did not take place after nonsyllabic resonants, which were probably characterized by strong friction at this stage, e.g. DZWµ ‘sleeping place’ , HZWµQ ‘seven’ , cf. also ONµDQHP ‘leave’ (with zero grade from the aorist, cf. JWDQHP ‘find’ next to JLWHP ‘know’ ), aor. ONµL (with restoration of the velar consonant on the basis of the present tense), Gr. �J?%7, |��%#�. It was posterior to the delabialization before *R and *U (PA 12c) because the labial feature was lost in JD\O ‘wolf’ , OHDUG ‘liver’ (gen. OHUGL), QHDUG ‘sinew’ (gen. QHUGL), Skt. W LB ÛZºL U ÛTO WB. The new *\ did not merge with the intervocalic reflex of PIE *\, which had been lost by this time, cf. HUHNµ ‘three’ vs. EHU ‘brings’ , Skt. USºZB ÛCIºSBUJ. b) Intervocalic *Q and preconsonantal *I became *Z, e.g. HUHZLP ‘appear’ , KDZU ‘of the father’ , Gr. %'z%7, %�-'²+. The development was apparently posterior to the merger of preconsonantal *Q, *K, *3 into *I (PA 12b). The new sequences *HZ and *RZ (the latter of which developed into X) did not merge with the reflex of PIE *HX and *RX, cf. HZWµQ ‘seven’ , XWµ ‘eight’ , NµXQ ‘sleep’ vs. OR\V ‘light’ , Gr. x%-9, ±�-I, Â%�#+, �J/�²+. The word NµR\U ‘sister’ shows that the development of *HX into R\ must be dated between PA 10 and PA 13. c) *� was lost between vowels and before U (cf. PA 12c). PA 14. Loss of the syllabic resonants, which received an epenthetic D. The word JD\O ‘wolf’ , Skt. W LB , shows that the loss of the syllabic resonants was posterior to the development of *N �

through *3� (PA 10) and *3 (PA 12c)

into *\ (PA 13), but anterior to the metathesis (PA 15).2 As was pointed out above, new *�DQ from postconsonantal *�Q and *�P did not merge with earlier *�DQ from PIE *� P because postvocalic word-final nasals had probably been lost by the time of the lenition (PA 10) already. I date the rise of a prothetic vowel before initial U to the same stage as the loss of the syllabic resonants, e.g. HUHNµ ‘three’ , HUHZLP ‘appear’ . This development was certainly posterior to the loss of the initial obstruent (PA 12b) because the latter would otherwise be reflected as Z (PA 13b). I also assume the rise of a prothetic schwa before initial clusters at this stage because it is a prerequisite for the metathesis (PA 15). Of course, the “ prothetic” vowel of HUHN ‘evening’ and RUFDP ‘vomit’ belongs to an older layer, cf. Gr. |'J�#+, y'JE�#µ��: it represents a PIE laryngeal (there was no initial *U in Proto-Indo-European).

2. Cf. Grammont 1918: 237 and Winter 1962: 261.


[104]

PA 15. Metathesis.3 As has just been pointed out, the metathesis was posterior both to the loss of the syllabic resonants, cf. JD\O ‘wolf’ , and to the rise of prothetic vowels, e.g. HáED\U ‘brother’ , Lat. IU WHU. PA 16. Apocope: the vowel of final syllables was lost. A lost *L or *X was preserved as an epenthetic \ or Z if the preceding vowel was *D, e.g. D\U ‘man’ , DUWDZVU ‘tear’ , Gr. w�=', 9�'/. The apocope was posterior to the metathesis in D\O ‘other’ , Lat. DOLXV, and JD\O ‘wolf’ , in which cases the preceding vowel was *D, but sufficiently early to inhibit the metathesis where the preceding vowel was less susceptible to infection, e.g. Rá ‘whole’ , PXQ ‘mute’ , VWHU ‘barren’ , P ‘middle’ , OIr. XLOH, Lat. P WXV, VWHULOLV, PHGLXV, where final *\ became . The apparent epenthetic *\ in P is probably recent: it is regular after *H before a palatal consonant and does not reflect a lost *\, cf. ã ‘donkey’ , Lat. HTXXV, where the palatal consonant has a different origin (Lamberterie 1978: 262-266). PA 17. Simplification of consonantal articulation: a) Final *�

� became Z, e.g. QDZ ‘ship’ , Lat. Q YLV.

b) *�� became Z before final U in DáELZU ‘spring’ , DOLZU ‘flour’ . This

development was blocked by the preceding resonant in WD\JU ‘husband's brother’ , Skt. EFW . c) Final *�

�3 became *�

� (or possibly *�) in gen.sg. DUHJ ‘sun’ , earlier

*DUH� � H3. It is clear that these developments were posterior to the apocope. PA 18. Loss of labialization. This development was probably posterior to the change of final *�

� into Z (PA 17).

PA 19. Shortening of *K, *3, *�, which became Wµ, Nµ, J� also N after V in VNHVXU ‘mother-in-law’ . This development was probably posterior to the loss of labialization. PA 20. Rise of new [ in [DFDQHP ‘bite’ , V[DOLP ‘err’ , Skt. LI E NJ, TLIºM NJ. I assume that the PIE clusters *S+, *W+, *N+ remained intact up to the lenition (PA 10), then became affricates or strongly aspirated occlusives Sµ, Wµ, Tµ, the latter of which now changed into the fricative [. The rise of this [ was probably posterior to the shortening of earlier *3 to Nµ (PA 19), not simply because the two did not merge (they may not have had the same place of articulation), but because it would otherwise remain unclear why [ from *N+ was not subject to the shortening. PA 21. Influx of Iranian loan words. This development brought a large number of new consonant clusters into the language. PA 22. Syncope: loss of pretonic *L and *X. The oldest layer of Persian loan words are subject to the syncope, e.g. QãDQ ‘sign’ , GVURY ‘blamed’ , Pehlevi

3. Cf. Grammont 1908: 231-236.


[105]

Q ã Q, GXVUDY. The relative chronology which I have given here is open to expansion and modification. I have not discussed separate etymologies, but only examined their compatibility within the limits of plausible phonetic developments. Thus, if one rejects the historical connection between JD\O and the Indo-European word for ‘wolf’ , it should be clear that there are no phonetic reasons for doing so. On the other hand, the compatibility with the rules does not prove that the etymology is correct: the word may still be a borrowing. The establishment of a chronology presupposes the clarification of the etymological correspondences. For this reason, several developments have not been included in the above chronology, e.g. the fate of initial *K (from PIE *+, *V, and *S) and the Armenian umlaut. The change of *H to R was obviously anterior to the syncope of medial *X (PA 22) in RUFDP ‘vomit’ , Gr. y'JE�#µ��. For the change of *R to D I slightly modify Pedersen's rule, which is unjustly disregarded by Greppin:4 initial *R became D- in open syllables unless the vowel of the following syllable was *R. This development must be dated after the syncope (PA 22) in view of RUFDP and R]QL ‘hedgehog’ , Gr. y3¥�#+. The rise of D\- from PIE initial RL�can be dated to a much earlier period (perhaps PA 2, cf. DZFDQHP ‘anoint’ from PIE *RQJ �

-), e.g. D\W ‘cheek’ , Gr. #« #+. Finally, it must be remarked that the establishment of a chronology presupposes that the material is homogeneous. It has been argued that Classical Armenian is a mixture of dialects (Winter 1966: 201-211). If this is correct, the method which is employed in this paper breaks down. For the time being, I am not convinced that the view of Armenian as a mixed language is correct. If it were, one would expect to find more doublets like WµDUãDPLP and WµD DPLP ‘wither’ . Moreover, there seems to be no semantic indication that different parts of the inherited vocabulary represent different underlying dialects. Indeed, I think that the reflex of PIE *UV offers the only trace of early dialectal diversity.5 [On the methodological problems see also Kortlandt 1996b [this vol., 110ff.]. I now think that aspiration had never arisen in Indo-European times but was a local development of the separate branches (cf. Kortlandt 2003). For *GZ > N see also Kortlandt 1989 [this vol., 88ff.]. On PHUM ‘near’ see Kortlandt 1985b [this vol., 58]. For the loss of final nasals see also Kortlandt 1985c [this vol., 63ff.]. For the loss of *K < *V see also Kortlandt 1996a [this vol., 117ff.]. For initial K� see also Kortlandt 1983b [this vol., 39ff.]. For the development of the PIE. laryngeals see Kortlandt 1984b [this vol., 54ff.], 1985b [this vol., 57ff.], 1987a

4. Pedersen 1900c: 98-100, Greppin 1973: 24; cf. also Meillet 1894a: 153-157 and Grammont 1918: 223-225. 5. On KRJL ‘breath’ and RJL ‘soul’ I agree with Greppin, o.c., 59-61.


[this vol., 75ff.], 1991 [this vol., 96f.], 1997 [this vol., 120f.]. For DUWDZVU ‘tear’ see Kortlandt 1985a [this vol., 65ff.]. For the development of *\ see also Kortlandt 1984a [this vol., 47ff.], 1994a [this vol., 104ff.], 1997 [this vol., 120f.], 1998a [this vol., 122ff.]. The development of initial *\� offers an example of early dialectal diversity in Armenian. The main difference between Job’ s chronology (1995: 307) and mine is that his stages (2b), (3), (4) are in the reverse order because of his unwillingness to recognize glottalized stops for Proto-Armenian. This deprives him of the possibility to explain why *UW and *UG changed places. The problem has an exact parallel in Germanic (cf. Kortlandt 1988b).]

34

[30]

On the Armenian personal endings*

Elsewhere I have argued that the Indo-European proto-language had totally different sets of endings in the thematic and the athematic flexion of the active present (1979a, 1979b) and made a fundamental distinction between transitive and intransitive middle paradigms (1981a, 1981b). In this article I intend to show the relevance of these findings for the explanation of the Armenian personal endings. I will limit the references to a minimum because a survey of earlier views can be found in Godel’s handbook (1975) and Jasanoff’s recent article on the subject (1979). 1. The PIE primary athematic endings have been preserved in the Armenian present tense: 1st sg. -P < *-PL 2nd sg. -V� < *-VL 3rd sg. -\� < *-WL 1st pl. -PNµ < *-PHV 2nd pl. -\Nµ < *-WH�V 3rd pl. -Q� < *-QWL The 2nd sg. ending -V represents the postconsonantal reflex of *-VL. After a vowel, the *V was lost phonetically and restored analogically. The 3rd sg. ending -\ and the 2nd pl. ending -\Nµ contain the intervocalic reflex of *W before a front vowel. The 2nd pl. ending received an additional *-V on the analogy of the 1st pl. ending. In the present subjunctive, the personal endings are preceded by the suffix -LFµH- < *-RL�VNH- (Pedersen 1905a: 207 [= Kl.S., 69]). 2. The PIE primary thematic endings are reflected in the aorist subjunctive: 1st sg. -LFµ� � < �� TL� 2nd sg. -FµHV� < �� TL�F�TJÛ 3rd sg. -Fµ < �� TL�F�UJÛ 1st pl. -FµXNµ� < �� TL�PNPN�TÛ 2nd pl. - LNµ� 3rd pl. -FµHQ�� < �� TL�F�OUJÛThe original 2nd sg. ending *-HL merged with the reflex of *-HVL as a result of the * Reprinted from $QQXDO�RI�$UPHQLDQ�/LQJXLVWLFV 2 (1981): 29-34.

2Q�WKH�$UPHQLDQ�SHUVRQDO�HQGLQJV 35

[31]

loss of intervocalic *V. When the *V was restored in *-QDL < *-Q VL and *-QXL < *-QXVL, it was also introduced into the thematic paradigm. The original 3rd sg. ending *-H was enlarged with the athematic ending *-WL. The 3rd pl. ending is analogical. The absence of P from the 1st pl. ending -FµXNµ is striking. Following a suggestion by Schindler (cf. Jasanoff 1979: 136), I assume that *P was lost before *X in Proto-Armenian. This development was posterior to the raising of *R to *X before *P. Thus, the expected reflex of the 1st pl. ending is -X < *-XPX < *-RPRP. It received an additional -Nµ on the analogy of the athematic ending. The loss of *P in Arm. -FµXNµ can now be adduced as additional evidence in support of the reconstruction *-RPRP for the primary 1st pl. ending of the PIE thematic flexion (cf. Kortlandt 1979a: 63f. and Beekes 1981b). 3. The 2nd pl. ending - LNµ is found not only in the aorist subjunctive, but also in the middle aorist imperative, whence it spread to the active aorist and middle present imperatives. I do not share the usual view that this ending replaced the original aorist subjunctive ending (Meillet 1936: 121, Godel 1975: 46) because such a replacement would be unmotivated and destroy the regularity of the paradigm. The ending rather contains a precious archaism from the time before the generalization of the subjunctive suffix -Fµ-. I think that - LNµ represents the middle optative ending *-LJ < *� EIXF, cf. Skt. � EIWBN. It survived because it was also used imperativally. The ending was enlarged with the personal pronoun * XNµ (Meillet 1936: 92) or rather *\XNµ < *M V, cf. Lith. M V, Av. \ ã. The latter development must be dated to the period between the apocope of *-H and the syncope of *-L-. The final part *-XNµ of the ending was eventually replaced with -LNµ on the basis of the regular 2nd pl. aorist ending. For the use of the enclitic pronoun after imperative forms cf. HO�GX |!J�KJ Gen. 12.1, HUNHUXNµ� GXNµ Q#�J¥)KJ Mt. 10.28 (Jensen 1959: 102). The ending - LNµ gave rise to a 2nd sg. middle imperative ending � LU. 4. The expected reflexes of the PIE secondary active endings, which underlie the Arm. active aorist indicative forms, are the following: 1st sg. zero after a vowel and *-Q after a consonant < *-P 2nd sg. �Nµ after a vowel and *-V after a consonant < *-V 3rd sg. zero < *-W 1st pl. *-P after a vowel and *-DP after a consonant < *-PH or *-PR 2nd pl. *-\ after a vowel and zero after a consonant < *-WH 3rd pl. *-Q < *-QW Moreover, the vowel of the stem-final syllable was apocopated in the 3rd sg. form, and in the thematic flexion also in the 1st sg. and 3rd pl. forms and analogically in the 1st pl. form. The 2nd sg. form escaped the apocope because it adopted a middle ending (cf. below). The historically attested paradigms are the following:


[32]

1st sg. EHU�L HWX� 2nd sg. EHUH�U�� HWX�U� 3rd sg. HEHU < *HEKHUHW HW < *HG W� 1st pl. EHU�DNµ WX�DNµ 2nd pl. EHU �Nµ, EHU�LNµ�� HWX�Nµ� 3rd pl. EHU�LQ�� HWX�Q The endings can be explained as follows. The 1st sg. ending *-Q < *-P after a consonant merged with the 3rd pl. ending -Q < *-QW and was therefore eliminated. The thematic ending *-X < *-RP merged with the corresponding primary ending *-X < *- and was probably extended to the athematic paradigm before the apocope. This development induced the spread of the primary athematic ending *-PL to the thematic present. The vowel of the stem-final syllable was *L in five types of aorist: (a) root aorists in *- - and *-L-, e.g. HGL < *HGL�X < *H�GK -. (b) root aorists in *-P- and *-Q-, e.g. HNL < *HNLP�X < *H�J � HP-. (c) derived aorists in *- -, cf. Gr. yµ9v�v, Sl. P Q [ . (d) sigmatic aorists with *- - in the root, cf. Skt. ºW L BN, Sl. Y V . (e) sigmatic aorists of derived stems in *-L-, cf. Sl. YR]L[ . When the vowel of the stem-final syllable was apocopated in the 3rd sg. form, it was reinterpreted as part of the ending in the 1st sg. and 3rd pl. forms, where it spread to the thematic flexion, and in the 2nd pl. form, where it was at first limited to the athematic flexion. The 2nd sg. ending �Nµ was eliminated when -Nµ became the characteristic plural marker. In the nominal flexion, the nom. and gen. sg. ending *-Nµ < *�V was eliminated for the same reason (cf. Pedersen 1905a: 215-221 [= Kl.S., 77-83]). I do not subscribe to de Lamberterie's judgment that “ le génie de H. Pedersen fait preuve de trop de virtuosité pour emporter la conviction” (1979: 325). The postconsonantal ending *-V merged with the sigmatic aorist marker. The ending -U < *-UR was taken from the middle paradigm (see below). The 1st pl. ending does not show the expected *P. I assume that the primary thematic ending *-RPRP replaced the secondary thematic ending *-RPR before the raising of *R to *X before *P and the loss of the final nasal. The final part *-PX of the resulting ending replaced the secondary athematic ending *-PH before the loss of *P before *X. After the apocope, the postconsonantal ending *-D < *-DPX spread to the thematic flexion. Finally, the 1st and 2nd pl. forms received an additional -Nµ from the primary athematic endings. 5. The PIE middle endings underlie the Arm. middle aorist and imperfect forms. Elsewhere I have argued that we can reconstruct the following endings for the proto-language (1981a: 128):

2Q�WKH�$UPHQLDQ�SHUVRQDO�HQGLQJV 37

[33]

2nd sg. transitive *�VW+R� intransitive *�W+R� 3rd sg. transitive *�WR�� intransitive *�R� 3rd pl. transitive *�QWUR�� intransitive *�UR In Armenian, as in Tocharian (cf. ibidem: 133f.), final *-UR was extended to the transitive 3rd sg. ending: *-QW : *�QWUR = *-W : *-WUR. The intransitive 3rd sg. and pl. endings were subsequently replaced with *-WR and *-QWR after the model of the active and transitive middle endings. These developments may have been common to a larger dialectal area. Since *W became characteristic of 3rd sg. forms, the 2nd sg. endings were replaced with *-VR, which was created on the analogy of the active endings. When intervocalic *V was lost, final *-UR spread to the 2nd sg. ending, so that we arrive at the following set of Proto-Armenian middle endings: 2nd sg. transitive *-UR intransitive *-R 3rd sg. transitive *-WUR intransitive *-WR 3rd pl. transitive *-QWUR intransitive *�QWR The 2nd sg. ending *�UR was generalized in the active and middle aorist and imperfect indicative. The transitive 3rd sg. ending *�WUR is found in the imperfect endings - U and �LZU, which I regard as dialectal variants from *�HWUR. The intransitive 3rd sg. and pl. endings *-WR and *�QWR are found in the middle aorist endings �DZ�< *-DWR and *DQ < *�DQWR or *�QWR after a consonant. It is generally assumed that the imperfect marker -L- reflects a preterital stem of the copula (e.g., Pedersen 1905a: 213 [= Kl.S., 75], Meillet 1936: 126, Jasanoff 1979: 140) or the PIE optative suffix (e.g., Winter 1975: 114). I think that outside the 3rd sg. form, the imperfect was obtained by simply adding the athematic aorist endings that were discussed in the preceding section to the present stem. Compare the following paradigms: 1st sg. impf. EHUHL� aor. HGL� 2nd sg. EHUHLU�� HGLU� 3rd sg. EHU U�� HG < HGK W� 1st pl. EHUHDNµ�� HGDNµ 2nd pl. EHUHLNµ�� HGLNµ 3rd pl. EHUHLQ�� HGLQ As was pointed out above, -L- represents the vowel of the stem-final syllable in the majority of athematic aorists. It was reinterpreted as part of the ending when it was apocopated in the 3rd sg. form. The discrepancy between the 1st sg.


[34]

middle aorist ending -D\ and the imperfect ending of D-stems -D\L betrays a chronological difference: the imperfect is a recent formation. 6. The 2nd pl. ending -DUXNµ is found next to -D\Nµ in the middle aorist indicative, where it was borrowed from the middle aorist imperative. As Pedersen pointed out already (1905a: 213f. [= Kl.S., 75f.]), it is most probably a pluralization of the 2nd sg. middle aorist imperative ending -LU with substitution of -D- for -L- on the basis of the indicative paradigm. I think that -XNµ was taken from the personal pronoun *\XNµ (cf. section 3 above). The 2nd sg. ending -LU is the expected reflex of the transitive middle ending in the athematic aorist, as was argued in the preceding section. The zero ending of the middle imperatives DUL ‘get up’ , QLVW ‘sit down’ , HUWµ ‘go’ , KD\HDFµ ‘look’ may represent the intransitive 2nd sg. ending *-R < *-VR. The zero ending of the active imperative developed regularly from the PIE thematic ending *-H, pl. - Nµ < *-HWH�V. The prohibitive imperative in -U belongs to the present system and cannot be connected with the middle aorist imperative in -U for two reasons. On the one hand, the prohibitive ending -U is common to the active and the middle, whereas the aorist imperative ending is limited to the middle voice. It is unlikely that the middle ending was generalized in the present imperative, where we find the active pl. ending �\Nµ�<�*�WH�V� On the other hand, the prohibitive -U is limited to the singular while the aorist imperative U is also found in the plural. I therefore stick to the traditional view that the final U of the prohibitive imperative represents a particle which can be compared with Gr. ·� (cf. Meillet 1936:119), e.g. EHUHU < *EKHUH�UD. [I now think that �Fµ� is the expected reflex of the PIE. sigmatic aorist marker *�V� in Armenian (see Kortlandt 1995a [this vol., 107ff.]). On the apocope of the stem vowel see also Kortlandt 1987b [this vol., 79ff.] and 1999 [this vol., 129f.]. On the reanalysis of �Nµ as a plural marker see also Kortlandt 1985c [this vol., 66f.]. For the loss of *P before *X cf. the dat.pl. ending (see Kortlandt 1984a [this vol., 49]). For the 2nd pl. pronoun see also Kortlandt 1998a [this vol., 122ff.].]

39

Notes on Armenian historical phonology III: K-*

J.A.C. Greppin has argued that Arm. K- originated not only from *S-, *V-, and possibly *N �

-, but also from the PIE laryngeals (1973: 37ff.). His study was severely criticized by P. Considine (1979). Here I will clarify my position with respect to the discussion. The disagreement about Arm. K- seems to be largely determined by two factors: the lack of a chronological perspective and the insufficiency of the evidence. Both factors require some comment. Greppin remarks that “since IE *V- passed to Arm. *K- and then finally to zero, we must acknowledge that Arm. K-, derived from IE *+-, had a different phonetic value than pArm. *K- from IE *V” (1973: 62). This is doubtless correct. Considine adds that Arm. K- from IE *S- must fit into the picture (1979: 358). As I have pointed out earlier in these “Notes” (1976, 1978a) [this vol., 1ff., 20ff.], several problems of Proto-Armenian phonology can be solved if the developments are considered in terms of features rather than segments. Since PIE *V and *S were voiceless sounds, it is reasonable to assume that their reflex in Armenian was also voiceless, at least up to a certain stage. On the other hand, the vocalization of the PIE laryngeals shows that these sounds were at least partly voiced in the proto-language and that the voiceless reflexes which we find in some languages are the result of a devoicing process. It is clear from the pres-ervation of K- from *+� in KRW ‘odour’ and the loss of *K- from *S- in RWQ ‘foot’ that voiced *+- cannot have merged with voiceless *K- before the loss of the latter in this position. The Armenian material which is suitable for comparison with other IE languages is not only small, but also insufficiently available. In a recent article (1982), Greppin has drawn attention to five words which occur both with and without initial aspiration before 1000 A.D.: �K�DJDQLP ‘put on clothes’, �K�DOHZRU ‘grey-haired’, �K�DUEHQDP ‘become drunk’, �K�DU U ‘plough’, and �K�RJL ‘spirit, soul’. These words have [-, which is the phonetic reflex of original K�, in the south-eastern dialects, where secondary K- appears as K-, e.g.

* Reprinted from 6WXGLD�&DXFDVLFD 5 (1983), 9-16.

[10]


in� �K�DP Q ‘all’ , �K�HUHN ‘evening’ . Greppin infers, no doubt correctly, that theforms with and without initial aspiration represent different original dialects. It follows that we cannot limit ourselves to the classical language when dealing with IE correspondences. Indeed, we urgently need a comparative grammar of the Armenian dialects. The starting-points of the following discussion are the relative chronology of sound changes which I have presented elsewhere (1980b) and the threefold reflex of PIE *S-, *V-, and *+-, viz. K-/[-, zero/[-, and zero/K-. Another prerequisite is a statement of the conditions under which PIE *R yielded D in Armenian, a problem which was discussed by Meillet (1894: 153-7), Pedersen (1900c: 93-100), Grammont (1918: 223-225), Greppin (1973: 9-36), and others. In the present context I will limit myself to stating my own view: unstressed *R was weakened to D in open syllables unless it was followed by the reflex of *Z or by a syllable which contained another R, e.g. ORJDQDP ‘bathe’ , NRJL ‘butter’ , KRYLZ ‘shepherd’ , RáRUP ‘pity’ , RáRNµ ‘prayer’ , RáRUN ‘smooth’ , RUR ‘lamb’ , RURU ‘sea-gull’ , also PR]L ‘calf’ (from *-]]-), PR DQDP ‘forget’ (from *-UQ- or *-Uã-). This development was more recent than the syncope of *L and *X, cf. R]QL ‘hedgehog’ , RUFDP ‘vomit’ . It suggests that J from *Z lost its labialization much later than I proposed in my chronology (1980b: 104 [this vol., 31]). The rise of D\� from PIE *RL- belongs to the earliest stages of Proto-Armenian, e.g. D\W ‘cheek’ (ibidem: 105 [this vol., 32]). It is generally accepted that PIE *S- yielded K-, e.g. in KLQJ ‘five’ , KHUX ‘last year’ , KD\U ‘father’ , KXU ‘fire’ (Godel 1975:67). This *K- was lost before *R, e.g. RWQ ‘foot’ , cf. KHW ‘footstep’ (Meillet 1936: 30). The latter development was posterior to the raising of *R to X before a nasal, as is clear from KXQ ‘ford’ from *SRQW-. I have dated the lenition of *S- to a bilabial fricative to stage 10 of my chronology (1980b: 100 [this vol., 28]) and its further development into K- to stage 12 (ibidem: 102) [this vol., 29]. Before pretonic L, this K- yielded \- in \LVXQ ‘fifty’ , probably at a recent stage. Since the raising of *R to X before a nasal was anterior to the lenition (ibidem: 101) and the loss of PIE *S- before R is most easily understood if the bilabial articulation had weakened but not yet been lost, I date the loss between stages 10 and 12. Since the reduction of *R to D in open syllables was posterior to the syncope at stage 22 (ibidem: 104), the absence of K- is regular in DVU ‘fleece’ , which must be derived from *SR X (Meillet 1936: 82), and in DOLNµ ‘wave’ , which can be identified with Gr. %#��²+. The existence of KDOHZRU next to DOHZRU ‘gray-haired’ shows that there was a second ablaut grade in Armenian, as there was in Greek. It corroborates Greppin's otherwise unjustified assumption that Armenian preserved different ablaut grades of this root (1973: 30, fn. 17). Before the rise of K- from *S-, PIE *V- became *K- at stage 5 of my chronology (1980b: 99 [this vol., 27]) and was mostly lost around stage 10

[11]

1RWHV�RQ�$UPHQLDQ�KLVWRULFDO�SKRQRORJ\�,,,� -h 41

(ibidem: 101) [this vol., 28f.], e.g. Dá ‘salt’ , HZWµQ ‘seven’ , Rá ‘healthy’ , XVWU ‘son’ . In my view, this K- was preserved before L, cf. KLQ ‘old’ , KLPQ ‘foundation’ , KLZWµ�µMXLFH¶��FI��'åDXNMDQ��,W�IROORZV�IURP�KLQ, which is the reflex of PIE *VHQ-, and from ]JHQXP, which is identical with Gr. {��/µ��, that the loss of *K from *V was posterior to the raising of *H to L before a nasal. The derivation of KHFDQLP ‘mount’ from *VHG- (Pedersen 1905a: 206 [= .O�6��@��'åDXNMDQ��*UHSSLQ��OHDYHV�ERWK�K- and -F- unexplained and must be abandoned, while LQM ‘leopard’ is probably of non-IE origin. The reflexive pronoun LZU ‘his’ is probably the unstressed form of *HZU: it represents the PIE genitive *VHZH, which developed into *HZ and then received -U- from the other pronouns (cf. Weitenberg 1982, section 11). The existence of KDUEHQDP next to DUEHQDP ‘become drunk’ suggests that Armenian preserved two ablaut grades of this root. While *K- would be lost before a vowel in the expected reflex of Slavic *VHUEM (Trautmann 1923: 294) and Alb. JMHUE, it would be preserved before U in the expected reflex of Lith. VUHEL�, which would receive a prothetic vowel at stage 14 of my chronology (1980b: 103 [this vol., 30]), cf. D X ‘brook’ from *KUXKL-, PIE *VUXWL-. I think that *K- was also preserved before a syllabic U and that the attested zero reflex was taken from a present tense which was identical with the Albanian form. The derivation of Arm. K- from PIE *N �

- cannot be maintained. I agree with Pisani (1950: 181f.) that the interrogative stem R- must be derived from *\R-, not *N � R�. Moreover, I think that the same holds for XU ‘where’ and KLP ‘why’ , cf. Polish MDN ‘how’ , corresponding to Russ. NDN. As in the case of *K- from *V-, the reflex of *\- was apparently retained before L and lost elsewhere. It was evidently lost after the preposition ] in ]L, ]LQþµ ‘what’ (cf. Meillet 1936: 38). As in the case of KLQJ ‘5’ and \LVXQ ‘50’ , it is probable that the preservation of K- was regular in monosyllabic forms only. Chronologically, the loss of *\- can be identified with the assimilation of *L to the following vowel in the genitive RUGZR\ of RUGL ‘son’ , which must be dated between the apocope at stage 16 and the syncope at stage 22, probably toward the very end of this period (cf. Kortlandt 1976: 99 [this vol., 8] and 1980b: 103f. [this vol., 31f.]). The derivation of - from *\- (Meillet 1936: 52) is not convincing. Now we turn to the reflex of the PIE laryngeals (cf. also Polomé 1980, especially 25-29). Greppin has argued that word-initially the phonetic reflex of the laryngeals is K- before a vowel and H-/D-/R- before a consonant, including syllabic U (1973: 47ff.). In my view, the former half of this statement must be limited to *+ � H- and *+ � H- and the latter half must exclude *+Z-. Armenian is in this respect half-way between Greek, where an initial laryngeal is lost before a vowel and vocalized before any consonant, and Hittite, where *+ � - yields K- also before Z and syllabic U. Incidentally, the Hittite evidence must be used with caution because we have to reckon with considerable analogical levelling, e.g.

[12]


KHQN]L ‘distributes’ , where the comparison with OIr. pFHQ ‘necessity’ and Gr. w�9�� requires the restoration of the root vowel (cf. Pedersen 1938: 184), and DãDQ]L ‘they are’ , DGDQ]L ‘they eat’ , where the initial laryngeal was apparently restored on the basis of Hã]L ‘is’ , H]]L ‘eats’ , PIE *+ � HVWL, *+ � HGWL.� Elsewhere I have proposed the view that the timbre opposition between the laryngeals was neutralized in the neighbourhood of PIE *R (1980c: 128). Since there is no reliable evidence for PIE initial vowels, we have to start from the four sequences *+ � H-, *+ � H-, *+ � H-, and *+R- in accounting for the initial vowels of the daughter languages. In view of the fact that *+ � - did not colour a following H, we can interpret *+ � H- and *+R- phonemically as /e-/ and /o-/. It is therefore not surprising that *+ � - and *+ � - were preserved as K- before *H, while *+ � - was lost, as they all were before *R. Armenian offers examples of *+ � H- in HNQ ‘came’ , HP ‘am’ , HV ‘I’ , HZ ‘and’ , H]U ‘shore’ , HUJ ‘song’ , perhaps HSµHP ‘cook’ , *+ � H- in KDQ ‘grandmother’ , KDWDQHP ‘cut’ , KDZ ‘grandfather’ , KDJDFLP ‘become addicted’ , KDUDZXQNµ ‘field’ , KDZ ‘bird’ , *+ � H- in KRYLZ ‘shepherd’ , KDFµL ‘ash tree’ , KRW ‘odour’ , perhaps KXP ‘raw’ , *+R- in R ‘rump’ , RUE ‘orphan’ , RUMLNµ ‘testicles’ , RURU ‘sea-gull’ , RVW ‘branch’ . The latter reflex merged with the zero grade of *+ � -, which is probable in RVNU ‘bone’ , Gr. ±)-z#�. The zero grade of *+ � can be distinguished by the absence of K- in D\U ‘cave’ , DU ‘bear’ , DUFDWµ ‘silver’ , DUJHO ‘obstacle’ . The Latin cognates DUJHQWXP and DUFH are apparently phonetically regular, whereas XUVXV has a deviating reflex, which was perhaps taken from XUF UH. Old Irish DUW ‘bear’ shows the Italo-Celtic vocalization, cf. also Latin DUPXV, DUGXXV, OIr. DUG, which are identical with Skt. UPi , UGKYi .� As in the case of *S- and *V-, the coexistence of forms with and without K- points to the preservation of different ablaut grades in roots with an initial laryngeal. An unassailable example is D\Fµ ‘inspection’ , next to KD\Fµ ‘inquiry’ , KD\FµHP ‘beg, beseech’ , which Greppin derives from zero grade and H-grade, respectively (1973: 56). If this interpretation is correct, we can also assume zero grade in D\F ‘goat’ , Gr. �©!, cf. Avestan ]D QD� ‘of leather’ . It seems more probable to me that D\Fµ represents the R-grade form *RLVN , from which OHG. HLVF Q, Eng. DVN is a derivative formation. The H-grade is attested in Latin DHUXVF UH. A second example is DUDZU, DU U, KDU U ‘plough’ , KDUDZXQNµ ‘field’ . The former word is identical with Gr. x'#-'#�, which probably represents *+ � U+ � WURP, and the latter word must be compared with Gr. x'#/'� from *+ � HU+ � ZU/Q-, the heteroclitic flexion of which was preserved in Old Irish DUERU ‘corn’ , gen. DUEDH (cf. Thurneysen 1946: 213). The full grade of Lith. iUNODV was taken from the verb iUWL, while DU- in Latin DU WUXP and DUYXP is ambiguous. A difficult example is �K�DJDQLP ‘put on clothes’ , ]JHVW ‘garment’ , DJDQLP ‘spend the night’ , DZWµ ‘place to spend the night’ . The zero reflex of PIE

[13]

1RWHV�RQ�$UPHQLDQ�KLVWRULFDO�SKRQRORJ\�,,,� -h 43

*+- before *Z is phonetically regular, as is clear from JRP ‘am’ , Hitt. KXLã]L ‘lives’ from *+ � ZHVWL, and JRJ ‘say’ , which represents *+ � ZHJ � K-, cf. Latin YRYH , Gr. J¾3#µ��. For DZWµ I assume that it shows the phonetic development of the R-grade form *RXWL-, cf. D\W ‘cheek’ , Gr. #« #+, and DZFDQHP ‘anoint’ from *RQJ �

-, Latin XQJX . This development can be dated to stage 2 of my chronology (1980b: 99 [this vol., 27]), cf. also DáDZWµNµ ‘prayer’ next to RáRNµ and DáDþµHP, with loss of *Z as in DþHP ‘grow’ , Latin DXJH . In my view, the Armenian evidence shows unambiguously that we have to start from two homonymous stems *+ � ZHV- ‘spend (the night)’ and ‘put on (clothes)’ , derived from two homonymous roots *+ � HX- ‘live’ and ‘wear’ . Greek and Hittite removed the homonymy by eliminating the initial laryngeal from the word with the latter meaning, probably after its loss before *R in the R-grade of the root, cf. Gr. {��/µ�� without D-, Hitt. XHãWD without K-. Note that the R-grade of the root is widely attested, whereas the corresponding H-grade can only be inferred (Pokorny 1959: 346). The form KDJDQLP presents the expected reflex of *+ � HZ-. The form without K must have arisen under the influence of either R-grade derivatives or prefixed formations, e.g. D DJDVW ‘curtain’ (Meillet 1936: 77). In the case of DJDQLP ‘spend the night’ , the R-grade is attested in DZWµ. A less clear instance is �K�RJL ‘breath, spirit, soul’ . If the word was not borrowed from an unknown language, it must be connected with KRY ‘wind’ , which can be compared with Gr. x�µ� ‘blow’ from *+ � ZH+ � PL (Winter 1965: 102) or rather with Lith. SXþL� ‘blow’ . The form without K- represents the R-grade of the root, whereas the analogical introduction of K� requires the existence of a different ablaut grade, which may be attested in KHZDP ‘pant’ . Though KDULZU ‘hundred’ can phonetically be derived from *+ � HUL-, which fits Gr. w'�Kµ²+ ‘number’ (cf. Hamp 1976), it is more probably a borrowing from an unknown source. The zero reflex of PIE *+- before *Z disproves the possibility of deriving the initial consonant of NµVDQ ‘twenty’ from *+Z- (Winter 1965: 107). Elsewhere I have argued that we have to start from PIE *GZLG WL with a glottalic *G which was preserved in the separate languages (1982). As in Greek, the buccal features of the initial consonant were lost by dissimilation, a development which yielded an initial glottal stop. Unlike Greek, where the initial laryngeals merged into a glottal stop after colouring the epenthetic vowel which originated from their vocalization, Armenian apparently preserved the ini-tial laryngeals longer as voiced consonants. This hypothesis accounts not only for the zero reflex before *Z and the preservation as K- in positions where *V- was lost, but also for the fact that the laryngeals did not merge with the glottal stop before *Z in NµVDQ. I assume that the glottal stop devoiced the following *Z in the same way as *K- from *V-, e.g. in NµR\U ‘sister’ , at stage 7 of my chronology (1980b: 100 [this vol., 28], cf. especially Kortlandt 1976: 96 [this

[14]


vol., 5]). The rise of the glottal stop must have been anterior to the monophonemicization of *GZ at stage 2 (1980b: 99 [this vol., 27], cf. 1976: 98 [this vol., 7]) and may have been a dialectal Indo-European development. The hypothesis that PIE *+ � H- yielded Arm. KD- is in contradiction with the common identification of DFHP ‘bring’ with Latin DJ and Gr. x�7. Greppin concludes that in this word “ there appears to be no laryngeal since the Armenian cognate (DFHP) is unaspirated” (1973: 64). I agree with Considine that “ one would like to be reassured that his selective use of the [laryngeal] theory is also a viable use” (1979: 361). In my view, the initial vowel represents the phonetic reflex of both the R-grade and the zero grade of the root. Note that DFHP is the exact correspondent of Latin JHU from PIE *+ � HV-. Similarly, the absence of K- in XWµ ‘eight’ is at variance with the possible reconstruction *+ � H WR+. This reconstruction is by no means certain, however, because it contains two full grade vowels. Moreover, the initial zero grade which the Armenian form presupposes can easily have been taken from the ordinal, cf. Vedic WXU \D� ‘fourth’ and Gr. -'9%J�� ‘table’ from *N � WZU-, and Old Prussian XVFKWV ‘sixth’ from *VX V-. The influence of the ordinal upon the cardinal is evident in Arm. YHFµ ‘six’ , where the elimination of the initial *K from *V is apparently due to its phonetic loss before the zero grade vowel in the ordinal. This analogical development was evidently posterior to stage 11 of my chronology (1980b: 101 [this vol., 29]) because the new initial consonant did not develop into *J�� Summarizing, we arrive at the following chain of events. The initial laryngeals of *+ � H- and *+R- were lost at an early stage, while *+ � - and *+ � - were preserved as a voiced laryngal glide before PIE *H, which was coloured to *D and *R, respectively. At stage 5, antevocalic *V- became *K-, which was lost around stage 10, except before *L. Initial *S- became a voiceless bilabial fric-ative at stage 10. The fricative was lost before *R and in other positions de-veloped into K- at stage 12. The merger of voiced *+- and voiceless *K- can be dated anywhere between stages 12 and 22. Initial *\- was lost not long before stage 22 except before L, where it merged with *K-. In the classical language we find a trace of complementary distribution between K- and \- in the coexistence of KLQJ ‘5’ and \LVXQ ‘50’ . New \- arose from L before a vowel. After the classical period, \- developed into K-. The distribution of K- in the modern dialects requires further investigation. [See further Kortlandt 1984b [this vol., 54ff.], 1986a [this vol., 68ff.], 1987a [this vol., 75ff.], 1998a [this vol., 122ff.].]

[15]

45

Proto-Armenian case endings* 1. ,QWURGXFWLRQ. One of the striking features of Armenian is the accurate preservation of the PIE case distinctions, in spite of the far-reaching changes in the phonemic make-up of the endings. The standard language of the Armenian SSR is in this respect even more archaic than contemporary Lithuanian, Polish, and Russian, which have best preserved the PIE case forms. Unlike these languages, Armenian was subject to a general apocope of final syllables. The remarkable preservation of the case functions in the historical language suggests that the original system had remained essentially unchanged up to the period which immediately preceded the apocope. In this paper I intend to discuss the form of the case endings at that stage. The chronological background of the present analysis was outlined in my contribution to the Philadelphia conference on Armenian linguistics (1980b [this vol., 26ff.]). In the following, the numbers of the stages will refer to the chronology given there. The rise of *K from V is dated to stage 5, the rise of new fricatives as a result of the lenition to stage 10, the loss of intervocalic and anteconsonantal fricatives to stage 13, the loss of the syllabic resonants to stage 14, the apocope to stage 16, the shortening of fricatives to stops to stage 19, and the syncope to stage 22. The raising of *R to X before a nasal consonant, the development of * into X, and the loss of non-syllabic final nasals were anterior to stage 10. 2. $FFXVDWLYH� The pre-apocope form of the acc.sg. ending was *-L in the L-stems, *-X in the X� and R-stems, *-D in the D-stems, and *-DQ in the consonant stems. The latter ending originated from the rise of an epenthetic vowel before postconsonantal *-Q at stage 14. Original postvocalic *-Q had apparently become a feature of the preceding vowel and subsequently been lost before stage 10.

* Reprinted from ,QWHUQDWLRQDO� V\PSRVLXP� RQ� $UPHQLDQ� OLQJXLVWLFV�� 5HSRUWV >0HåGXQDURGQ\M� VLPSR]LXP� SR� DUPMDQVNRPX� MD]\NR]QDQLMX�� 'RNODG\@� �(UHYDQ�� $1�Arm. SSR, 1984), 97-106.

[98]


There is no evidence for Godel’s view that the final nasal was preserved in the nom.sg. form of the Q-stems in PIE - Q and - Q (1975: 100f.). As Pedersen has conclusively demonstrated, the nom.acc.sg. form of the Q-stems represents the original accusative, not the nominative (1905a: 216f. [= Kl.S., 78f.]). The original nominative was preserved in HU Fµ ‘elder’ and PDUGLN ‘people’ . The generalization of the nom.sg. form in the U-stems and of the acc.sg. form in the Q-stems is a consequence of the merger with the neuter gender, which preserved the final nasal of PIE *-PQ in accordance with the sound laws and never had a final nasal in the U-stems, e.g. DáELZU ‘spring’ , DZU ‘day’ , KXU ‘fire’ . The pre-apocope form of the acc. pl. ending was probably *�LV in the L-stems, *�XV in the X� and R�stems, and *�DV in the D� and consonant stems. It cannot be determined if the neuters had already joined the latter type at this stage. 3. 1RPLQDWLYH� As Pedersen has shown in detail, the Arm. nom.pl. ending -Nµ is the phonetic reflex of PIE postvocalic *-V (1905a: 209-227 [= Kl.S., 71-89]). In my chronology, final *-V became *-K at stage 5, then *-[ at stage 11, and eventually -Nµ at stage 19. This creates a problem with the nom.sg. form of the masc. and fem. L-, X-, and R-stems, which does not end in -Nµ. The ending was evidently eliminated under the combined pressure of several morphological factors. On the one hand, the final consonant characterized the nom.pl. ending of the D�� Q-, and U-stems, which had a zero ending in the corresponding singular form. On the other, the masc. and fem. acc.sg. form was identical with the neuter nom.acc.sg. form in the L-, X-, and R-stems. As in the case of the Q- and U-stems, the elimination of the distinction between nom. and acc.sg. forms is a consequence of the elimination of grammatical gender.1 The analogical development can be compared with the substitution of the acc.sg. for the nom.sg. of the R-stems in Slavic, where the barytone neuters merged with the masculines. In Armenian, the motivation for the substitution was strengthened by the merger of the nom. and acc.sg. forms of the D-stems as a result of the loss of nonsyllabic final nasals. A third factor may have been the zero ending of the demonstrative pronoun *VR, which must have been preserved because it served as a model for the attested demonstrative and interrogative pronouns, e.g. QR\Q ‘that very’ , RY ‘who’ . The main objection which can be put forward against the point of view advocated here is the zero nom.pl. ending of the attributive adjective.2 This objection is easily removed by the assumption that the adjective had pronominal endings, as it had in Germanic, e.g. Gothic EOLQGDL ‘blind’ , cf. GDJ V ‘days’ . The 1. Cf. Lamberterie 1979: 327. 2. Cf. Meillet 1900b: 382 [= 1962: 52] and 1911: 9 [= 1962: 131].

[99]

3URWR�$UPHQLDQ�FDVH�HQGLQJV 47

apocope yielded the attested endings. I think that there is additional evidence for the view that the PIE nom.sg. ending *-V yielded -Nµ in the remarkable abundance of SOXUDOLD� WDQWXP in classical Armenian, the origin of which must probably be sought in deverbal nouns such as HONµ ‘outcome’ , [DZVNµ ‘speech’ . Thus, I arrive at the following reconstruction of the pre-apocope endings. The nom.sg. ending was *-L in the L-stems, *-X in the X-, R-, and masc. and fem. Q-stems, *-D in the D-stems, and *�R in certain pronominal forms. The nom.pl. ending was *-X[ in the X- and R-stems, *-D[ in the D-stems, �H[ in the consonant stems and probably in the *L-stems, and *-H in the adjective (see below). 4. *HQLWLYH� Like Indo-Iranian and Greek, Armenian adopted the pronominal gen.sg. ending *�RV\R in the nominal R-stems. This may have been a dialectal Indo-European innovation. The absence of final *�V from this ending provided in my view the model for the analogical elimination of the final consonant from the endings *-LRV in the L�stems, *�XRV in the X-stems, and *-RV in the consonant stems. The ending -L of the D-stems represents most probably PIE *-L+RV.3 The LD-stems had apparently adopted the ending of the LR-stems before the apocope already in order to remove the homonymy with the L- and D-stems. The ending *�\RV is evidently preserved in the form JHá of JLZá ‘village’ and in WX Q HDQ (with secondary -HDQ) of WLZ ‘day-time’ . The form NQR of NLQ ‘woman’ is probably a pre-apocope extension of gen.sg. *NLQR with the same ending. I wonder if the coexistence of the gen.sg. forms PLR\ and PLR of PL ‘one’ is a vestige of the earlier gender distinction, with -R\ and -R representing the masc. and fem. ending, respectively. In this conception, the ultimate origin of the final - is the antevocalic form *-\- of the L-stem formative, which was frequent with feminine nouns. Thus, I arrive at the following reconstruction of the pre-apocope form of the gen.sg. ending: *�R\R� in the R-stems, *-LR\R in the LR- and LD-stems, �L�\�R in the L- and D-stems, *-XR or *-X\R in the X-stems, and *-R in the consonant stems. The gen.pl. ending -Fµ represents PIE *-VNRP, the pre-apocope vocalism of which is preserved in the form QRFµXQ of QR\Q ‘that very’ .4 The same vowel is found in the gen. form XUXNµ of RNµ ‘somebody’ , which must be derived from *-URP. Thus, I reconstruct *-FµX as the regular pre-apocope ending and *-UX in certain pronominal forms, e.g. in the genitive PHU of PHNµ ‘we’ , cf. Latin QRVWUXP�� 3. Cf. Lamberterie 1979: 328. 4. Cf. Pedersen 1905a: 237 [= Kl.S., 99] and Kortlandt 1978c: 295.

[100]


5. /RFDWLYH� The main problem which poses itself in connection with the loc.sg. forms is the following: why is the ending of the LD-stems different from those of the D�� R-, and LR-stems? It is the only flexion class where the loc.sg. ending differs from both the dat. and the acc. ending. Since the PIE loc.sg. ending �L would undoubtedly merge with the dat.sg. ending *�HL as a result of the apocope, we must start from an extended zero ending in order to explain the loc.sg. form of the LD-stems. There is indeed a zero ending in Vedic loc.sg. Y N , which is the most likely candidate for comparison. The usual connection of the ending -R with Gr. -#K� meets with two fatal objections. First, it is highly improbable that an adverbial derivative of the R-stems should be preserved precisely in the paradigm of the LD-stems. Second, the connection requires the supplementary extension of the new ending with an additional vowel, which was subsequently lost as a result of the apocope. In my view, the ending was created on the analogy of the gen.sg. ending -R from which it probably differed in the apocopated vowel. If my conjecture about gen.sg. PLR\ and PLR is correct, we may suspect that loc.sg. PLXP and PLR exhibit the masc. and fem. pronominal endings. Thus, I tentatively reconstruct the following pre-apocope loc.sg. endings: *-L�\�L in the L- and D-stems, *�X�\�L in the X-stems, *-H in the R-stems (see below), *-L in the consonant stems, and finally *-XPL and *-R L in pronominal forms. The introduction of *-R L into the paradigm of the nominal LD-stems must have been anterior to the generalization of the gen.sg. ending *�R\R� The loc.pl. ending -V represents the PIE ending *-VX, which was apparently preserved in the consonant stems up to the apocope. After a vowel *V was lost and the vowel sequence was subject to contraction, cf. NµR\U ‘sister’ from *VZHV U, pl. NµRUNµ from *VZHVRUHV, and ERN ‘barefoot’ from *EKRVR�. Since the stem vowel was apocopated in the loc.pl. form, the ending must have been restored after the apocope. The pre-apocope form of the ending may have been *-R\X in the R-stems, *-X in the X-stems and probably in the L-stems, perhaps *-R from *-DX in the D-stems. It is possible that the ending *-VX replaced *-X in the L� and X-stems before the apocope already, but in that case the absence of the stem vowel before the case ending would be remarkable. 6. ,QVWUXPHQWDO� The endings can be derived from PIE *-EKL, which received an additional *-V in the plural. The pre-apocope endings were sg. *-EL and pl. *-EL[ after a consonant, and the same endings with *Z instead of *E after a vowel. The absence of PIE *EK from the Armenian dat. and abl.pl. endings is a strong indication that this consonant was originally limited to the instrumental case because such a distinctive case marker would hardly have been eliminated in a language which preserves the PIE case distinctions so well. This point of view is

[101]

[102]


supported by the Gothic adverbs in -ED, which represent earlier instrumental forms. The PIE inst.pl. ending of the R-stems *- LV, which would yield the pre-apocope form *�H[ (see below), may have been eliminated at any stage before or after the apocope. 7. 'DWLYH� The dat. form XPHNµ of RNµ ‘somebody’ �shows that PIE *- L yielded *-H before the apocope, cf. Vedic WiVPDL� The same can be assumed for *�RL in view of XUHNµ ‘somewhere’ , which is apparently a loc. form of the same stem as gen. XUXNµ of RNµ ‘somebody’ .5 Most probably, all final L-diphthongs merged into *-H before they were apocopated. If this is correct, the pre-apocope dat.sg. ending was *-H in all flexion types. It is not probable that final *�HL yielded *-L before the apocope because in that case the dat. and loc.sg. forms would have merged in all flexion types. Indeed, I think that the merger of *- L and *-RL in the R-stems gave rise to the analogical dat.sg. ending *�R\H in order to differentiate it from the loc.sg. ending. The PIE dat.pl. ending was *-PXV according to the Germanic, Baltic, and Slavic evidence.6 This reconstruction is supported by the Armenian material because *P was lost before *X in this language, so that the expected pre-apocope form of the ending would be *-X[.7 The form was apparently replaced with the gen.pl. ending. In Indo-lranian and Italo-Celtic, it was remodelled on the basis of the inst.pl. ending. It is possible that the Greek pronouns 1st pl. * VPH and 2nd pl. *XVPH which are attested in Aeolic acc. xµµJ and ¾µµJ must be derived from dative forms, cf. Vedic acc. DVP Q, \X P Q. 8. $EODWLYH� As Pedersen has shown, the abl.sg. ending - originated from an enclitic particle, perhaps PIE *HWL, which was added to the loc.sg. form after the apocope in order to remove the homonymy of the two case endings (1905a: 221f. [= Kl.S., 83f.]). I conclude that the homonymy resulted from the apocope. In the R-stems, the expected pre-apocope ending *�X which developed from PIE *� G was apparently replaced with *�R\X on the basis of the gen. and dat.sg. endings. This analogical development was particularly well motivated if the other flexion classes had adopted the ending *-X at that time. The theory advanced here is for two reasons more attractive than the alternative view that the abl.sg. ending of the R-stems had been replaced with the gen. sg. ending before the apocope. First, it would be unclear why the two cases have not entirely merged, as they have in Greek and Balto-Siavic, because they were not 5. Cf. Pedersen 1905a: 235-237 [= Kl.S., 97-99]. 6. OHG. WDJXP requires *-PXV, not *-PRV, which would yield **WDJDP. 7. Cf. Kortlandt 1981c: 30 [this vol., 35].

[103]


distinguished in the other flexion classes. Second, the ablative cannot be derived from the genitive in the D� and L-stems, where the stem vowel is absent before the historical ending, e.g. DP ‘year’ , EDQ ‘word’ , gen.sg. DPL, EDQL, abl.sg. DP , EDQ , inst.sg. DPDZ, EDQLZ.8 Here again, as in the explanation of the ending -R in the paradigm of the LD-stems, I think that we have to start from a pre-apocope loc.sg. ending *-L which corresponds with Vedic Y N . This ending apparently served as a model for the creation of a monosyllabic abl.sg. ending in the paradigm of the D-stems, perhaps *�X, which was lost as a result of the apocope. The reconstruction of the PIE abl.pl. ending is a problem. The merger with the dative in Indo-Iranian, Italo-Celtic, and Germanic, and with the genitive in Balto-Slavic, Armenian, and Greek suggests that the original abl.pl. ending contained neither *�EK� nor *�P� and was liable to replacement with a more distinctive ending. I reconstruct PIE *�LRV�� Under the influence of the instrumental, the ending was replaced with *-EK\RV in Indo-Iranian and with *-EKRV in Italo-Celtic, whereas it adopted the *�P� from the dative in Germanic. In my view, Armenian has preserved a remarkable trace of the PIE ending *�LRV in the pronominal forms 1st pl. P Q and 2nd pl. M Q , where - continues earlier *-\-. These are the only forms with a distinct abl.pl. ending in Armenian, cf. 1st sg. LQ Q, 2nd sg. Nµ Q. They cannot be compared with the case forms in �R , which are limited to the singular. The form LQ Q may represent earlier *LP Q, which can be identified with Gothic gen. PHLQD of LN ‘I'.9 Thus, Armenian provides an important link in the reconstruction of the PIE abl.pl. ending. 9. *HQGHU� Several authors have observed that the loss of the PIE gender distinction in Armenian seems to be at variance with the accurate preservation of the case distinctions. As Meillet has remarked, the loss of grammatical gender is really the loss of gender concord in the adjective (1900b: 382 [= 1962: 52]). I have argued above that the neuter had merged with the masculine before the apocope, at least in the singular. The zero nom.pl. ending of the attributive adjective is best derived from masc. *�RL and fem. *-DL, which had merged into *�H before the apocope. The D- and LD-stems had adopted a number of case endings from the L- and LR-stems, respectively. As a result of these developments, the importance of grammatical gender was strongly reduced before the apocope already. When the vowel of the final syllable was lost, the gender distinction was eliminated from the nom. and acc. forms, and subsequently from the other cases. I have suggested above that a trace of the earlier gender distinction has been preserved in the oblique cases of PL ‘one’ , gen.sg. PLR\, PLR , dat.loc.sg. PLXP, PLR . 8. Cf. Godel, o.c., 106 top. 9. Cf. Pedersen 1905a: 226 [= Kl.S., 88].

[104]


10. 6XPPDU\� On the basis of the considerations which have been put forward I arrive at the following tentative reconstruction of the pre-apocope endings: sg. pl. nom. �X, �D, �L, �R, -L �X[, �D[, �H[, �H, -D gen. �R�L�X�\�R, �R, ��R� R, -UX ��R�D�L�X�FµX, -UX dat. �R�L�X�\�H, ��XP�H, -�R� H ��R�D�L�X�FµX, -X�[�� acc. �X, �D, �L, �DQ, -L �XV, �DV, -LV, -D abl. �R�L�X�\�X, �X, -�R� X -�R�D�L�X�FµX, - R�[� inst. -R�D�L�XZL, -EL -R�D�L�XZL[, �EL[ loc. �H, �L�X�\�L, -�XP�L, -�R� L ��R\�X, �R, -VX [On the merger of nominative and accusative singular forms see further Kortlandt 1985c [this vol., 63ff.]. For the pronouns see also Kortlandt 1983a [this vol., 52f.] and 1998a [this vol., 122ff.]. On the flexion of PL ‘one’ and the loss of grammatical gender see further Weitenberg 1984. For the gen.pl. ending cf. Kortlandt 1978c. On the flexion of the X�stems see also Kortlandt 1985a [this vol., 60ff.] and 1996a [this vol., 117ff.].]

[105]

52

Demonstrative pronouns in [Balto-Slavic,]

Armenian [and Tocharian]* [...] Starting from a single original demonstrative pronoun (VR, WR�), a single anaphoric pronoun (H/L�), and three deictic particles ( L, DX, DQ), we can derive the Slavic and East Baltic paradigms in a straightforward way. The main problem which remains to be discussed is the rise of the Prussian forms. I think that the clue to a possible solution is provided by the Armenian system. Unlike the other Indo-European languages, Armenian possesses three anaphoric pronouns, corresponding to the three demonstrative pronouns: D\V ‘hic’ VD ‘is’ VR\Q ‘idem’ D\G ‘iste’ GD ‘is’ GR\Q ‘idem’ D\Q ‘ille’ QD ‘is’ QR\Q ‘idem’ The radical element of the three series is identical with the one of Lith. ãuV, WjV, DQjV. As Holger Pedersen has pointed out (1905b: 34f. [= Kl.S., 38f.]), the starting-point of the Armenian development must be sought in the forms GD, which represents WR�, and D\Q, which is the reflex of DQLR�, Skt. DQ\i� ‘other’ . The latter word is a derivative of the proposed particle DQ, which has been preserved in the adverb DQG ‘there’ , cf. also Skt. iQWDUD� ‘other’ , Lith. DxWUDV ‘second’ . Thus, we can derive the Arm. system from the one which has been reconstructed for Balto-Slavic. The correctness of this analysis is corroborated by the forms DZDVLN ‘voici’ , DZDGLN, DZDQLN ‘voilà’ . Pedersen already identified �VLN with Slavic VLF as a derivative in �NR� of PIE. L (1982: 41 [= Kl.S., 45]). It gave apparently rise to a correlating form DQLN, which in its turn supplied the initial vowel for the creation of a form DVLN. Similarly, the initial vowel of DVW ‘here’ was evidently taken from DQG ‘there’ . The initial part DZ� may represent the PIE. particle DX. Perikhanian’ s hypothesis that the forms under discussion are of Iranian origin does not explain the �L� (1966: 22). Of course, Iranian K X is itself

* Reprinted from 6WXGLHV�LQ�6ODYLF�DQG�*HQHUDO�/LQJXLVWLFV 3: 'XWFK�FRQWULEXWLRQV�WR�WKH��WK�LQWHUQDWLRQDO�FRQJUHVV�RI�VODYLVWV��/LQJXLVWLFV (1983), [311-]316-317[-322].

[317]

'HPRQVWUDWLYH�SURQRXQV�LQ�$UPHQLDQ 53

a conflation of VR and DX.Û [...] [The absence of *K� before *DQ� is regular after prepositions. It is possible that DQG ‘there’ actually reflects *\� plus *DQ� (see Kortlandt 1998a [this vol., 122ff.]). The threefold article of Macedonian Slavic was probably borrowed from the local variety of Armenian in the 13th century.]

54

PIE. *+- in Armenian* It is clear from Brian Joseph’s discussion (1984) that lack of explicitness on my part has given rise to some misunderstanding. In view of this, I gladly accept the Editor’s invitation to elaborate on the subject. 1. Since the publication of R.S.P. Beekes’ monograph on the development of the PIE. laryngeals in Greek (1969), the triple representation of “ schwa” is a solved problem. Lindeman's little book (1982) has convinced me that nothing can be gained from reopening the issue. It must be noted that the Greek evidence is crucial, and that the material from other languages can only play a subsidiary role. More specifically, the Armenian material must be considered in the light of the Greek evidence. For this reason I shall not go into a discussion of the basic hypothesis that there were three PIE. laryngeals *+ � , *+ � , *+ � , which yielded a “ prothetic” H�, D�, R� in Greek, a hypothesis which I consider as well-proven as a hypothesis in linguistic reconstruction can be. Incidentally, note that in Lindeman's chronology (p. 69f.) the early coloring of PIE. *H by a contiguous laryngeal is gratuitous because it is subphonemic only, and that he agrees to date the merger of the laryngeals after the separation of Anatolian from the original linguistic community, i.e. after the disintegration of Proto-Indo-European. 2. Lindeman calls my assumption that there was an opposition between *+ � H- and *+R- in Proto-Indo-European “ wholly circular” . There are three reasons why this is obviously not the case. 2.1. The comparative evidence points to three laryngeals and two vowels, viz. *H and *R� to neutralization between the laryngeals in the neighborhood of the rounded vowel, and to the absence of word-initial vowels. If we start from the four sequences *+ � H-, *+ � H-, *+ � H-, *+R-, which can be reconstructed without taking the Armenian evidence into account, it is clear that the initial laryngeal is automatic in the first and in the last instance, but not in *+ � H- and *+ � H-. It is therefore hardly surprising if we find a zero reflex of *+ � before *H and of any laryngeal before *R and a nonzero reflex of *+ � and *+ �

* Reprinted from $QQXDO�RI�$UPHQLDQ�/LQJXLVWLFV 5 (1984), 41-43.

[42]

3,(��*H-�LQ�$UPHQLDQ 55

before *H. I claim that this is indeed what we find in Armenian. 2.2. We know more about PIE. ablaut than can be inferred from the Armenian material. If KRW ‘odor’ can be identified with Latin RGRU, which is an original V-stem, we expect H-vocalism in the root. If we find D\Fµ ‘inspection’ next to OE. VFH ‘idem’ < *R\VN and KD\FµHP� ‘beg’ next to Latin DHUXVF UH ‘idem’ < *D\VRV-, we expect *R in the former and *H in the latter word. 2.3. The same distribution can probably be established for Hittite, e.g. KDUS- ‘separate’ < *+ � HUEK-, DUN� ‘mount’ < *+RU K�H\- (Oettinger 1979: 415), Gr. ±'Q��²+, ²'3�+. The matter cannot be pursued here. 3. Joseph's objection that I deal with an abstract phonological representation is based on a misunderstanding. I do not generally recognize ordering of underlying forms as a viable mode of linguistic argumentation. The correct view is stated in Joseph's footnote 5: *+ � H differed phonetically from *+R in the proto-language. In my article on *+ � R I have called attention to the Shuswap language, which offers the closest typological parallel for the PIE. laryngeals. That language distinguishes phonetically between e.g. + � V�+ � HV�W ‘sunny’ and + � R\�W ‘withered’ , where the labialization of the pharyngeal resonant is automatic before the rounded vowel of the latter word, but constitutes an independent distinctive feature before the H- and zero grade vocalism of the former (cf. Kuipers 1974). A wider knowledge of more exotic languages could save Indo-European linguistics quite a bit of misguided argumentation. 4. The PIE. word for ‘name’ can be reconstructed as *+ � QH+ � �PQ (Latin Q PHQ, Hitt. O PDQ, Skt. Q PD, Arm. DQXQ), obl. *+ � Q+ � �PHQ- (Gr. ²�#µ�, OIr. DLQP, OPr. HPPHQV, Russ. tPMD, Alb. HPsU). The initial labialized laryngeal is warranted by Gr. �I�/µ#+ ‘nameless’ < *Q�+ � -. No conclusions can be based on the Laconian name �/µ��'�-? �+: if it is cognate at all, the initial laryngeal may have lost its rounding by dissimilation, as the second laryngeal did in Toch. A xRP, B xHP, which point to *�+ � �Q P�Q�. It is clear from KHUX ‘last year’ that Armenian did not lower H to D before X in the following syllable.

5. The word DUDZU ‘plough’ can be identified with Gr. x'#-'#� < *+ � U+ � �WURP, with zero grade before the thematic suffix. The H-grade is attested in KDUDZXQNµ ‘field’ , which can be identified with OIr. DUERU ‘corn’ < *+ � HU+ � �ZU, gen. DUEDH < *+ � U+ � �ZHQV. The dialectal variant KDU U apparently adopted the H-grade of the verb in the same way as Lith. iUNODV and Latin DU WUXP adopted the verbal stem. The original zero grade was preserved in Lith. uUNODV ‘oar’ < *+ � U+ � �WORP, Gr. y'J-µ²�. 6. The form KDJDQLP ‘put on clothes’ shows the expected reflex of + � HZ- and can be compared with Lith. D WL, Slavic -XWL, Latin -X , whereas the variant DJDQLP shares the R-vocalism of Umbrian DQRXLKLPX ‘induitor’ <

[43]


*-+RZ�H\-. The former variant is original in the verb, while the latter must have arisen under the influence of R-grade derivatives and prefixed formations such as D �DJDVW ‘curtain’ . In a similar way, Greek and Hittite eliminated the initial laryngeal of *+ � ZHV- ‘put on clothes’ on the analogy of the R-grade root form *+RZ-, e.g. Hitt. ZHã- ‘dress’ next to XQX- ‘adorn’ , thereby removing the homonymy with *+ � ZHV- ‘spend the night’ , where the H-grade root form *+ � HZ- was more vigorous, e.g. Gr. ¨�E7, aor. x(Ü)J)�. Armenian shows expected absence of K- in DZWµ ‘place to spend the night’ < *+RZ- and YD\U�DJ ‘living in the field’ , and analogical elimination in DJDQLP ‘spend the night’ . [See further Kortlandt 1986a [this vol., 68ff.] and 1987a [this vol., 75ff.]. On Anatolian see now my forthcoming article in 2USKHXV 13-14 (2004).]

57

Notes on Armenian historical phonology IV*

1. PXNQ, MXNQ.� The words PXNQ ‘mouse’ and MXNQ ‘fish’ must be derived from PIE. acc.sg. *PX+VP and *G KX+P respectively (cf. Pedersen 1905a: 215 [= Kl.S., 77], Winter 1965: 104). The structure of these words can be compared with that of RWQ ‘foot’ , GX Q ‘door’ , MH Q ‘hand’ , from *SRGP, *GKXUP, * KHVUP. Since it has been shown that the glottalic feature of East Armenian N is probably very old (cf. Kortlandt 1978a: 15 [this vol., 25]), we must reconsider the possibility that it represents the phonetic reflex of the PIE. laryngeal in PXNQ and MXNQ. If this conjecture is correct, the oralization of the laryngeal is comparable to the same development in English KDFN, TXLFN, VWDNH from *ND+Z-, *J � L+Z-, *VWD+Z- (cf. Lehmann 1965: 216). Oralization of a glottal stop is a well-known development in Danish (cf. Skautrup 1928: 32-33). Werner Winter has proposed that the velar stop in PXNQ and MXNQ is a secondary modification of aspirated Nµ before the suffix -Q and that this Nµ was the phonetic reflex of *-+V in the nom.sg. form (1965: 105). Since it has been shown that -Nµ is the phonetic reflex of PIE. final *-V (cf. Pedersen 1905a: 209-227 [= Kl.S., 71-89], Kortlandt 1980b: 99-104 [this vol., 27-31]), the laryngeal is redundant in this explanation. Moreover, it is improbable that postvocalic Nµ would have been preserved before -Q (cf. Kortlandt 1980b: 102 [this vol., 29f.]). We must therefore consider the possibility that the stem-final spirant in the word for ‘mouse’ was simply lost and that the laryngeal was oralized before the syllabic nasal. I see no objection to this view. It obviously presupposes the syllabification of Vedic acc.sg. GKt\DP ‘thought’ , Y N\jP ‘she-wolf’ , not GHY P ‘goddess’ . 2. DNQ, XQNQ. The reason for Winter to derive the velar stop of PXNQ and MXNQ from earlier Nµ is the alternation in DNQ ‘eye’ , gen.sg. DNDQ, nom.pl. DþµNµ, gen.pl.



DþµDFµ, which represent acc.sg. *RN � P, nom.du. *RN � L+. The initial D- is the phonetic reflex of R- in open syllables (cf. Kortlandt 1983b: 10 [this vol., 40]) and represents both the R-grade and the zero-grade vocalism of the root. The expected nom.sg. form is *RNµ. It is improbable that the velar stop would have been preserved in the oblique cases, cf. JD\O ‘wolf’ (from *-O\-), OHDUG ‘liver’ , Vedic Y NDV, \iN W. I therefore assume that it was restored on the analogy of the type *PXNµ, *MXNµ, acc.sg. PXNQ, MXNQ. The original stop was preserved in the plural, where -þµ- represents the cluster *-N � \-. The word XQNQ ‘ear’ consists of XQ-, which is the expected reflex of acc.sg. *XVP, and -NQ, which was apparently taken from DNQ. The plural DNDQ Nµ remains unexplained. 3. PHUM, PDZUXNµ. The perfect correspondence between PHUM ‘near’ and Greek µz3'� suggests that PIE. *- K- was not depalatalized before -U- (cf. Kortlandt 1978b: 239). This is at variance with the depalatalization of *- - in PDZUXNµ ‘beard’ , which is identical with Vedic Pi UX, cf. Irish VPHFK ‘chin’ from *VPHN . It now seems to me that Armenian probably shared the development of Balto-Slavic and Albanian, e.g. Lith. VPmNUDV, Alb. PMHNsU. The palatal stop of PHUM shows that we have to reconstruct *PH KVUL and that the medial *-V- was preserved at the time of depalatalization before resonants. The word is evidently a compound of *PH- ‘in’ and the locative of the word for ‘hand’ *- KVUL. 4. VUXQNµ, N XQN�� The preservation of the palatal in VUXQNµ ‘shin-bone’ and the absence of metathesis in N XQN ‘crane’ show that there was a vowel between the initial consonant and the following -U-. This vowel was either L from * or X from * . The comparison with Latin FU V and JU V points to an original vowel alternation in the root. This is corroborated by the depalatalized stop of N XQN, which must have been taken from the cases where the palatal was in immediate contact with the following -U-. The full grade is attested in Lith. JpUY , SCr. å U Y, Gr. �z'��#+, the zero grade in OHG. NUDQXFK. I wonder if N XQN represents a metathesized form of acc.sg. *JUX+QP, with oralization of the laryngeal as in PXNQ and MXNQ. Similarly, the velar suffix of the Germanic word may be due to oralization of the laryngeal. It is noteworthy that the velar suffix is limited to Armenian and Germanic. The coexistence of derivatives in -X- and -Q- is reminiscent of Latin FRUYXV ‘raven’ and FRUQ [ ‘crow’ . I think that we have to start from PIE. masc. * HU+X-, metathesized to * HUX+- before a consonant, and fem. * HU++Q-, cf. Welsh JDUDQ with double zero grade. The syncopated vowel in VUXQNµ and N XQN was probably an analogical lengthened grade *- - as in Gothic T QV ‘wife’ and Greek �%�' ‘liver’ , where

[10]

1RWHV�RQ�$UPHQLDQ�KLVWRULFDO�SKRQRORJ\�,9 59

the original full grade is preserved in Old Irish EHQ, Slavic åHQD, Latin LHFXU, Vedic \iN W. The model for the introduction of the lengthened grade vowel was provided by monosyllables such as PIE. * UG ‘heart’ , *P PV ‘meat’ , *GL ZV ‘heaven’ . The different generalization of the initial consonant in VUXQNµ and N XQN is a consequence of the phonetic developments and their relative chronology (cf. Kortlandt 1980b: 100-103 [this vol., 28-30]). In the case of VUXQNµ, the original alternation between antevocalic palatal and anteconsonantal velar was changed by the lenition and subsequent loss of the velar, which motivated the generalization of V-. In the case of N XQN, the velar stop was not lost, but only subject to metathesis shortly before the apocope. I think that the velar had been generalized at that stage and that the metathesized stem form was eliminated in order to remove the awkward alternation. [See also Kortlandt 1985c [this vol., 63ff.]. The laryngeal was not oralized in *�L+P, as is clear from the original accusative PLQ of PL ‘one’ (see Kortlandt 1994b [this vol., 98]). ]

[11]

60

Arm. DUWDZVU ‘tear’ * 1. The Indo-European word for ‘tear’ has the following manifestations: (1) *D UX in Vedic i UX, Toch. A NlU, Lith. mãDUD. (2) *GD UX in Greek 9�'/, Old Irish GpU, Old Breton GDFU-,

Gothic WDJU, OHG. ]DKDU� (3) *GUD �U�X in Arm. DUWDZVU, pl. DUWDVXNµ, Latin G�ODFULPD

(Hamp 1972: 294), OHG. WUDKDQ. (4) Hittite LãKDKUX. The derivation of the Armenian word presents a difficulty. 2. Meillet (1936: 55) and Pedersen (1906: 408 [= Kl.S., 186]) derived the Armenian word from *GUD X-, with epenthesis of the lost *X in the singular, and adduced DZU ‘day’ as a parallel. This presupposes a reconstruction * P U of the latter word. I agree with Greppin (1978: 284) that it is better to start from * P , which is in accordance with Homeric �µ�'. Even if we start from * P U, the word does not provide evidence for X-epenthesis because *P was lost before *X in Proto-Armenian (cf. Kortlandt 1981c: 30 [this vol., 35]). 3. Greppin nevertheless assumes X-epenthesis in DUWDZVU and adduces PDZUXNµ ‘beard’ , Vedic Pi UX, as a parallel (1978: 286). This position is untenable, first because there is no X-epenthesis in the plural DUWDVXNµ, and second because the palatovelar was lost in PDZUXNµ. The stop in the latter word was evidently depalatalized, as it was in Lith. VPmNUDV and Alb. PMHNsU, and subsequently subject to lenition and development into -Z-, like the dental stop in PDZU ‘mother (gen.)’ , Gr. µ�-'²+ (cf. Kortlandt 1985b: 10 [this vol., 58] and 1980b: 102 [this vol., 29f.]). 4. Though Greppin's position cannot be maintained, the comparison with PDZUXNµ points the way to a solution of the problem. It must be noted that DUWDVXNµ is the regular outcome of *GUD X-, while the expected reflex of *GUDNUX is **DUWDZU. The latter form seems to have adopted the -V- of the former. This shifts the problem to the establishment of the original formation and the flectional paradigm of the word. 5. It is clear that the Indo-European words for ‘tear’ cannot be reduced

* Reprinted from $QQXDO�RI�$UPHQLDQ�/LQJXLVWLFV 6 [Fs. Djahukian] (1985), 59-61.

[60]

$UP��artawsr µWHDU¶ 61

to a single reconstruction. If we regard *GD UX as a simplification of *GUD UX, we are still left with three forms with a stronger mutual resemblance than could have been produced by accident. Moreover, the reconstruction *GUD UX cannot be recast in laryngealist terms and must therefore be rejected as a Proto-Indo-European form. Both formal and semantic considerations lead to the conclusion that the word is a compound, the second component of which is *D UX. 6. Formally, the second component *D UX can hardly be separated from the root *+ � H - ‘sharp, bitter’ , zero grade *+ � - in composition. The vocalization of the laryngeal suggests that the first component ended in an obstruent. The preservation of the simplex in Indo-Iranian and Baltic may be due to the fact that a laryngeal between obstruents was not vocalized in these languages. The development of * before U in Hittite LãKDKUX is possibly the same as in KDUãQL ‘head (loc.)’ and KDUã]L ‘plows’ , Vedic U D L, NiU DWL, N iWL.� 7. Semantically, the word for ‘tear’ is typically a compound ‘eye-water’ in many languages. For *GUD UX both ‘eye-water’ and ‘water-bitter’ have been proposed (cf. Van Windekens 1976: 14). The obvious source seems to be *GU �+ � UX ‘eye-bitter’ with dissimilation of the first * , cf. Gr. '9�#+ ‘eye’ , Vedic G - ‘sight’ , Irish GHUF ‘eye’ , GUHFK ‘face’ from *G . The compound may have competed with the phrase *GU RP� + � H UX in late Proto-Indo-European.1 8. The Hittite word for ‘eye’ ãDNXL, pl. ãDNXD, is generally derived from *VHN �

-, OHG. VHKDQ ‘see’ . The expected word for ‘tear’ is now *VN � �+ � UX, which apparently yielded LãKDKUX by loss of the labialization and assimilation of the velar to the following laryngeal. There is a trace of the first component in Old Irish URVF ‘eye’ , pFRVF ‘appearance’ from *SUR�VN � RP, *LQ�NRP�VN � RP. 9. Like other X-stems, *GUD UX received an ending -U in Armenian. The development of * into -V- in DUWDZVU shows that *GUD UXU was dissimilated to *GUD XU before the depalatalization of * before *U, which was a dialectal Indo-European development. The addition of -U� must therefore be dated to Proto-Indo-European times. This early date is corroborated by the plural *�GU�D UXQD, which can be reconstructed on the basis of Latin G�ODFULPD (Hamp 1972: 296), OHG. WUDKDQ, Lith. mãDUD, and Toch. B DNU QD. The Arm. plural DUWDVXNµ was evidently built upon the singular *GUD XU. 10. We have now arrived at a reconstructed paradigm *GUD XU, obl. *GUDNUX-, which yielded an alternation *-DVXU, *-DZUX- in pre-apocope Armenian. This alternation was not quite exceptional, as is clear from *PD��LU ‘mother’ , *KZH�K�XU ‘sister’ , obl. *PDZU-, *KZHKU- on the one hand, and *VLUX-, 1. Professor C.J. Ruijgh points out to me that Homeric J'�- denotes specifically a fixed look, as opposed to *+ � HN � � ‘see’ , and that we may have to conceive of *GUN�DNUX as tears that gather when the eye is fixed upon an event which rouses emotions.

[61]


VUXQNµ ‘shin-bone’ , obl. *�H�UX-, Latin FU V on the other. Most probably, insertion of -Z- from the oblique cases yielded a pre-apocope form *DUWDZVXU, from which the oblique case forms adopted the -V- after the apocope. The plural DUWDVXNµ was not affected. [A similar analysis has now been proposed by Pinault (1997).]

63

The syncretism of nominative and accusative singular in Armenian*

1. Apart from the personal pronouns 1st sg. HV, acc. LV, and 2nd sg. GX, acc. NµH], classical Armenian does not make a distinction between nom.sg. and acc.sg. forms. The origin of this syncretism may be threefold: (1) the nom. and acc.sg. forms merged phonetically; (2) the nom.sg. form replaced the acc.sg. form; (3) the acc.sg. form replaced the nom.sg. form. In the following I intend to show that all of these developments must be assumed for Proto-Armenian. 2. As Godel points out, the choice between the three hypotheses on the origin of the syncretism “ depends on how one figures out the fate of final nasals” (1975: 99). There can be no doubt that syllabic final nasals are reflected as -Q, cf. HZWµQ ‘seven’ , LQQ ‘nine’ , WDVQ ‘ten’ , also VHUPQ ‘seed’ , JHáPQ ‘fleece’ , HUPQ ‘heat’ . The words MH Q ‘hand’ , RWQ ‘foot’ , DNQ ‘eye’ , GX Q ‘door’ can

therefore be identified with the PIE. acc.sg. forms, cf. Greek 3J¥'�, %² �. The correctness of this identification is strongly supported by the archaic instrumental form MHUE ‘by hand’ , cf. MHUEDNDO ‘prisoner’ , and by the absence of the nasal in the plural stems MH -, RW-, Dþµ-, GXU-. Nonsyllabic final nasals are reflected as -Q in monosyllables, cf. NµDQ ‘than'; LQþµ ‘something’ , Latin TXDP, Vedic Nt �FLG. It is clear from these examples that original -P and -Q merged in Proto-Armenian. 3. If the identification of -Q with the acc. ending is correct, it follows that the consonant stems mentioned above generalized the acc.sg. form as the casus rectus. There is a clear motivation for this generalization in the merger with the neuter gender because neuter consonant stems in -PQ were particularly frequent. The mere absence of a distinction between nom. and acc. forms in the neuter does not suffice to motivate the elimination of the distinction in

* Reprinted from 5HYXH�GHV�eWXGHV�$UPpQLHQQHV, n.s. 19 (1985), 19-24.


non-neuter paradigms, however, especially in view of its preservation in the plural. It is therefore probable that the nom. and acc.sg. forms merged phonetically in one or more flexion classes. This suggests that nonsyllabic final nasals were lost in polysyllables, so that the acc.sg. ending of the vocalic stems was lost and the acc.sg. form could be identified with the bare stem in a number of flexion classes. As a result, the nom. and acc.sg. forms merged in the -stems. 4. In contradistinction to the instances cited above, the nom.sg. form was generalized in the U-stems, cf. KD\U ‘father’ , PD\U ‘mother’ , HáED\U ‘brother’ , NµR\U ‘sister’ , WD\JU ‘husband's brother’ , GXVWU ‘daughter’ , D\U ‘man’ . All of these nouns denote persons, where the nominative was more frequent than in the case of inanimate objects. Incidentally, neuter nom.acc.sg. forms in -U were frequent, e.g. DZU ‘day’ , DáELZU ‘spring’ , KXU ‘fire’ , DUWDZVU ‘tear’ , RVNU ‘bone’ . 5. According to Godel (1975: 99f.), the weak point in the hypothesis that nonsyllabic final nasals were lost while syllabic final nasals were preserved as -Q is that the syllabic nasals must have developed into -DQ� cf. HZWµDQDVXQ ‘seventy’ , WDVDQRUG ‘tithe’ . The objection does not hold because it does not take into account the relative chronology of the sound changes. The word JD\O ‘wolf’ < *J � DO\R- < *J � O[R- < *ZON � R- shows that the elimination of the syllabic resonants was posterior to the lenition of the intervocalic velar stop, but anterior to the metathesis of the resonants (cf. Kortlandt 1980b: 103 [this vol., 31]). Nonsyllabic final nasals in polysyllables had probably been lost by the time of the lenition of intervocalic stops: I assume the early rise of nasal vowels and their subsequent denasalization, cf. XV ‘shoulder’ , Vedic i VD . The rise of nasal vowels was posterior to the rise of K from V, but anterior to the lenition which eliminated the nasal in D\U ‘man’ , DZU ‘day’ , Greek w�=', �µ�'. 6. It has been objected against the hypothesis that nonsyllabic final nasals were lost in polysyllables that final -Q seems to be preserved in JD Q ‘lamb’ , Greek (Ü)�'=� (e.g., Godel 1975: 100f.). The objection does not hold because the -Q can easily be explained as a result of analogy (cf. Pedersen 1905a: 217 [= Kl.S., 79], Meillet 1936: 80). It has been suggested that the original nom.sg. form of the non-neuter Q-stems was preserved in HU Fµ ‘elder’ , PDUGLN ‘people’ , DQGUDQLN ‘first-born’ (Pedersen 1905a: 218 [= Kl.S., 80]), Dá LN ‘maid’ , PDQXN ‘child’ (Meillet 1936: 80). These nouns denote persons. Nouns denoting inanimate objects apparently adopted the neuter ending -Q. 7. It follows from what has just been said that MLZQ ‘snow’ and VLZQ� ‘column’ do not match Greek 3�I� and � 7�, the expected reflex of which is +ML, +VLZ. The Armenian words must be derived from the acc.sg. form, Greek

[20]

[21]

7KH�V\QFUHWLVP�RI�QRPLQDWLYH�DQG�DFFXVDWLYH�LQ�$UPHQLDQ 65

3�²��, � #��. Note that MLZQ is indeed the phonetic reflex of * KLH�RPP because final -P yielded -Q and -PQ yielded -ZQ after a vowel, cf. PU LZQ ‘ant’ , gen.sg. PU PDQ. The radical nasal of MLZQ is preserved in MPH Q ‘winter’ , which must be derived from acc.sg. * KLPHUP, like DPD Q ‘summer’ from *VP+ � HUP. For the same reason, the suffix -WµLZQ cannot be identified with Latin -WL but must be compared with the acc.sg. form -WL QHP. It appears that -QP shared the development of -PP. Similarly, the derivation of ãXQ ‘dog’ and WXQ ‘house’ from * Z Q and *G P is questionable: these words can equally be derived from the acc.sg. forms * ZRQP and *GRPP, cf. Vedic Y QDP, Greek Æµ�. The latter derivation is actually to be preferred in view of the syllabification of RV. X 86.4 XY , Greek �E7�, and Welsh FL (not +SL), the expected Armenian reflex of

which is +VX, not ãXQ. 8. The instances adduced so far point to a tendency toward generalization of the nom.sg. form in the case of persons and the acc.sg. form with other nouns. This is reminiscent of the development in French and Bulgarian. Modern French has generally lost the original nominative, which survives in SUrWUH, DQFrWUH, SHLQWUH, WUDvWUH, V°XU (Arm. NµR\U), and in the doublets FRSDLQ, JDUV, VLUH and the pronoun RQ beside FRPSDJQRQ, JDUoRQ, VHLJQHXU, KRPPH (cf. Price 1971: 98). Since the development seems to be parallel, we may look for doublets in Armenian as well. 9. Weitenberg has recently suggested that the “ additional” (\DZHODNDQ) -Q of modern eastern dialects represents the PIE. acc.sg. ending in a number of instances (1985). The word DVWá ‘star’ has an additional -Q in Goris and Lo i, and gen.sg. -DQ in Van. The dialectal form DVWHáQ may represent acc.sg. *DVWHáQ < *+ � VWHOP. The word þLZá ‘branch’ has an additional -Q� LQ�0HáUL� DQG�.DUDEDJ��and gen.sg. -DQ in Muš. Its archaic instrumental þHáE shows that it is an old consonant stem, to be compared with Greek �z�#+ ‘missile’ , �J�²�� ‘needle’ , Lith. JHOXRQuV ‘sting’ (cf. Pedersen 1906: 393 [= Kl.S., 171]). 10. Conversely, the coexistence of classical Q-stems and forms without -Q�in the same dialects or in the classical language points to original doublets. In the case of DVHáQ ‘needle’ , forms without -Q� DUH� IRXQG� LQ�0HáUL�� $JXOLV�� DQG�Goris, and in the classical bird name DVáDNWXFµ“ needle-beak” . The word is evidently identical with Slavic RVOD ‘whetstone’ and Old English HJOH ‘awn’ . Weitenberg's theory offers an ingenious explanation for the word NDWµQ ‘milk’ , which is represented in the dialects as ND[Fµ� �0HáUL�� $JXOLV�� DQG� NDWµQ (e.g., Karabag): these forms can be derived from nom. *NDáFµ < *JONWV and acc. *NDáWµQ < *JONWP��UHVSHFWLYHO\��FI��'åDXNMDQ�� 11. The hypothesis that nonsyllabic final nasals were lost in

[22]


polysyllables allows us to identify the nom.acc.sg. form of the R�, L�, and X-stems with the original accusative. This identification actually imposes itself because the phonetic reflex of final -V is -Nµ in Armenian, as Pedersen has shown in detail (1905a: 209-227 [= Kl.S., 71-89], cf. Kortlandt 1980b: 99-104 [this vol., 27-31]). The sigmatic nom.sg. ending was evidently eliminated under the combined pressure of several morphological factors, the most important of which was the sigmatic nom.pl. ending. Here again, French offers a parallel. The regular paradigm of the R-stems in Old French was the following: sg. (Latin) pl. (Latin)

nom. PXUV� < PXUXV�� PXU� < PXUL� acc. PXU� < PXUXP� PXUV� < PXURV Modern French has substituted the accusative for the nominative both in the singular and in the plural. The same development must evidently be assumed for Armenian in the singular. The model for the replacement was provided by the D�, Q�, and U-stems, where the sigmatic ending was limited to the plural. 12. A second factor in the development was the merger with the neuter gender (cf. Lamberterie 1979: 327). When nonsyllabic final nasals were lost in polysyllables, the acc.sg. form of the masc. and fem. L- and X-stems merged with the neuter nom.acc.sg. form, which was identical with the bare stem. Moreover, the acc.sg. form of the -stems merged with the nom.sg. form, which was again identical with the bare stem. The substitution of the accusative for the nominative in the paradigm of the R-stems has a parallel in Slavic, where the barytone neuters merged with the masculines (cf. IlOLþ-6YLW\þ��-130 = 1979: 108-114). 13. A third factor in the elimination of the sigmatic nom.sg. form was the asigmatic ending of the demonstrative pronoun *VR, which served as a model for the attested demonstrative and interrogative pronouns, e.g. QR\Q ‘that very’ , RY ‘who’ . The introduction of pronominal endings into nominal paradigms is also found in gen.sg. -R\, loc.sg. -XP, -R (cf. Kortlandt 1983d: 67). 14. The main objection which can be raised against the hypothesis that final -V developed phonetically into -Nµ is the zero nom.pl. ending of the attributive adjective (cf. Meillet 1900b: 382 [= 1962: 52] and 1911: 9 [= 1962: 131]). The objection is easily removed by the assumption that the adjective had pronominal endings, as it had in Germanic, e.g. Gothic EOLQGDL ‘blind’ beside GDJ V ‘days’ . If this assumption is correct, the apocope yielded the attested endings.

[23]

7KH�V\QFUHWLVP�RI�QRPLQDWLYH�DQG�DFFXVDWLYH�LQ�$UPHQLDQ 67

15. As in the case of the consonant stems, one may look for survivals of the original nom.sg. form. Since the sigmatic nominative merged phonetically with the plural, such instances may be expected to have adopted the plural inflexion. I think that the sigmatic nom.sg. ending lies at the root of the remarkable abundance of pluralia tantum in Armenian, many of which have an abstract meaning. In the locative, these nouns take demonstratives and possessives in the singular, e.g. L�NHDQVQ�NµXP ‘in your life’ (cf. Meillet 1911: 12 [= 1962: 134]). The adoption of the plural inflexion was facilitated by the absence of a clear number distinction in abstract concepts. The origin of the type must probably be sought in deverbal L-stems, e.g. HONµ ‘outcome’ , [DZVNµ ‘speech’ . 16. To summarize, we have to assume the following Proto-Armenian developments: (1) The nom. and acc.sg. forms merged phonetically in the -stems; neuters had a single nom.acc. form in the proto-language already. (2) The nominative replaced the accusative in KD\U, PD\U, HáED\U, NµR\U, WD\JU, GXVWU, D\U, HU Fµ, PDUGLN, DQGUDQLN, Dá LN, PDQXN; it joined the plural in abstract nouns such as HONµ and [DZVNµ. (3) The accusative replaced the nominative in the R�, L�, X-, and consonant stems, e.g. MH Q, RWQ, DNQ, GX Q, MLZQ, VLZQ, �WµLZQ, ãXQ, WXQ, JD Q, MPH Q, DPD Q. (4) The nom. and acc.sg. forms are both preserved in DVWá, *DVWHáQ, þLZá, *þHáQ, *DVá, DVHáQ, *NDáFµ, NDWµQ. [See also Kortlandt 1996a [this vol., 117ff.]. Stempel objects that we do not find �Nµ < *�V in the nom.sg., gen.sg., and 2nd sg. endings (1994). It seems to me that this argument has adequately been countered above, cf. also Kortlandt 1984a [this vol., 46f.] on the noun and 1981c [this vol., 34f.] on the verb.]

68

Armenian and Albanian* 1. Ten years ago I wrote an article, which was published five years ago (1980a), on the earliest relations between Albanian and Armenian. I defended the thesis that these languages developed from contiguous Indo-European dialects, constituting a link between Balto-Slavic on the one hand and Greek on the other. Since the recent publication of Huld’s etymological glossary (1984), the Albanian material has become more easily accessible to students of Indo-European, not only through the rich bibliography, but especially as a result of the laudable inclusion of ample information on the dialects, which often play a decisive role in the reconstruction of Proto-Albanian. We must be grateful to the author for placing this material at our disposal and can only hope for an amplified future edition to replace Meyer’s etymological dictionary (1891). In the following I intend to reconsider the earliest relations between Armenian and Albanian in the light of Huld’s monograph. I shall not discuss the points of disagreement with Huld’s views which have no bearing on the subject. 2. Huld lists the following correspondences between Albanian, Armenian, and Greek: (1) Alb. sQGsUU < *DQGsUUs, Arm. DQXU , Gr. ²�J�'#+ ‘dream’ , which must be derived from *+ � QU\R-, *+ � Q U\R-, *+ � QHU\R-, respectively. Huld assumes for Albanian full grade in the root, which is in my opinion incorrect (cf. below). (2) Alb. HGK, Arm. D\F, Gr. �©! ‘goat’ . The problem with this word is the absence of K- in both Armenian and Albanian. Hamp suggests that the K- was lost by analogy with Alb. GKL ‘goat’ if the latter is related to Lith. P T, Skt. DMi �(1965: 140), but the comparison of this word with German =LHJH < *GLJK , Laconian ?��, Arm. WLN is preferable (cf. Huld 1984: 59).1 Peters suggests that Gr. �©! may represent both full grade *+ � HL - and zero grade *+ � L - (1980: 83). Though this would solve the problem of Armenian and Albanian, it is improbable, as Peters demonstrates himself (1980: 113). The word is

* Reprinted from /D�SODFH� GH� ODUPpQLHQ� GDQV� OHV� ODQJXHV� LQGR�HXURSpHQQHV (Leuven: Peeters, 1986), 38-47. 1. Laconian ?�� may be a mistake for �©�� < *DLJ\D.

probably an “ orientalisches Kulturwort” (Pokorny 1959: 13, Solta 1960: 405): I [39]

$UPHQLDQ�DQG�$OEDQLDQ 69

reconstruct dialectal IE. *D\ -. (3) Alb. sQGH, Gr. x�K#+ ‘flower’ , Arm. DQG ‘field’ , to which can be added Skt. iQGKD ‘plant’ , Toch. A QW, B QWH ‘plain’ , all of which can be derived from *+ � HQGKRV� *+ � QGKHV-. The equation is not evident. (4) Alb. WKHOOs ‘deep’ , Arm. VR\O ‘cave’ , Gr. �#¥�#+ < *NyZLORV ‘hollow’ , which Huld derives from * RX+OR- ‘curved in’ (1978). Since the evidence for a laryngeal origin of Gr. -�- is insufficient, I think that Armenian and Albanian have preserved the original form of the word, while Gr. -��#+ is of analogical origin, cf. %#��?�#+ ‘spotted’ , Skt. SH DOi . (5) Alb. YMHKsUU, YMHKUUD, Arm. VNHVXU, Gr. x�/' ‘husband's mother’ represent *VZH XU , which was created on the basis of *VZH XURV ‘husband's father’ (Alb. YMHKsUU, YMHKUUL) to replace PIE. *VXH UX+V. Huld assumes a metathesized form with irregular stress *XpV XURV in order to explain the Albanian words (1979: 198). This DG� KRF solution is not attractive. Pedersen's statement that intervocalic *- U- yielded -KU� in the word for ‘mother-in-law’ (1900a: 339) is in conflict with the development in PMHNsU < *VPH U ‘chin, beard’ , where the palatovelar stop was regularly depalatalized before the following resonant (cf. below). I think that we must assume an intermediate stage *VZHQU V, where the medial fricative was adopted from the word for ‘father-in-law’ , and subsequent development into -KU-. A similar analogy accounts for the development of the initial cluster (cf. below). (6) Alb. LP, Arm. LP, Gr. yµ²+ ‘mine’ . The Albanian form represents L� HP (Pedersen 1900b: 341 [= Kl.S., 2]). I think that it reflects PIE. zero grade *+ � P, which relates to acc. *+ � PH, poss. *+ � PRV in the same way as Gr. -E ()E) and Alb. X < *VX ‘self’ relate to acc. *WXH, *VXH, poss. *WXRV, *VXRV. (7) Alb. QMHUt, Arm. D\U, Gr. w�=' replace PIE. *XL+URV as the common word for ‘man’ . The Alb. suffix is a late addition. PIE. *+ � QHU- was a honorific term. (8) Alb. WKsUt < * RQLG , Arm. DQLF, Gr. �#�?+ ‘lice eggs’ have full grade in the root, as opposed to the zero grade in OE. KQLWX <* QLG , SCr. JQM GD (with depalatalization and voicing of the palatovelar stop before the nasal resonant, palatalization of the latter in accordance with Trubetzkoy 1930: 392, and “ lengthening” in accordance with Winter's law). The loss of the initial stop in Armenian betrays an earlier alternation * RQ-, * Q-, the latter of which yielded *Q- at stage 12 of my chronology (1980b: 102 [this vol., 29]), cf. OX, Gr. ��/-²+ ‘famous’ . The zero reflex of the initial stop was evidently extended analogically to the antevocalic position in DQLF, probably at a stage when it still was a weak fricative. (9) Alb. SXQs ‘work’ , Arm. SµR\Wµ ‘zeal’ , Gr. )%#/ = ‘haste’ . This comparison is of little value. The connection with the Arm. word cannot be maintained (cf. Pedersen 1905a: 200 [= Kl.S., 62]). (10) Alb. TDM < * ODXQ\ , Arm. ODP, Gr. ��(Ü)?7 ‘cry, weep’ . This equation

[40]


[41]

seems preferable to the usual derivation of the Arm. word from *O -: I reconstruct * OH+ � X-. For the loss of the semivowel in Armenian cf. DþHP < *DX- ‘grow’ , Rþµ < *RX� ‘not’ . (11) Alb. ]MDUP ‘fire’ , Arm. HUP, Gr. KJ'µ²+ ‘warm’ , Toch. A lUPH ‘heat’ have H-grade vocalism, while the R-grade is found in Baltic, Germanic, Latin, and Indo-Iranian.21 (12) Alb. GRUs <* K VU , Arm. MH Q, Gr. 3J?', Toch. A WVDU, B DU, Hittite NHããDU represent the PIE. word for ‘hand’ . (13) Alb. QMs, Arm. PL, Gr. µ?�, Toch. A Vl , B VDQD ‘one’ . This derivation of the Alb. word seems preferable to its comparison with Skt. DQ\i . 3. In addition, the following comparisons were adduced by Pedersen (1900b), Jokl (1963), and Solta (1960): (1) Alb. DUGKt, Arm. RUWµ ‘vine’ . This connection cannot be maintained (cf. Weitenberg 1975: 68f.). (2) Alb. ]RJ, Arm. MDJ ‘young bird’ . The initial stop is reminiscent of the one in Alb. ]s, Arm. MD\Q ‘voice’ . (3) Alb. EXWs ‘soft’ , Arm. EXWµ < *-NW- ‘blunt’ , to which can be added Lith. EXN�V ‘blunt’ , OIr. ERFF ‘soft’ . (4) Alb. V, Arm. þµ, Gr. #½�? ‘not’ . (5) Alb. TH < NOH ‘was’ , Arm. OLQLP ‘become’ , Gr. �� 7 ‘incline’ . (6) Alb. KD, KsQJUD, Arm. XWHP, NHUD\ ‘eat, ate’ , which must be derived from *-+ � HG-, *-J HU+ � -. The prefix of the present tense may represent *+ � HX- in Albanian and *VRP- in Armenian (cf. Skt. iYD�JLU-, ViP�DG-). (7) Alb. KHUGKH, Arm. RUMLNµ ‘testes’ , also Gr. ²'3�+, MIr. XLUJH. I reconstruct *+ � HU KL-, *+ � U KHL- (cf. below). (8) Alb. GKMDPs ‘fat, lard’ , Arm. WDPXN ‘humid’ , Gr. �µ²+ ‘fat’ . This connection cannot be maintained (cf. Huld 1984: 60). (9) Alb. J�O�LVKW ‘finger’ , Arm. þLZá ‘branch, finger’ , which must be an old consonant stem in view of its archaic instrumental þHáE. (10) Alb. GUHGK, GURGKD, Arm. GD QDP, GDUMD\ ‘turn, -ed’ . (11) Alb. YsWK, Arm. JLQG ‘ring’ . This comparison can hardly be maintained. (12) Alb. WULP ‘brave’ , Arm. WµDUP ‘fresh’ , which can be derived from the same root as Gr. -z'�� ‘soft’ , (13) Alb. GKDOOs ‘buttermilk’ , Arm. GDO, GD\O ‘colostrum’ , The comparison of the Alb. word with Gr. �9�� ‘milk’ (e.g., Huld 1984: 57) does not explain the initial stop, which should have been depalatalized if it represents * O-. It is interesting that both Pedersen and Jokl changed their mind about the Alb. word, abandoning their earlier support for the connection with Gr. �9�� (cf. Hamp 2. The H-grade may have been taken from the root noun *J � HU- which is attested in SCr. åDU beside åDUD ‘heat’ .


1960: 277, Pedersen 1905a: 201 [= Kl.S., 63]). The Arm. word must be derived from *GKDOL-, which is a possible origin for the Alb. word, cf. also Gr. K��= ‘teat’ . 4. According to Huld, the “ most striking feature that separates Albanian from the Satem languages is the environmentally conditioned depalatalization of Indo-European palatals before non-vocalic resonants. This feature, a corollary to the Albanian neutralization of palatals before non-vocalic sounds, is entirely absent in Indo-Iranian and Armenian and quite different from the internal variation found in Baltic and Slavic, e.g. Lith. DNPX} ‘stone’ , mãPHQ\V 'edge (of a tool)’ , both from *$H PHQ-” (1984: 160). Elsewhere I have suggested that the conditions for the depalatalization of palatovelars before resonants were essentially identical in Albanian and Balto-Slavic and independent of the syllabification (1978b: 242). In spite of Huld's statement to the contrary, the Balto-6ODYLF� GHYHORSPHQW� LV� QRW� VSRUDGLF�� ýHNPDQ� OLVWV� �� LQVWDQFHV� RI�“ Gutturalwechsel” in Baltic and Slavic, not counting the onomatopoeic cases (1974). The Lith. formation in -PHQ� is productive, e.g. UDã� ‘write’ , UDãPX} ‘letter’ , VNDLW- ‘count’ , VNDLWPX} ‘figure’ , similarly mãPHQ\V ‘blade’ from the root Dã- ‘sharp’ , but DNPX}�= Skt. i P ‘stone’ . In Indo-Iranian the depalatalization was apparently limited to the position before PIE. *U (cf. Meillet 1894b: 297f.). I now think that the depalatalization operated in Armenian in the same way as in $OEDQLDQ��FI��'åDXNMDQ��.RUWODQGW��E��gOEHUJ�� (1) Arm. PDZUXNµ ‘beard’ , Alb. PMHNsU, Lith. VPmNUDV ‘chin’ . (2) Arm. DQLF ‘louse egg’ , Alb. WKsUt, Gr. �#�?+ (cf. above). (3) Arm. OX ‘famous’ , Gr. ��/-²+, Skt. UXWi . (4) Arm. OLQLP ‘become’ , Alb. TH ‘was’ , Gr. �� 7 ‘incline’ (cf. above). (5) Arm. ODP ‘weep’ , Alb. TDM, Gr. ��(Ü)?7 (cf. above). (6) Arm. *NDáFµ, NDWµQ ‘milk’ , Gr. �9�� (cf. Weitenberg 1985). (7) Arm. N XQN ‘crane’ , Gr. �z'��#+, German .UDQLFK. (8) Arm. DUJHO ‘obstacle’ , where German 5LHJHO ‘bolt’ points to an original vocalization *+ � UH O-, cf. also Lith. UmNWDV ‘key’ . The Arm. vocalism was probably adopted from a derived verb with zero grade root vowel which can be compared with Gr. w'�z7, Latin DUFH . (9) Arm. NµHUHP, NµRUHP ‘scratch’ , where the initial stop was taken from the zero grade form of the root, cf. Gr. �J?'7, ��'-. (10) Arm. VUXQNµ ‘shin-bone’ reflects * UX-, while Latin FU V represents the oblique stem form. (11) Arm. PHUM < *PH KVUL ‘near’ , Gr. µz3'�, shows that the medial *-V- had not been lost at the time of the depalatalization, as is also clear from Lith. L ãPDV

[42]


[43]

‘spit’ , Gr. �¨3µ= < *- VP- (cf. de Saussure 1892b: 90f. [= 1922: 459]).32 5. Another characteristic feature of the satem languages is the retraction of PIE. *V to ã after L, X, U, N (cf. Pedersen 1895). This is a shared innovation of Balto-Slavic, Indo-Iranian, Armenian (YHãWDVDQ < *- V- ‘sixteen’ ), and probably Albanian. The problem with the latter language is that ã is the usual reflex of PIE. *V in most environments, e.g. VKRK ‘I see’ , VKWDWs ‘seven’ , VKWsUSLQM < *VHUSpQ- ‘vermin’ , PLVK ‘meat’ (Arm. PLV), GKDVKs < *GRG+ � V- ‘I gave’ , VKRVK ‘sieve’ , VKWHJ ‘path’ , DVKW ‘bone’ , sVKWs ‘is’ , MDVKWs ‘outside’ (Gr. y3K²+). Word-initially before a stressed vowel the phonetic reflex of PIE. *V is a voiced palatal stop, e.g. JMDUSsU < *VpUSHQ- ‘snake’ , JMDVKWs ‘six’ , JMDOSs ‘butter’ (Gr. |�%#+, Toch. A lO\S, B DO\SH), JMXPs ‘sleep’ . The intervocalic reflex of PIE. *V is zero, e.g. DMR < *-V ‘she’ , also X < *VX ‘self’ . The word NRKs ‘time’ cannot be connected with Russ. þDV because the initial stop is not palatalized and the ending represents *- . The word YHVK ‘ear’ (Gr. #À+) shows the anteconsonantal reflex of the root-final obstruent. The reflex of PIE. *VN is K, e.g. KLH ‘shade’ (Gr. )�� , Toch. B VNL\R), KHGK < *VNHXG � ‘throw’ (OE. VF RWDQ), DK < *RVNR- ‘beech’ (Arm. KDFµL), verbs in -K < *-VN . As in Armenian, PIE. *V was lost before nasals in Albanian, e.g. MDP ‘am’ , PMHNsU ‘beard’ . 6. Both Armenian and Albanian have a particular reflex of PIE. *VZ: (1) Arm. NµXQ ‘sleep’ , Skt. VYiSQD , but zero grade in Gr. Â%�#+, Alb. JMXPs. (2) Arm. NµR\U ‘sister’ , Skt. VYiV . (3) Arm. NµLUWQ ‘sweat’ , Alb. GLUVs <*VZLGU-, Gr. ¬ 'I+, full grade in Latvian VYLrGUL, Skt. VYpGD . (4) Alb. GLHOO < *VZHO- ‘sun’ , Gr. {��, J�� ‘sun's heat’ . (5) Alb. GHUJMHP < *VZRUJK- ‘am ill’ , Lith. TFSHÒ, TJ HUJ, Toch. A VlUN, B VDUN�‘illness’ . (6) Alb. YMHKsUU ‘father/mother-in-law’ , Gr. x�/'²+, - < *VZH-. As in the case of JM- < *V-, I assume that G- < *VZ- developed before stressed vowels only (cf. Pedersen 1900a: 290). The word YMHKsUU shows that the initial fricative was lost in unstressed *VZH-, as it was in the clitic X < *VX ‘self’ . (7) Alb. YHWs < *VZH- ‘self’ . (8) Alb. YMHUU ‘hang’ , Lith. TWFSJÒ, TWF UJ ‘weigh’ . This verb must be separated from YpUWL ‘pierce’ and Gr. wJ?'7 ‘lift’ . Elsewhere I have argued that the remarkable development of PIE. *VZ- in Armenian and Albanian can be understood if we assume that the cluster became monophonemic at an early stage in these languages (1980a: 249 [this vol., 18], 1980b: 99 [this vol., 27]).

3. The form �¨!µ' is attested in Myc. acc. pl. DL�ND�VD�PD.


7. Armenian and Albanian also have a particular reflex of PIE. * Z and * KZ: (1) Arm. ãXQ ‘dog’ , Skt. Y . (2) Arm. ã ‘donkey’ , Skt. i YD ‘horse’ . (3) Alb. VRUUs < * Z UQ ‘crow’ , Bulg. VYUiND ‘magpie’ , Lith. ãiUND. (4) Arm. MD\Q ‘voice’ , Alb. ]s, ]k < * KZRQRV, Russ. ]YRQ ‘ringing’ . Here again I assume that the clusters became monophonemic at an early stage in the two languages. The original affricate þ- of Alb. VRUUs is preserved in the Rumanian loanword FLRDU . The alleged reflex V of * before syllabic X (Huld 1984: 144f) cannot be maintained, as Ölberg has convincingly demonstrated (1968). 8. Elsewhere I have argued that PIE. *+ � H- is reflected in Arm. KDQ ‘grandmother’ , KDWDQHP ‘cut’ , KDZ ‘grandfather’ , KDJDFLP ‘become addicted’ , KDUDZXQNµ ‘field’ , KDZ ‘bird’ , PIE. *+ � H- in KRYLZ ‘shepherd’ , KDFµL ‘ash tree', KRW ‘odour’ , perhaps KXP ‘raw’ , PIE. *+R- and vocalized *+ � - in R ‘rump’ , RUE ‘orphan’ , RUMLNµ ‘testicles’ , RURU ‘sea-gull’ , RVW ‘branch’ , RVNU ‘bone’ , XWµ ‘eight’ , and vocalized *+ � - in D\U ‘cave’ , DU ‘bear’ , DUFDWµ ‘silver’ , DUJHO ‘obstacle’ , DUDZU ‘plough’ , DFHP ‘bring’ (1983b: 12ff. [this vol., 41ff.]). I wonder if a similar distribution can be established for Albanian. The following instances are relevant (cf. Hamp 1965, Ölberg 1972): (1) Alb. KDS < *+ � HS- ‘open’ , Gr. w%², Skt. iSD.43 (2) Alb. KLQM, I K < *+ � HX- ‘enter’ , Skt. iYD. This etymology is probably false (cf. Ölberg 1972: 123). (3) Alb. KXW < *+ � HXW- ‘empty’ , Gr. �¾-7+, Gothic DXìHLV. (4) Alb. KLSsQM, K\SLM < *+HXS- ‘mount’ , Gr. Á%z', Skt. XSiUL. This etymology is questionable. (5) Alb. KHUGKH < *+ � HU K- ‘testicle’ , Gr. ²'3�+, Arm. RUM ‘male’ (with R- or zero grade). (6) Alb. KD < *+ � HX- ‘eat’ , Skt. iYD- (cf. above). (7) Alb. KDOs < *+ � HOVQR- ‘alder’ , Latin DOQXV (Huld 1984: 151). (8) Alb. DUGKXU < *+ � RU K�XQ- ‘come’ (part.), Gr. |'3#µ��. (9) Alb. DWKsW < *+ � R - ‘sour’ , Gr. ²�'�+, ±!E+. (10) Alb. DPs < *+ � RGP ‘stench’ , Gr. ± µ=, ±)µ=, Arm. KRW < *+ � HGRV, Latin RGRU. (11) Alb. DK < *+ � RVNR- ‘beech’ , Arm. KDFµL < *+ � HVN- ‘ash’ , OE. ¿VF, Gr. ±!E� ‘beech’ . (12) Alb. DVKW < *+ � RVW- ‘bone’ , Arm. RVNU, Gr. ±)-z#�, Skt. iVWKL. 4. Cf. Latin DE ‘from’ , DSHUL ‘open’ .

[44]


[45]

(13) Alb. D < *+ � RX ‘whether, or’ , DV ‘neither, nor’ , Arm. Rþµ ‘not’ , Gr. ±/�? (Pedersen 1900a: 322), H-grade in Gr. �À, �À-J ‘again’ , Latin DXW ‘or’ . (14) Alb. HQM, sM < *+RLG�Q- ‘swell’ , Arm. D\WQXP, Gr. #¨ z7, #¨ 9�7 (cf. 22 below). (15) Alb. HGK < *DL - ‘kid’ , Arm. D\F, Gr. �©! (cf. above). (16) Alb. DUt < *+ � UW R- ‘bear’ , Arm. DU , Gr. x'�-#+, Skt. N D . (17) Alb. DUs < *+ � U+ � - ‘field’ , Arm. KDUDZXQNµ < *+ � HU+ � -, DUDZU < *+ � U+ � - ‘plough’ , Gr. x'#/'�, x'#-'#�, Latin DUYXP, DU WUXP. (18) Alb. HOE < *+ � OEKL- ‘barley’ , Gr. x�Q�. (19) Alb. sQGH < *+ � QGK- ‘flower’ , Gr. x�K#+, Arm. DQG ‘field’ (cf. above). (20) Alb. sQGsUU < *+ � QU- ‘dream’ , Gr. ²�J�'#+ < *+ � QHU�, Arm. DQXU < *+ � Q U-(cf. above). (21) Alb. HPsU < *+ � Q+ � PHQ- ‘name’ , Gr. ²�#µ�, Russ. tPMD, Arm. DQXQ < *+ � Q+ � PQ, Latin Q PHQ, Skt. Q PD. (22) Alb. sQM, sM < *+ � Q+ � - ‘blow’ , Gr. x�Jµ#+ < *+ � Q+ � HPRV ‘wind’ , ��Jµ?� < *Q�+ � Q+ � HP- ‘calm’ , Skt. iQLWL < *+ � HQ+ � WL ‘breathes’ (cf. 14 above). (23) Alb. PEL, PEs < *+ � PEKL ‘on’ , Arm. DPE�Rá ‘complete’ , Gr. wµQ? ‘about’ , Skt. DEKt, OE. \PE�H�. (24) Alb. Qs, QGs < *+ � QGK- ‘in’ , Gr. |�K�, Skt. iGKL. (25) Alb. LP, Arm. LP, Gr. yµ²+ ‘mine’ (cf. above). Albanian does not share the Graeco-Armenian vocalization of an initial laryngeal before a heterosyllabic consonant, e.g. QsQGs < *+ � QHXQ- ‘nine’ , PMHO < *+ � PHO - ‘milk’ , QMHUt < *+ � QHU- ‘man’ , OLJ <*+ � OLJ- ‘bad, ill’ , Gr. y��z� (Arm. LQQ), wµz��7, w�=' (Arm. D\U), ±�?�#+. 9. I conclude that Armenian and Albanian share not only a number of lexical items, which they have in common with Greek, but also a number of phonetic developments: the depalatalization of palatovelars before resonants, which is also found in Balto-Slavic, the monophonemicization of PIE. *VZ, * Z, * KZ, which may have been a parallel development of the two languages, and perhaps the twofold nonzero reflex of PIE. *+ � - and *+ � -, which were vocalized before a tautosyllabic resonant, as in Greek, and yielded aspiration before PIE. *H. The similarities between the two languages support the idea that they originated from contiguous Indo-European dialects, situated between Greek in the south and Balto-Slavic in the north (cf. also Hamp 1966).

[I now think that the intervocalic reflex of PIE. *V in Albanian is VK (cf. 1987c: 221).]

75

Notes on Armenian historical phonology V*

The fourth number of the Arbejdspapirer udsendt af Institut for Lingvistik, Københavns Universitet ($3,/.8 4, December 1984) contains three articles on the development of the Proto-Indo-European laryngeals in Armenian. Muller claims that interconsonantal laryngeals were lost before a single consonant but yielded D before a cluster (1984a: 97). I think that this rule is indeed correct. A similar rule holds for Indo-Iranian, as Beekes has demonstrated (1981a). The only genuine exception according to Muller is GDODU ‘green, fresh’ , Gr. K��J'²+ ‘blooming, fresh’ , which seems to point to *GK+ � O+ � URV or *GKDO+ � URV (1984b: l00). The Greek form must be derived from *GK+ � OHURV, however, and cannot therefore be identified with the Armenian word, which may represent the zero grade of the root *GKHO- ‘bright’ , cf. GHáLQ ‘yellow’ , GDOXNQ ‘jaundice’ . Thus, the comparison must probably be abandoned. Unlike Muller's rule for the development of the PIE. laryngeals in medial position, Olsen's thesis that word-initial anteconsonantal laryngeals yielded D- in Armenian (1984: 111, cf. Klingenschmitt 1982: 105, fn. 27) cannot be maintained. Perhaps the most remarkable thing about her article is the absence of any reference to Greppin's little monograph on the subject (1973). Even if there are good reasons to disagree with Greppin's views and methodology (cf. Kortlandt 1983b [this vol., 39ff.]), the simple disregard of his work is not only at variance with common standards of scholarly behavior, but also harmful to the discipline. We can only hope for progress in the field if earlier scholarship is taken seriously. In my view, the development of the PIE. initial laryngeals before a consonant in Armenian is identical with the development in Greek, *+ � / � / � - yielding a threefold reflex according to the original color of the laryngeal. Unlike the Greek reflexes, which were preserved unchanged up to historical


times, the Armenian vowels were subject to later developments, *H- yielding L- before a nasal and R- before a rounded vowel in the following syllable (beforethe syncope), and *R- yielding D- in open syllables (after the syncope, cf. Meillet


[62]

1894a: 154, Pedersen 1900c: 100, Kortlandt 1980b: 105 [this vol., 32]). As a result, the normal reflex of *+ � - and *+ � - is indeed D-, but *+ � - is reflected as L�, H�, R-, and the expected reflex of *+ � - in historically closed syllables is R-. Moreover, initial *+- before *Z was lost in Armenian (but not in Greek), cf. JRJ ‘say’ , JRP ‘am’ , Gr. J¾3#µ��, Hitt. KXLã]L. An initial laryngeal before *\ was lost both in Armenian and in Greek. If we confront Olsen's view and mine with the material, it will be clear that the two of us are in agreement on the reflex of *+ � -, e.g. in D\U ‘man’ , DVWá ‘star’ , DUHZ ‘sun’ , and on the reflex of *+ � - in open syllables, e.g. DQLFDQHP ‘curse’ , \�DZHOXP ‘increase’ , also aor. \�DUHD\ ‘rose’ , Gr. ²�J� #+, ±Qz��7, ±'#º)��. The vocalism of the present tense \�D QHP ‘rise’ , Gr. ²'�/µ��, is in my view analogical. Since anteconsonantal *+ � - merged with *+R-, it is hardly possible to ascertain the original vocalism of RáE ‘lament’ , R ‘rump’ , RUE ‘orphan’ , RUMLNµ ‘testicles’ , RVW ‘branch’ , RVNU ‘bone’ , XWµ ‘eight’ . None of these examples is mentioned by Olsen. If any of them represents the zero grade *+ � -, her thesis is disproved. I consider zero grade probable for RáE in view of Gr. ±�#Q '#µ�� and Lith. XOE~RWL, which point to *+ � OREK-, *+ � OEK-, and for XWµ (cf. Kortlandt 1983b: 14 [this vol., 44]), and possible for RUE, RUMLNµ, RVNU, Gr. ±'Q��²+, ²'3�+, ±)-z#�. Olsen does mention HáXQJQ ‘nail’ , Gr. ²�/!, which is irregular anyhow (if it is cognate at all). This word is best derived from an alternating paradigm *+ � QRJ K-, *+ � QJ K-, which yielded Arm. *RQRJ-, *RQJ- in accordance with my rule. The generalized stem form *RQRQJ- was apparently subject to a rounding dissimilation which yielded *HQRQJ-, and subsequently to nasal dissimilation and raising yielding HáXQJQ, where -Q represents the PIE. acc.sg. ending. The rounding dissimilation requires *R- as the reflex of *+ � - and thereby supports my view. Initial *+ � - is reflected as L- in LQQ ‘nine’ , LP ‘mine’ , LQHZ (from *LPH�EL) ‘with me’ , Gr. y��z�, yµ²+, yµz, PIE. *+ � QHXQ, *+ � PRV, *+ � PH. Olsen assumes a metathesis of *QHZ to *HQZ for both languages, a totally arbitrary hypothesis which yields the wrong forms both for Armenian (**LQJQ) and for Greek (**HQD, **HLQD). She does not mention the pronoun. Initial *+ � - is reflected as H- in HOX]DQHP ‘extract’ , HUHN� ‘evening’ , perhaps HUJ ‘song’ , Gr. y�JE)#µ��, |'J�#+, Hitt. DUNXZDL-. Olsen proposes a “ possible explanation of the disturbing initial H-” of HOX]DQHP on the basis of an original augmented aorist * OXGKH- yielding *LOX]H- and subsequently *HOX]H-, referring to the dialectal form OL]X beside classical OH]X ‘tongue’ in support of the “ phonetic principle” surmised for the latter development (p. 109). There can be little doubt that OL]X is an adaptation of OH]X to the verb OL]HP ‘lick’ (cf. Pedersen 1906: 419 [= Kl.S., 197]). Thus, her explanation is entirely DG�KRF. She proposes that HUHN represents *UHJ

- rather than *+ � UHJ - (similarly Klingenschmitt, l.c.)

and asserts that the word “ seems to stress the necessity of reconstructing a [63]

1RWHV�RQ�$UPHQLDQ�KLVWRULFDO�SKRQRORJ\�9 77

system with an initial *U-” (p. 112). The reconstructed absence of initial *U- from Proto-Indo-European is not based on its absence in Greek or Armenian, as Klingenschmitt and Olsen suggest, but on the absence of unextended PIE. roots with an initial *U-, which Lehmann demonstrated a long time ago (1951: 17). It must be regretted that this point has escaped many later investigators. Initial *+ � - is reflected as R- in RUFDP ‘vomit’ , Gr. y'JE�#µ��, and possibly in RURþHP ‘ruminate’ and RURQHP ‘search’ . Initial *H- was rounded before U plus rounded vowel in Armenian, cf. RUR � ‘lamb’ , HULQ ‘heifer’ , Gr. |'�Q#+. Olsen rejects both *+ � UHXJ- and *UHXJ- “ as pre-stages of Arm. RUFDP, which leaves us only one possibility of explaining the initial R-: it must be a preverb, as assumed by Beekes (1969, p. 88)” (p. 110), cf. Klingenschmitt (l.c., with the same reference). This is hardly a correct rendering of Beekes' statement that “ it may be asked whether we are in fact concerned here with a phonetic development of the laryngeal. (We might envisage a preverb)” . The connection of RURþHP with Latin U G , U G , Vedic UiGDWL (Greppin 1973: 28) is difficult because the latter words represent PIE. *+UH+ � G- (cf. Lubotsky 1981: 134). Olsen's derivation of RURþHP from *N URZ\H- is arbitrary. The connection of RURQHP with Gr. y'J/�97 is phonetically difficult, but the word can be identical with Homeric y'JJ?�7, with RURQ- from *+ � UHXHQ-. Alternatively, both RURþHP and RURQHP may contain the prefix *SUR-. Olsen adduces three instances in support of her view that *+ � - yielded D- in Armenian, viz. DQXQ ‘name’ , DWDPQ ‘tooth’ and DZHWHP ‘bring good news’ . Since the comparison of the latter word with Gr. �½ = imposes *+ � -, she posits a stem *+ � VX�+ � ZHG�H\H- and declares that “ the problem is solved” (p. 108). This method of reconstruction can hardly be taken seriously. Both DQXQ and DWDPQ actually represent *+ � -, as is clear from Gr. ²�#µ�, ²�/µ�, ± I�, ± #E+, and especially �I�/µ#+, �7 ²+, which are archaic formations, cf. w�I�/µ#+, w�² 7�, and also �¬µ7 z7. The Laconian name �/µ��'�-? �+ cannot be taken as evidence for an original *+ � - in the word for ‘name’ . Apart from the Greek and Armenian material, the Phrygian word #�#µ�� also points to *+ � -. The Aeolic form | #�-J+ can easily be explained as secondary on the basis of the verb | 7 (cf. Beekes 1969: 55). The root *+ � HG- ‘bite, sting’ is also attested in Lith. ~RGDV, Latv. X{GV ‘gnat’ (with lengthening in accordance with Winter's law) and in Gr. ± E�� ‘pain’ , cf. �I /�#+. The latter word cannot be cognate with either Arm. HUNQ ‘pang’ , which can be related to Gothic IDXUKWMDQ (cf. Pedersen 1906: 399 [= Kl.S., 177]), or Irish LGX ‘pain, pang’ , which is best compared with Gothic ILWDQ. Schindler's identification of the three words (1975) is precluded by formal and semantic

[64]


difficulties on the Armenian side (cf. Pedersen, l.c.) and by the incongruous vocalism of Irish and Greek. Thus, there is no evidence for Arm. D- from PIE. *+ � -. [For Greek K��J'²+ ‘blooming’ I now reconstruct *GKO+ � HURV, which renders a comparison with the Armenian word possible (cf. Clackson 1994: 118-120, with references).]

79

Sigmatic or root aorist?* The large majority of Armenian aorists must be derived from thematic formations, e.g., 3rd sg. -Fµ < *�VNHW, HOLNµ ‘left’ , HJLW ‘found’ , HEHU ‘bore’ , HKDUFµ ‘asked’ , DUDU ‘made’ , Gr. |��%J, |QJ'J, �'�'J, Skt. iYLGDW, iEKDUDW, iS FFKDW. There are clear traces of athematic formations. Since intervocalic *-V- was lost, it is not always obvious whether these represent sigmatic or root aorists. Consider the following paradigms: 1st sg. HNL ‘came’ HGL ‘put’ HWX ‘gave’ 2nd sg. HNLU�� HGLU� � � HWXU 3rd sg. HNQ� � HG� � � HW 1st pl. HNDNµ� � HGDNµ� � � WXDNµ 2nd pl. HNLNµ� � HGLNµ� � � HWXNµ 3rd pl. HNLQ� � HGLQ� � � HWXQ The 3rd sg. forms are the phonetic reflexes of the root aorists *HJ HPW, *HGK W, *HG W, cf. Vedic iJDQ, iGK W, iG W. The derivation of these forms from a sigmatic formation requires the additional hypothesis that the *-V, which was not of PIE. date, was again eliminated by an analogical process. Though this is a possibility in the case of HG and HW, it is highly improbable in the case of HNQ, where the final nasal can hardly be due to restoration. When we turn to the other persons, the situation is slightly different. Since the thematic 1st sg. endings *- and *-RP merged into Proto-Armenian *-X, it is probable that this ending spread in pre-apocope times. Bonfante's derivation of HGL and HWX from *HGK VRP and *HG VRP (1942) is therefore most probably correct. The form HNL cannot be derived from a sigmatic formation. It has been argued that *P was lost before *X in Proto-Armenian (cf. Kortlandt 1981c: 30 [this vol., 35]). If this is correct, HNL is the phonetic reflex of


*HJ HPRP. Thus, we find that HGL and HWX represent sigmatic formations and HNL a root aorist. The 2nd sg. form is irrelevant because the personal ending *-V merged

[50]


[51]

with the formative *-V-, so that the two types of athematic aorist cannot be distinguished. The ending -U < *-UR was taken from the transitive middle paradigm (cf. Kortlandt 1981c: 32 [this vol., 36f.]). The absence of *P in the 1st pl. ending -DNµ can be explained if we start from Proto-Armenian *-DPX (cf. Kortlandt, ibidem). The forms HNDNµ and HGDNµ therefore seem to reflect the zero grade root aorist forms *HJ PH, *HG � PH, cf. Gr. |KJµJ�. The form WXDNµ must be the result of an analogical development. No conclusions can be based on the 2nd and 3rd pl. forms. Apart from the verbs which have been mentioned thus far, we can suspect an earlier athematic aorist in the following paradigms: 1st sg. Há ‘became’ NHUD\ ‘ate’ OXD\ ‘heard’ 2nd sg. HáHU�� NHUDU� � � OXDU 3rd sg. HáHZ� � HNHU� � � OXDZ 1st pl. HáHDNµ� � NHUDNµ� � � OXDNµ 2nd pl. Há Nµ� � NHUD\Nµ� � � OXD\Nµ� 3rd pl. HáHQ� � NHUDQ� � � OXDQ The 3rd sg. form HáHZ probably replaces earlier *Há, which can be identified with Vedic i UHW < *H OHLW and Albanian TH, NOH ‘was’ (which must be a sigmatic formation, cf. Pedersen 1900b: 341 [= Kl.S., 2]). The velar á, which arose phonetically in word-final position, apparently spread to the other forms of the aorist paradigm, but did not reach the imperative OHU. The generalization of the stem HáH- disambiguated the paradigm from that of HO- ‘ascended’ . The 1st sg. and pl. forms Há , HáHDNµ may represent *HNOHLVRP, *HNOHLVPH. The 3rd sg. form HNHU reflects the root aorist *HJ HU � W, cf. Vedic subj. JiUDW. The vocalization of the laryngeal explains the D-stem of the paradigm. In the case of OXD\ I think that there is evidence for a sigmatic aorist in the present tense OVHP ‘hear’ . The assumption of a N-present of this root (Meillet 1936: 133, Klingenschmitt 1982: 157f.) is arbitrary, More probably, OVHP represents a sigmatic stem with a secondary nasal infix *NOXQV-, just as PSHP ‘drink’ reflects the reduplicated stem with an analogical nasal infix *SLPE-, Skt. StEDWL. Nasal presents derived from sigmatic aorist stems are not rare in Armenian, as Holger Pedersen argued a long time ago (1905a: 206 [= Kl.S., 68], 1906: 423ff. [= Kl.S., 201ff.]), e.g., DQLFDQHP ‘curse’ (DQ F, Gr. ²�J� #+), [DFDQHP ‘bite’ (H[DF, Skt. NK GDWL), KHFDQLP ‘mount’ (with prefixed K-, Gr. {�#µ��), PXFDQHP ‘introduce’ (HPR\F, cf. PWDQHP ‘enter’ , HPXW), HOX]DQHP ‘extract’ (Gr. y�JE)#µ��), VX]DQHP ‘plunge’ (HVR\], Gr. �JEK7), KHáMDQHP ‘suffocate’ (HKHáM, cf. [HáG), DQFµDQHP ‘pass’ ( DQFµ, Gr. x�-#µ��), OXFµDQHP ‘light’ (HOR\Fµ, cf. OR\V), OXFDQHP ‘untie’ (Gothic JD-, XV�O NDQ), QNOX]DQHP ‘submerge’ ( QNOR\], cf. QNOQXP), NRUXVDQHP ‘lose’ (NRUR\V, cf. NRUQþµLP).

6LJPDWLF�RU�URRW�DRULVW 81

Godel’s derivation of PXFDQHP, [DFDQHP, OXFµDQHP from a stem in *-\H- (1975: 82, followed by Klingenschmitt 1982, who consistently disregards Pedersen’s views) is highly improbable in view of P ‘middle’ < *-GK\-, JRþµHP ‘-call’ < *-N \-. Thus, it appears that sigmatic aorists were rather widespread at an earlier stage in the development of Armenian. The prehistoric spread of the sigmatic aorist in Armenian supports the derivation of the Fµ-aorist from a sigmatic formation rather than from the present stem. Godel's derivation of 3rd sg. QN Fµ ‘threw’ from *VRQJ H\H�VNHW (1969: 256, 1975: 128) cannot be upheld because *-H\H- yields -H-, not - -, cf. HUHNµ ‘three’ < *WUH\HV. Moreover, the Ionic iterative preterit in -HVNH cannot be directly compared because it is a recent formation (cf. most recently Ruijgh 1985: 146f.). It seems more attractive to derive QN Fµ from a sigmatic form in *-HLV�W� to which *-VNH�W��was added. Before a vocalic ending, the sequence *-\V- was lost, so that QNHFµ- is the expected stem form. Thus, I think that the vowel alternation betrays the earlier existence of a sigmatic aorist.1 In the case of denominatives, the 3rd sg. ending -HDFµ points to an earlier aorist stem in *-HVD- or *-LVD- which may have arisen along similar lines as Gr. yQ?��)� of Q��z7. In the same way, the middle aorists in -HD\ are most easily derived from sigmatic extensions of -stems, to be compared with the Greek aorist in -�� (cf. Godel 1975: 121). The original stative meaning of this class precluded the rise of a Fµ-form. As in Greek, the D-flexion of the sigmatic aorist probably originated in the endings which began in a nasal. I think that this was indeed the starting-point for the generalization of the D-flexion in the Armenian mediopassive aorist. There is no evidence for a suffix *- - in this category. Summarizing, we have found traces of an original root aorist in HNL, HNQ, HNDNµ, HG, HGDNµ, HW, HNHU, and evidence for a sigmatic formation in HGL, HWX, Há , OXD\, -HD\, Fµ-aorists, and nasal presents of stems in F, ], M, Fµ, V of roots in dentals and palatals. The flexion of HGL and HWX points to the coexistence of sigmatic and asigmatic forms in the same paradigm, a situation reminiscent of Slavic. Most probably, the forms *HGK P and *HG P were replaced by the reflexes of *HGK VRP and *HG VRP when the final nasal was lost, and a similar

1. There is little in favor of Godel’s etymology; it seems more likely that Q� is a prefix. I wonder if the stem can be identified with Gr. �9��7, aor. |��#�. Pedersen assumes that *-OV- yielded -áV- in VWHáFDQHP ‘create’ , 3rd sg. aor. HVWHáF, )-z��7, also NHáFNµ ‘feint’ < *GZHOV-, Latin EHOOXP (1906: 427, 379 [= Kl.S., 205, 157]), so that QNHFµ� may

represent earlier QNHOF�. It is conceivable that - Fµ is the phonetic reflex of *-HOV�W�. For the loss of -á- before an aspirate cf. NDWµQ ‘milk’ < *-ONW-, Gr. �9��, -�--. The absence of

-O- in the present QNHQXP ‘throw’ differentiates the verb from QNOQXP ‘plunge’ . The root-final laryngeal of PIE. *J � HO+ - was lost before a vowel. I am unable to see a motivation for the general elimination of this laryngeal in Armenian.

[52]


substitution may have taken place in the plural, e.g. the reflex of *HG VDPH replacing that of *HG � PH. The paradigms were eventually normalized shortly before the beginnings of the written tradition. As I have pointed out earlier (1981c: 31 [this vol., 35f.]), the historical personal endings to some extent reflect the stem formation of the athematic aorist, which in part represents a sigmatic formation. [I now think that �Fµ� is the expected reflex of the PIE. sigmatic aorist marker *�V� in Armenian (see Kortlandt 1995a [this vol., 107ff.]). On the development of the verbal system see further Kortlandt 1996b [this vol., 110ff.] and 1999 [this vol., 129f.].]

83

The Thraco-Armenian consonant shift* Vladimir Georgiev has strongly argued in favor of a strict distinction between Dacian and Thracian, the former language being identified as the ancestor of Albanian (e.g., 1977). Neither language shows a particular affinity with Phrygian, which now appears to be rather closely related to Greek (cf. Lubotsky 1988). One of the major characteristics distinguishing Thracian from Dacian is a consonant shift, as a result of which PIE. *W, *G, *G �

are reflected as W �

, W, G in Thracian, e.g. 8WXV ‘water’ , Dacian -XGH, Gr. Â 7'. Since we find a similar consonant shift in Armenian, we may wonder if the development has a common origin in the two languages. It has often been suggested that there is a special relationship between Armenian on the one hand and the ancient Balkan languages on the other (cf. especially Djahukian 1970, and the lucid survey in B lt ceanu 1980).1 This relationship can only be specified if the chronological relations can be determined. Such an evaluation requires the establishment of a relative chronology with respect to the internal development of the separate languages. I shall not go into the severe limitations which the scanty material imposes on our knowledge of the ancient Balkan languages because this aspect has received wide attention in the scholarly literature. In the following I intend to examine the compatibility of the evidence for a close relationship between Armenian and the Balkan languages with the internal chronology of the Armenian developments. Djahukian lists the following major characteristics of Armenian, Thracian, and Phrygian historical phonology (1970: 163):

* Reprinted from %DONDQVNR�(]LNR]QDQLH / /LQJXLVWLTXH�%DONDQLTXH 31 (1988), 71-74. 1. I am indebted to Professor R.S.P. Beekes and to Dr. J.J.S. Weitenberg for supplying some of the relevant literature.

Arm. Thr. Phr.

consonant shift yes yes yes satem language yes yes no *V yields K yes no no


*-P yields -Q yes yes yes D-coloring of syll. res. yes no yes * yields R yes no yes close * (further to X) yes yes yes *HX yields RL yes no no preservation of *+ yes no no prothetic vowel yes no no This is a simplified picture of the actual state of affairs. The alleged Phrygian consonant shift cannot be maintained (cf. Lejeune 1979). The preservation of PIE. *V in Phrygian holds for word-initial position, but intervocalic *V was lost in this language (cf. Lubotsky 1988). The preservation of PIE. *R in Armenian does not hold for non-final open syllables, where it mostly developed into D. Before a consonant, an initial laryngeal is reflected as a prothetic vowel in Phrygian, as it is in Greek, whereas a consonantal reflex before a vowel would probably remain unmarked in the orthography. There is no evidence for a non-zero reflex of an initial laryngeal in Thracian, but there is no counter-evidence in the scanty material either. Thus, we may conclude that Phrygian largely agrees with Greek (except for the preservation of initial V- in Phrygian and the devoicing of the aspirates in Greek), while Armenian appears to be equally close to Thracian and Phrygian on the basis of the above list of correspondences. The problem we now have to face is the relative chronology of the developments which led to the attested state of affairs. Elsewhere I have proposed the following relative chronology of Armenian sound changes (1980b [this vol., 26ff.]): 1. Loss of aspirated stops. This was a dialectal Indo-European development which Armenian shared with Celtic, Germanic, Balto-Slavic, Albanian, and Iranian. The resulting consonants remained distinct from the earlier unaspirated voiced stops, which were injectives at this stage. 2. Rise of new labialized stops. Elsewhere I have argued that this may have been a shared development of Albanian and Armenian (1980a: 249 [this vol., 18]). It may also have been common to Greek and other Balkan languages. 3. Assimilation of * to * . This may also have been a common development of Albanian and Armenian. 4. Assimilation in *VZH XU ‘mother-in-law’ , which became * ZH XU at this stage and subsequently developed into VNHVXU. 5. *V became *K unless it was preceded by a consonant or followed by an obstruent. 6. Development of * (from PIE. * ) into V. 7. Redistribution of labialization.

[72]

7KH�7KUDFR�$UPHQLDQ�FRQVRQDQW�VKLIW 85

8. Palatalization. 9. *W

and *G became *N

and *J .

l0. Lenition: *S, *W, *N, *N became fricatives.

11. Rise of new velar fricatives. 12. Simplification of consonantal articulation. 13. Loss of intervocalic and anteconsonantal fricatives. 14. Loss of the syllabic resonants, which received an epenthetic D. 15. Metathesis. 16. Apocope. 17. Simplification of consonantal articulation. 18. Loss of labialization. 19. Shortening of fricatives. 20. Rise of new [. 21. Influx of Iranian loanwords. 22. Syncope. When we compare the above list of correspondences with this relative chronology, it turns out that Armenian was much closer to Thracian than to Phrygian at the earliest stages of its development: (1) Unlike Greek, Thracian and Phrygian may have shared the loss of the voiced aspirates with Balto-Slavic, Daco-Albanian, Armenian, and Iranian. Unlike Phrygian, Thracian may have shared the devoicing of the glottalic (unaspirated voiced) stops with Armenian. This devoicing can be dated between the assibilation of * to * (stage 3) and the palatalization of velars (stage 8), as I pointed out earlier (1980b: 100 [this vol., 28]). Thracian did not share the Armenian lenition of voiceless stops (stage 10). (2) Unlike Phrygian, Thracian is a satem language like Armenian and Daco-Albanian. (3) Thracian did not share the development of PIE. *V to K which Armenian has in common with Greek and Iranian. This development is not very early either in Armenian (stage 5) or in Iranian, where it was posterior to the separation from Indic. It is reasonable to assume that *V was relatively weak in the southern dialects of the Indo-European proto-language. (4) The development of final *-P to -Q which all the Balkan languages have in common with Armenian was probably anterior to the lenition (stage 10, cf. Kortlandt 1980b: 101 [this vol., 28f.]) and may have been very early. (5) The elimination of the syllabic resonants was recent in Armenian (stage 14) and in Greek, where we find a twofold reflex: R-coloring of syllabic U and O in Arcado-Cyprian and Aeolic, and D-coloring elsewhere. The epenthetic X in Thracian does not therefore conflict with the hypothesis that any two of these languages were closely related at an early stage. (6) The development of PIE. *R to D in Armenian can be dated after the syncope

[73]


[74]

(stage 22, cf. Kortlandt 1980b: 105 [this vol., 31f.]). Similarly, the same development in Thracian may have been recent. (7) The narrowing of * was anterior to the lenition (stage 10, cf. Kortlandt 1980b: 101 [this vol., 29]) and may have been very early. It distinguishes Thraco-Armenian from Daco-Albanian. (8) The Armenian development of *HX to R\ can be dated after the lenition (stage 10). It is of no relevance to the earliest relationships. (9) Since there is no evidence for the development of initial laryngeals in Thracian, this point must be left out of consideration. Thus, it appears that Thracian shared the earlier developments of Armenian (loss of the voiced aspirates, devoicing of the glottalic stops, satem assibilation, preservation of the syllabic resonants, narrowing of * ), while the discrepancies between the two languages seem to be the result of later developments in Armenian (lenition of voiceless stops, development of *V to K, epenthetic D, development of *HX to R\) and Thracian (epenthetic X, development of *R to D). We may therefore think of Thracian as an early dialect of Proto-Armenian. This view leads to a reconsideration of the following points. Firstly, the devoicing of the glottalic stops can be dated to an early stage. Since Thracian did not share the development of PIE. *V to K (stage 5) whereas Daco-Albanian did not share the devoicing of the glottalic stops though it probably shared the chronology of the assibilation of the palatovelars (stage 3, cf. Kortlandt 1980a: 250 [this vol., 19]), it is reasonable to date the devoicing to stage 4, which may then be called Thraco-Armenian. Secondly, the development of clusters consisting of obstruent plus *Z deserves more attention because these are reflected in particular ways in the Balkan languages as well as Armenian. The development of PIE. * Z in Thr. (VEHQXV (Skt. i YD ‘horse’ ) is reminiscent of the development in Arm. VNHVXU ‘mother-in-law’ , where original *VZ- was evidently assimilated to * Z- as in Skt. Yi XUD and Lith. ã ãXUDV ‘father-in-law’ , whereas the loss of the resonant in Thr. 6DEDGLRV (OCS. VYRERG ‘free’ ), = UXQWKRQ (OCS. ]Y U ‘beast’ ) suggests a comparison with Arm. ã ‘donkey’ , MD\Q� ‘voice’ (Pol. G ZL N ‘sound’ ). It seems reasonable to assume that Thracian shared the early rise of geminated labialized stops which subsequently developed along different lines in the separate languages, cf. also Gr. %%#+, etc. [The Thracian name 7DUX�WKLQQDV, -WKLQDV, -VLQDV (Gr. ²'/ ‘spear’ , )?�#µ�� ‘harm’ ) is of particular interest here because it suggests that we may have to do with the development of aspirated *F

from an earlier geminated *W .]

Thirdly, the close relationship between Thracian and Armenian should lead us to expect non-zero reflexes of PIE. initial laryngeals in Thracian. Though initial K-, which is found in Armenian and Albanian (cf. Kortlandt 1986a: 43f. [this vol., 73f.]), would probably remain unmarked in the orthography, prothetic

7KH�7KUDFR�$UPHQLDQ�FRQVRQDQW�VKLIW 87

vowels should be demonstrable. I see no compelling evidence for the view that Thracian sides either with Daco-Albanian, or with Greek, Phrygian, and Armenian in this respect. Thus, I find no evidence against the hypothesis that there was a single Thraco-Armenian consonant shift which separated this branch from the other Balkan languages. This hypothesis is compatible with the chronological relationships which can be established. It supports the view that the Proto-Indo-European glottalic stops were preserved at the stage of development under consideration. [I now think that aspiration had never arisen in Indo-European times but was a local development of the separate branches (cf. Kortlandt 2003).]

88

The making of a puzzle* In an earlier issue of this journal, Theo Vennemann has proposed an explanation in terms of syllable structure for the alleged sound change of *GZ to UN in Armenian, e.g. in HUNX ‘two’ , a development which he dubs “ the Great Armenian Puzzle” (1986: 29). In a footnote he refers to the present author as someone who “ disagrees strongly both with the theory and method of my approach and with a number of etymological assumptions adopted here from other sources” (1986: 55). It may therefore be expedient that I explain the nature of the disagreement to the readers of this journal. I. Theory. According to Vennemann's theory, NµR ‘yours’ and HUNX ‘two’ must be derived from *WZR- and *GZ by the following rules: (1) Armenian consonant shift: Vennemann assumes this development “ to have preceded all other sound changes between Proto-Indo-European and Classical Armenian” (1986: 28). He therefore writes “ Pre-Armenian” *WµZR-, *WZ . (2) Glide strengthening: the change of *Z to J yielded *WµJR-, *WJ -. (3) Voicelessness assimilation: *WµNR-, *WN . (4) “ Velar aspiration” : *N yielded Nµ “ in a syllable onset from which an aspirate is lost” (1986: 34). This development yielded *NµR- from *WµNR-, but left *WN unchanged. (5) Prothesis: *HWN from *WN . (6) Coda weakening: intervocalic *WN yielded UN. This chain of developments differs from the one proposed by the present author (1980b [this vol., 26ff.]) in the following respects: (1) The main problem in connection with the date of the Armenian consonant shift is the preservation of a distinction between the voiced reflexes of


PIE. voiceless stops after resonants, e.g. PDUG ‘man’ , KLQJ ‘five’ , Gr. �'#-²+, %z�-J, and the voiceless reflexes of PIE. “ voiced” (glottalic) stops in the same position, e.g. VLUW ‘heart’ , JRUF ‘work’ , Gr. ��' ?�, |'�#�. Vennemann does not

[44]

7KH�PDNLQJ�RI�D�SX]]OH 89

address this problem at all. The voicing rule did not affect KDUVQ ‘bride’ , Latin SURFXV, where the PIE. palatovelar had evidently been assibilated at an early stage. I am inclined to date the devoicing of the glottalic stops after the loss of palatal articulation in DUW ‘field’ , Gr. w�'²+ (1980b: 100 [this vol., 28]). Most previous authors seem to agree on an early date for the Armenian consonant shift. The notable exception is Meillet, who writes about the proposed development of *GZ- in HUNX ‘two’ that “ U est un reste de l'articulation sonore G: l'altération du groupe GZ est donc antérieure à la mutation consonantique arménienne” (1936: 51). The incongruity of this hypothesis was already noted by Grammont, who suggested that U continues the following *Z and that the resulting cluster *NU- “ a subi l'interversion habituelle en -UN-” (1918: 252). It seems better to leave the Armenian consonant shift out of consideration altogether in the discussion of the development of *GZ, as Vennemann in effect does. (2) Elsewhere I have argued that the chronology of the change of *Z to J in relation to the loss of final syllables, the palatalization before front vowels, and the rise of intervocalic Z from labial stops points to two chronologically distinct developments: “ It is probable that the rise of buccal features in this sound and the transformation of the resonant into a resonantal feature date from the earliest stages of the Armenian language, whereas the transformation of the sound into an occlusive and the loss of the labialization feature are relatively recent. This state of affairs is perfectly compatible with the hypothesis that Georgian �YLQR was borrowed from Armenian” (1976: 95). Vennemann does not discuss these problems at all. In my view, the change of *Z to J is partly analogous to the change of final *-V to -Nµ (1980b: 101ff. [this vol., 28ff.]). Vennemann does not even mention this point. (3) The development of *WZ- via *WµJ- and *WµN- to Nµ- to which Vennemann subscribes (1986: 55) was stated explicitly by Zabrocki (1951: 157), whom Vennemann does not mention, and by Winter (1962: 262). It is based on Meillet's statement that after a consonant “ le Z devient aussi guttural; certaines consonnes précédentes perdent leur point d'articulation propre, mais toutes conservent leur caractère de sourde ou de sonore, d'aspirée ou de non aspirée qui est attribué à la gutturale” (1936: 50). It is interesting to note that Meillet was apparently not quite clear in his own mind about the chronology of the developments because he continues: “ ainsi *VZ-, devenu *KZ-, donne, avec assourdissement du Z par K, arm. Nµ” , and “ * Z donne avec le traitement normal de * et assourdissement de Z, arm. VN” (ibidem). I subscribe to the idea that *Z was devoiced by a preceding voiceless consonant before the rise of velar articulation (1980b: 99-101 [this vol., 27-29]) and therefore reject the chain of developments advocated by Zabrocki and Vennemann. (4) Vennemann does not tell us why the dental stop was lost in *WµNR-

[45]


but not in *WN . He compares the rise of Nµ- in the former word with the development in NµR\U ‘sister’ from *VZHV U, but does not tell us why the cluster was preserved in VNHVXU ‘mother-in-law’ , V[DOLP ‘fail’ , RVNU ‘bone’ , which can be compared with Skt. Yi XUD , VNKDODWL, Gr. ±)-z#�. He ignores the rise of palatal ã in ãXQ ‘dog’ , ãXQþµ ‘breath’ , ã ‘donkey’ , Skt. Y , YiVLWL, i YD , and the loss of the labial (or velar) element in MD\Q ‘voice’ , Russ. ]YRQ. He does not “ offer a coherent description” (1986: 28) but presents a series of rather DG�KRF rules to account for a limited material in a far from cogent way. His “ velar aspiration” is no improvment when compared with Meillet's formulation quoted in the preceding paragraph. (5) Vennemann does not distinguish between “ prothetic” vowels of laryngeal origin, which he attributes to “ sporadic prothesis” (fn. 15), and real prothetic vowels, which arose in Armenian at a much more recent stage. I have dated the rise of a prothetic vowel before initial U after the loss of PIE. initial voiceless stops before U because these would otherwise be reflected as Z (1980b: 105 [this vol., 32]), e.g. HUHNµ ‘three’ from *WUH\HV, cf. KDZU ‘father's’ from *SDWURV. Vennemann ignores this point. (6) The alleged development of *W to U before N differs from the development of PIE. *W to zero in HUHNµ ‘three’ , to Z in KDZU ‘father's’ , and to \ in EHU ‘carries’ , all of which are the result of a weakening process. Vennemann only remarks that “ the weakest consonantal member of the dental order is the central liquid” (1986: 54) and does not discuss the discrepancy at all. II. Method. The major point of disagreement between Vennemann and the present author appears to concern the role of the material in a linguistic analysis. The point may be exemplified by a quotation from the opening paragraph of Vennemann's article: “ My goal is not to provide new evidence for or against these sound changes but rather to describe them in a uniform phonological framework and thereby to derive explanations for them from certain preference laws for phonological structure” (1986: 27). The linguistic facts are taken for granted and merely serve as an illustration of a general theoretical framework. If the framework provides a generalization which fits the selected material, this is considered to be an explanation of the facts. This primacy of theoretical considerations contrasts sharply with the empiricist outlook of those who regard counter-evidence as the driving force behind all scholarly progress. In my view, a new insight means a new way of looking for counter-evidence rather than a new way of looking at known facts. The difference appears several times in Vennemann's footnotes. He writes e.g. that “ Kortlandt posits voiceless, glottalic, and aspirated plosives for Proto-Indo-European” and that “ his chronology is motivated in part by his own assumptions about both the Proto-Indo-European

[46]


speech sounds and their development in Armenian, assumptions which I do not share” (fn. 5). Similarly: “ In my opinion the correct assumption is positional frication of +7µ, with no need arising for the positing of late parts of an Armenian Consonant Shift” (ibidem). The terms ‘positing’ and ‘assumption’ suggest a freedom of choice to which I do not subscribe. Much better is his formulation that “ the glottalic plosives are identified as preglottalized voiced” , which suggests that there is actual evidence for this reconstruction (as I think there is). The primacy of the theoretical interest is particularly clear in the following statement: “ However this may be, the fact that matters to my argument is that +\ was eliminated word-initially, together with +Z and +U; whether it was strengthened, or lost or dissimilated, makes no difference for my purpose” (fn. 16). Referring to the hypothesis that *GZ yielded UN, Vennemann writes: “ Those repeating Meillet's ‘supposition’ have not been convinced by Pedersen's contrary argument (the salvaging part of which is that the portion HU� of HUNX ‘two’ was prefixed to the reflex +NX of the Proto-Indo-European etymon on the model of HUHNµ ‘three’ ). However, even for those convinced that Meillet's equation is correct the change itself has so far remained problematical from a phonological point of view” (fn. 22). The real point is that there is counter-evidence for the supposed development, a fact which Vennemann does not even mention. It is possible that the counter-evidence can be accounted for if it is properly examined. The amazing fact, however, is that it is simply disregarded by those who, like Vennemann, believe that the hypothesis is correct. It is remarkable that those who are so eager to accept Meillet's bold suggestion do not pay attention to the careful attitude toward the material which guided him in his research in the first place. III. Etymology. In his earliest discussion of HUNX (1894a: 160), Meillet stuck to Bugge's derivation of �NX from *GZ : “ Cela suppose que HU� est une addition postérieure à la chute des finales” because -X would otherwise have been lost. “ Il n'existe pas d'autre exemple du traitement de GZ- initial en arménien; mais les cas analogues justifient l'hypothèse arm. N = i.e. GZ” (ibidem). In a later issue of the same volume he already reports a second example: “ LJ ‘joint, uni’ et comme deuxième terme de composé, ‘compagnon’ , cf. v.h.-a. ]ZLVN; HUL V ‘trois fois’ , cf. v.h.-a. GULVN” (1894b: 296). For the semantics of NLFµ cf. Dutch JHWZHHsQ ‘two together’ . In the same volume Meillet derives HUNQþµLO ‘fear’ from HUNX, adducing parallels from other languages (1894c: 255): “ Le fait qu'il s'agit d'un doute ayant un intérêt particulier pour le sujet est indiqué par la voix dans skr. EKi\DWH, gr. J Ü?)�#µ��, arm. HUNQ LO et par la forme réfléchie de v.sl. ERM �V ,

[47]


lit. ELMD V.” 1 The semantic relationship has more recently be/en stressed by Benveniste (1954: 255): “ *GZHL- ‘craindre’ signifie ‘être en double, GRXWHU’ au sens où GRXWHU est pris en ancien français (= fr. mod. UHGRXWHU).” In the next volume of the same journal Meillet compares HUND\Q, HUNDU ‘long’ with Lith. F EWBT ‘spacious’ (1896: 150). In a later volume he adds PHáN ‘soft’ from *PHOGZL- (l900: 594). This is the place where we find the earliest formulation of what Vennemann calls “ Meillet's famous hypothesis” : “ HUNX représente phonétiquement *GZ , c'est-à-dire que i.-e. *GZ- initial est représenté en arménien par HUN-; et en effet on a HUNLZá ‘crainte’ , HUNQ LP ‘je crains’ ” , which represent “ i.-e. *GZL- ‘craindre’ ” . In the second edition of the (VTXLVVH he adds HUNDU ‘long’ from *GZ UR- to these examples (1936: 51). Holger Pedersen did not accept Meillet's hypothesis. Since his objections have not received due attention, it may be appropriate to quote the relevant passage in full (1906: 398f. [= Kl.S., 176f.]):

M e i l l e t Esquisse s. 29 will nachweisen, dass anlautendes GY- im Armenischen als HUN- auftritt (HUNX ‘zwei’ ); U soll ‘un reste de l'articulation sonore G’ sein; unglücklicherweise spricht dagegen nicht nur NUNLQ ‘doppelt’ , sondern auch die übrigen ableitungen des zahlwortes: NXá ‘verdoppelung’ , NL ‘vereinigt, genosse’ (M e i l l e t MSL VIII 296), N V ‘hälfte’ , NRLV ‘seite’ (§ 38). HUNQ LP ‘fürchte’ , HUNLXá ‘furcht’ wird keineswegs zu gr. z#+ gehören; schon die verwandten wörter HUN ‘mühe, beschwerde, kummer’ und HUNQ gen. HUNDQ ‘geburtsschmerzen, furcht’ widerlegen diese kombination vollkommen; lautlich unanstössig ist dagegen eine kombination mit got. IDXUKWMDQ (über frühere etymologien dieses wortes vgl. Z u p i t z a Gutturale s. 5); idg. wurzel *SHUJ- oder *SHUJ

-. Dass die Meilletsche vermuthung über GY- sich für eine populäre „esquisse“ eignet, kann ich nicht finden.

Pedersen derives N V ‘half’ and NR\V ‘side’ from *GZRL R- and *GZRX , respectively (1906: 400 [= Kl.S., 178]). We may add NHáDNDUF ‘doubtful’ from *GZHO- (Pedersen 1906: 379 [= Kl.S., 157]). In his answer to Pedersen, Meillet mentions the following possibilities to account for the counter-examples NUNLQ, NXá, NLFµ, N V, NR\V (1909: 353): 1. Professor C.J. Ruijgh points out to me that the form * J Ü?)�#µ�� should be asterisked. The attested form y J ?)�J-# (Aristophanes) is often corrected to y J ?--J-#, but cf. Hesych. J ?)�J-# ‘frightened’ . The transitive meaning of the form J� ?))#µ��, J ?))#µ�� is already attested in Homer.

[48]


(1) Twofold reflex: “ étant donné qu'il s'agit de l'initiale du mot, on pourrait à la rigueur admettre qu'on est en présence d'un fait de phonétique syntactique; comme les faits de ce genre n'existent guère en arménien, l'explication paraîtrait sans doute un peu désespérée.” (2) False etymology: “ rien n'oblige à croire que des mots signifiant ‘pli’ , ‘compagnon’ , ‘moitié’ , ‘côté’ soient tirés du nom de nombre ‘deux’ .” (3) Special development: “ si NUNLQ n'a pas de U avant le N, c'est sans doute par dissimilation” . Meillet rejects Pedersen's comparison of HUNQþµLP with Gothic IDXUKWMDQ, pointing out that “ le S initial n'aurait laissé aucune trace, ce qui, au moins devant un H, serait surprenant” (1909: 354). This certainly is a valid, though not decisive, objection. The expected K- was regularly lost in R-grade forms and after prefixes. Meillet also rejects the connection of HUNQþµLP with HUN and HUNQ: “ Il n'y a aucune raison pour grouper HUN ‘travail pénible’ , HUNQ ‘douleur intense’ , HUNXQNµ ‘douleurs de l'enfantement’ , avec le mot signifiant ‘craindre’ ; l'idée de ‘craindre’ et celle de ‘souffrir cruellement’ ne sont pas si proches que les deux mots doivent appartenir à un même groupe” (ibidem). Grammont accepts Meillet's hypothesis that *GZ yielded UN but rejects his idea that the U represents a trace of the *G. He assumes that *GZ� became *WZ� as a result of the consonant shift and then *NZ� “ par assimilation des points d'articulation, de même qu'en allemand WZ- est devenu NZ�, par exemple dans all. TXDUN ‘fromage mou’ de m.h.a. WZDUF, all. TXHQJHOQ ‘geindre’ en face de m.h.a. WZHQJHQ, all. TXHU ‘en travers’ de m.h.a. WZsU, all. TXHWVFKH ‘pruneau’ en face de ]ZHWVFKH, all. TXLUO ‘moussoir’ de m.h.a. WZLUO” (1918: 251f.). The group *NZ� then became *NU�� which yielded HUN�. According to Pisani (1934: 185), *GZ became *J � which changed to *JU under the influence of *WUH� ‘three’ and then developed into HUNX. He derives NUNLQ ‘double’ from *NUXNLQ, where *NUX- represents *GZ , and concludes that the metathesis of the initial consonant group was limited to words which were monosyllabic at the right moment, such as HáED\U ‘brother’ and HUND\Q ‘long’ . Though this appears to be an DG�KRF solution, Szemerényi calls it “ obviously right” (1960: 96). Pisani compares HUND\Q with Latin JUDQGLV ‘big’ (1934: 184), which may indeed be correct. In a later study he derives HUNDU ‘long’ from the same rout (1944: 164). Pedersen had already identified -DU in HUNDU as a suffix of collective origin (1906: 479 [= Kl.S., 257]). In search of additional evidence for the change of *GZ to UN, Dumézil proposes to derive VHUNHDQ ‘today's’ from *VH�GLZ- “ avec chute précoce de L en syllabe non finale” (1938: 51). This is at variance with the development of *GLZ- to WX- in the oblique cases of WLZ ‘day’ , cf. also DQLZ ‘wheel’ , gen. DQXR\ (Kortlandt 1976: 98f. [this vol., 7f.]), so that Frisk's judgment that the proposal “ ist lautlich nicht einwandfrei” (1966: 261, fn. 3) is an understatement. It is also

[49]


unsatisfactory from a morphological point of view, as Dumézil recognizes (1938: 51): “ Il ne faut pas regarder de trop près le H de VH-: l'allemand KHXWH suffit à rappeler que les composantes de ce genre de mots ont été parfois ‘métamorphisées’ de façon imprévisible; VH- peut cependant représenter un ancien cas” , which he identifies forty years later (1979: 108): “ Il faut sans doute partir du locatif, dont la désinence était anciennement H dans certains pronoms” . This H represents *-RL (cf. Kortlandt 1984a: 101f. [this vol., 48f.]) and is incompatible with the demonstrative stem *VL-, which is found in DZDVLN ‘voici’ (cf. Pedersen 1905b: 37 [= Kl.S., 41], Kortlandt 1983a: 316 [this vol., 52]). Dumézil's proposal is also unacceptable for chronological reasons: “ U ne pouvant provenir que d'une sonore, *GZ est passé à UN avant la mutation qui a assourdi G, E, etc. en W, S, etc.” (1979: 107), which puts the consonant shift after the syncope and therefore after the influx of Iranian loan words. The adverb VHUN ‘today’ remains unexplained. In his later study Dumézil even proceeds to deriving HUNLQ ‘heaven’ and HUNLU ‘earth’ from *GLZLQ- ‘celui du ciel’ and *GLZLUD- ‘ayant le Ciel pour homme’ . These words are better compared with Irish JUtDQ ‘sun’ and Welsh HUZ ‘field’ , respectively (cf. Pedersen 1905a: 197 [= Kl.S., 59], Solta 1960: 151). In 1944 Frisk proposed to derive HUNQ ‘pain’ from *HGZ Q, identifying the word with Gr. ± E�� (1966: 259), a comparison which Schindler has extended to Irish LGX (1975). Both connect the etymon with the PIE. root *+ � HG- ‘eat’ . The main problem with this etymology is the initial R- of the Greek word. There is a variant H- in the phrase -#¹+ �q' ± ²�-�+ | #�-�+ �z�#/)�, ��¤ -q+ ± E��+ y E��+ (Greg.Cor., 12th c. AD, cf. Schindler 1975: 55). This can hardly be regarded as evidence for an original H-. According to Frisk “ ist mit einem alten qualitativen Ablautswechsel zu rechnen” (1966: 260), whereas Schindler assumes “ ganz einfach Vokalassimilation” (1975: 56). These conjectures are unsatisfactory because they do not explain the distribution of the variants. The archaic character of R- in Gr. ± E�� ‘pain’ as well as ± I�, ± #E+ ‘tooth’ is clear from the negative compounds �I /�#+ ‘painless’ , �7 ²+ ‘toothless’ , later w�I /�#+, w�² 7�, also �¬µ7 ?� ‘sensation of having the teeth set on edge’ , and the agreement with Arm. DWDPQ ‘tooth’ , with D- from R- in an open syllable. The Aeolic form | #�-J+ ‘teeth’ can easily be explained as secondary on the basis of the verb | 7 ‘eat’ . I have therefore proposed to start from a PIE. root *+ � HG- ‘bite, sting’ , which is also found in Lith. ~RGDV ‘gnat’ (1987a: 63 [this vol., 77]). It follows that HUNQ must be separated from Gr. ± E��. Neither of these words can be identified with Irish LGX. Schindler remarks: “ Theoretisch besteht jedoch noch die Möglichkeit, dass LGX gar kein alter Q-Stamm, sondern ein Q -Stamm war” (1975: 60). He rejects this possibility: “ Diese Annahme erübrigt sich jedoch, weil bei LGX erstens jeder positive synchrone Hinweis auf einen Q -Stamm fehlt (die Nominative LGQD Cóir

[50]


[51]

Anmann 121, LGKQD SG 1.19.1 sind natürlich Akkusative in nominativischer Verwendung), und weil zweitens fürs Idg. nur der Q-Stamm gesichert ist” (ibidem). But these nominatives are precisely the evidence we have for an original Q -stem, and the etymological connection remains to be substantiated. Schindler continues: “ Bisher wurde LGX mit got. ILWD ‘Ë ?�7’ verbunden (...). Lautlich und semantisch (...) ist gegen den Vergleich nichts einzuwenden” (ibidem). The word HUNQ�cannot therefore be adduced as evidence for the alleged development of *GZ into UN. It can hardly be separated from the R-stem HUN ‘toil’ . After this review of the material and the arguments presented in the scholarly literature, I conclude that *GZ yielded N, which is or may be attested in HUNX, NUNLQ, NXá, NLFµ, N V, NR\V, NHáDNDUF, PHáN, and perhaps HUND\Q, HUNDU, if these are related to Lith. F EWBT rather than Latin JUDQGLV. The numeral HUNX ‘two’ adopted HU- from HUHNµ ‘three’ in the same way as Eng. IRXU has I- from ILYH and Russ. GHYMDW ‘nine’ has G- from GHVMDW ‘ten’ . The form NU- in NUNLQ ‘double’ is found with prefixed HU- in HUNLU ‘second’ , also HUNURUG, while NLFµ ‘conjoined, contiguous, adjoining, together’ is found with prefixed HU- in HUNLFµV ‘twice’ , OHG. ]ZLVN ‘double’ (Meillet 1936: 101). The verb HUNQþµLP may either be a recent derivative of HUNX, or a cognate of OHG. IXUKWHQ ‘fear’ with analogical elimination of the initial *K�, or a word of unknown origin. Thus, the “ Great Armenian Puzzle” turns out to be a red herring. [Meillet’ s early comparison of Arm. HUND\Q, HUNDU ‘long’ with Lith. F EWBT ‘spacious’ is definitely better than Pisani’ s comparison of these words with Latin JUDQGLV, which is probably a borrowing from an unknown source. Harkness proposes a development of *GZ > *WJ > *NU > *UN (1996), dismissing the counter-evidence and disregarding the arguments adduced above. De Lamberterie’ s recent treatment of “ Meillet’ s law” (1998) is very useful but has not convinced me of its correctness for the reasons stated above.]

96

[2]

Arm. FDQDZWµ ‘known’ * The PIE. sequence *&5+& appears to have a twofold reflex in Armenian, cf. þDQDþµHP ‘I know’ , FDQDZWµ ‘known’ , DáDþµHP� ‘I implore’ , DáDZWµNµ ‘prayer’ (e.g., Winter 1965: 103f., Klingenschmitt 1982: 167, fn. 13). Scholars are in disagreement on the origin of these forms. The obvious source for þDQDþµHP is * Q+ � VNH�R-, which should yield *FDQDFµHP. Meillet assumes “ élargissement d'un présent en *-VNH- par le suffixe *-\H-” (1936: 109) and “ assimilation du F�initial d'un ancien *FDQDþµHP” (p. 29). This is doubtless correct. The addition of the present formative *-\H- was motivated by the spread of *-VNH- as an aorist and subjunctive marker. Though Pedersen originally derived the present formative þµ from palatalized *VN (1905a: 207 [= Kl.S., 69]), he later changed his mind (thus explicitly 1906: 422 [= Kl.S., 200]) in favor of a derivation from single *N, which he identifies with the velar in Gr. |K��J and Latin IDFL (1906: 348 [= Kl.S., 126]). But the latter is not a present formative. Moreover, the expected reflex of the Greek and Latin suffix should rather be Arm. V, certainly to those who do not accept three unconditioned series of velar obstruents for the proto-language. Meillet's theory is certainly preferable both formally and semantically. Winter (1965: 104) and Godel (1975: 80) derive þDQDþµHP and FDQDZWµ from *-N\- and *-NW-, respectively. While Godel relegates the origin of this *N to “ the gloomy history of PA” (1975: 124), Winter derives it from the root-final laryngeal. The problem with this derivation is that it separates þDQDþµHP and DáDþµHP from the other þµ-presents, which are left out of consideration, and introduces a zero-grade \H-present where we would expect either a zero-grade VNH-present, which we find in other languages, or a full-grade \H-present on the basis of the root aorist. Zero-grade \H-presents have stative or passive meaning, e.g. Gr. µ�?�J-�� ‘he is mad’ , Skt. E~GK\DWH ‘he awakes’ , and reflect the type


which underlies the Arm. passive L-presents (cf. Meillet 1936: 107). While Schmitt adopts Meillet's derivation of the þµ-presents from *-VN-\H- (1981: 136), Klingenschmitt follows Pedersen's early derivation of -þµH-

$UP� canawt‘ µNQRZ¶ 97

from *-VNH- (1982: 68, 84), though he derives KDUFµDQH- ‘ask’ and KD\FµH- ‘request’ from *SU �VNH- and *+DLV�VNH-, respectively, with dissimilation of *-VN- on an DG� KRF basis. This is clearly an arbitrary way of dealing with the facts. Assuming that þDQDþµHP represents * Q+ � VN�\H-, we conclude that the expected reflex of PIE. *&5+& is &D5D& in Armenian. For FDQDZWµ we must start from the participle * Q+ � WRV and the verbal noun * Q+ � WLV, which should yield *FDQDZ and *FDQD\, respectively, as is clear from the 3rd sg. middle ending -DZ < *-DWR and the word ED\ ‘verb’ < *EDWL- < *EK+ WLV. Since the characteristic *-W- was lost in these formations, it is reasonable to expect an analogical replacement or restoration of the suffix. Pedersen has observed that “ das suffix *-WL- nur da produktiv geblieben ist, wo es mit anderen suffixalen elementen zu einer umfangreicheren endung verwachsen war [e.g., JRYHVW ‘praise’ , LPDVW ‘understanding’ , HUHZR\Wµ ‘appearance’ , DáDZWµNµ ‘prayer’ ]. Die erweiterung des urspr. suffixes scheint oft dadurch zustande gekommen zu sein, dass man die endung -WL- in ganz unursprünglicher weise direkt an den konsonantischen auslaut eines suffixes gefügt hat [e.g., HUDãW ‘dryness’ < *-VN-, FQXQG ‘childbirth’ < *-Q-, WµDNµXVW ‘concealment’ < *�X�V�]´ (1905a: 218f. [= Kl.S., 80f.]). The motivation for this development is of course the loss of *-W- after a vowel. Thus, I think that the dental suffix was restored in *FDQDZ- < * Q+ � WRV after stage 13 of my chronology (1980b: 102 [this vol., 30]) before the rise of new intervocalic obstruents. The generalization of -DZWµ rather than -D\Wµ may be due to the influence of DZWµ ‘passing the night’ and the suffix -R\Wµ, as in HUHZR\Wµ ‘appearance’ , a formation which Bugge already compared with Gr. -J/)�+ (1893: 80) and which became productive in later Armenian. This suggests that the introduction of -DZWµ in FDQDZWµ and DáDZWµNµ was anterior to the development of *HX into R\. The theory proposed here accounts for FDQDZWµ on the basis of a single phonetic reflex of PIE. *&5+& and without invoking an unmotivated introduction of a suffix of unknown origin. [On the present tense suffix �þµ� see further Kortlandt 1996b [this vol., 110ff.].]

[3]

98

Proto-Armenian numerals* Ten years ago I discussed the formation of the Greek numerals against the background of (my variety of) the so-called glottalic theory (1983c). In the present contribution I intend to examine the development of the PIE. numerals in Armenian on the basis of the proposed reconstructions.

1. PL ‘one’ and the initial syllable of PHWDVDQ ‘eleven’ seem to reflect *VPLD, Gr. µ?� (cf. Winter 1992: 348), but see below. The masculine stem form *VHP� was eliminated because it yielded an awkward paradigm: nom. *KLV (cf. DPLV ‘month’ ), acc. *KLZQ (cf. MLZQ ‘snow’ , PU LZQ ‘ant’ , Kortlandt 1985c: 21 [this vol., 64f.]), loc. *KLP (homophonous with KLP ‘why’ ), cf. KLQ ‘old’ from *VHQ�. The alternant form PLQ ‘one’ may reflect the accusative of PL (cf. Kortlandt l.c.). It is probable that the oblique forms PLR\ and PLXP on the one hand and PLR on the other represent the original masc. and fem. sets of pronominal endings, respectively (cf. Kortlandt 1984a [this vol., 45ff.], Weitenberg 1984). The ordinal D D LQ ‘first’ is derived from D D ‘before’ , which is in its turn derived from D ‘at’ , Gr. %'², cf. YHU LQ ‘last’ , YHU ‘end’ , L� YHU ‘above’ , Gr. Á%z'.

2. HUNX ‘two’ represents �NX from *GZ with transfer of initial HU� from HUHNµ ‘three’ (cf. Kortlandt 1989 [this vol., 88ff.], Winter l.c.). Uninflected *GZR is preserved in HUNRWDVDQ ‘twelve’ and HUNRNµLQ, HUNRNµHDQ ‘both’ , and the composition form *GZL� in HUNHDP ‘two years old’ , also in HUNLFµV ‘twice’ , HUNLU, HUNURUG ‘second’ , but not in HUNHULZU ‘two hundred’ , which probably represents *GZRL (Winter l.c.). I think that NUNLQ ‘double’ from *NLULNLQ replaced *NLQ ‘double’ after the rise of *HULNLQ ‘triple’ , which was replaced by HUHNµNLQ after the syncope.1

* Reprinted from ,Q� KRQRUHP� +ROJHU� 3HGHUVHQ�� .ROORTXLXP� GHU� ,QGRJHUPDQLVFKHQ�*HVHOOVFKDIW [Kopenhagen, 1993] (Wiesbaden: Reichert, 1994), 253-257. 1. Dr. Alexander Lubotsky draws my attention to Vedic trisyllabic WUHGK beside WUtGK ‘in three ways’ from WUD\� plus �LGK , cf. WUi\DV ‘three’ and GYtGK ‘in two ways’ , which in its turn has a later analogical variant GYHGK . This is a nice parallel.

3URWR�$UPHQLDQ�QXPHUDOV 99

[254]

[255]

3. HUHNµ ‘three’ , acc. HULV reflect *WUHLHV, *WULQV. The initial part of HUHNµWDVDQ ‘thirteen’ is analogical. The composition form *WUL� is found in HUHDP‘three years old’ , also in HULFµV ‘three times’ , HULU, HUURUG ‘third’ .

4. þµRUNµ ‘four’ and the initial part of þµRUHNµWDVDQ ‘fourteen’ are usually compared with Doric -z-#'JO. The medial cluster *�WZ� may have been lost phonetically as a result of the monophonemicization (Kortlandt 1980b [this vol., 27ff.], stage 2), velarization (stage 9), lenition (stage 10), delabialization (stage 12c), and loss of intervocalic fricatives (stage 13a). The expected reflex of the composition form *N � WZU� is NµDU�, but we find NµD �, reflecting *�UV�, which was probably taken from NµD DPHD\ ‘quadrennial’ , cf. DP ‘year’ , Vedic ViP , analogically H DPHD\ ‘triennial’ . The ordinal NµD RUG ‘fourth’ replaces *NµD\URUG from *N � WZU�NRUWR�, cf. Lith. NDUWDV ‘time’ ; for the phonetic development cf. JD\O ‘wolf’ from *ZON � RV, Lith. YLONDV. The final part �URUG spread to the other numerals as an ordinal suffix. I agree with Szemerényi (1960: 95) that the short ordinal suffix �U spread from the original form *N � WXUR�, which was evidently replaced by the precursor of NµD RUG when the initial *N �

- was lost. This chain of developments explains the rise of the new ordinal suffix.21

5. KLQJ ‘five’ and the initial part of KQJHWDVDQ ‘fifteen’ represent *SHQN � H with restoration of the final velar on the basis of the ordinal *SQN � R�.

6. YHFµ ‘six’ and the initial syllable of YHãWDVDQ� ‘sixteen’ represent *VZH V with transfer of *X from the ordinal *VX VR-. The substitution was evidently triggered by the particular development of *VZ (stages 5, 7, 11 of Kortlandt 1980b).32

7. HZWµQ ‘seven’ reflects *VHSWP. The variant H WµQ from *HDZWµQ, for which Winter reconstructs a dialectal form *DZWµQ (1992: 350), rather suggests a reduced grade vowel which replaced zero grade vocalism in the ordinal and was later introduced into the cardinal.

8. XWµ ‘eight’ is parallel with Elean ±%-I. The absence of initial 2. Winter derives the short ordinal suffix �U from *�V after *L, *X, *U, e.g. HULU ‘third’ from *WULV ‘three times’ (1992: 355). Apart from the syntactic and semantic complications, this is improbable for phonetic reasons. The nom.acc.sg. ending �U in the X-stems is characteristic of original neuters, e.g. FXQU ‘knee’ , PHáU ‘honey’ , DVU ‘fleece’ , DUWDZVU ‘tear’ , and must therefore be derived from *�U, not *�V. The original nom.sg. ending *�V after *L is preserved in deverbal L-stems, e.g. HONµ ‘outcome’ , [DZVNµ ‘speech’ (Kortlandt 1985c: 23 [this vol., 67]). The 2nd sg. ending �U is characteristic of the middle aorist imperative and must therefore be derived from *�U�, not *�V. 3. The substitution of *KXH� for *KZH� in the cardinal ‘6’ on the basis of *KX� in the ordinal is parallel to the substitution of *GXR� for *GZR� ‘2’ on the basis of *GX� in other Indo-European languages, e.g. Latin GXSOXV, GXSOH[ ‘double’ , GXFHQW ‘200’ , GXELXV.


[256]

aspiration shows that it adopted the zero grade vocalism of the ordinal (cf. Kortlandt 1983b: 14 [this vol., 44]).

9. LQQ ‘nine’ similarly adopted the zero grade vocalism of the ordinal *HQXQR�, with regular vocalization of the initial laryngeal. The variant form LQXQNµ is the phonetic reflex of *HQXQHV.

10. WDVQ ‘ten’ represents *GH PW with adoption of the reduced grade vowel which replaced zero grade vocalism in the ordinal *G PWR-. The composition form �WDVDQ in the numerals from ‘eleven’ to ‘sixteen’ apparently reflects the ordinal.

20. NµVDQ ‘twenty’ reflects *GZLG PWL (Kortlandt 1983b: 14 [this vol., 43f.], 1983c: 100). Partial dissimilation of the initial glottalic consonant yielded a glottal stop which devoiced the following *Z in the same way as *K� from *V�, e.g. in NµR\U ‘sister’ from *VZHV U. The initial Nµ� is the phonetic reflex of unvoiced *Z (Kortlandt 1980b: 100 [this vol., 28]).

30. HUHVXQ ‘thirty’ is usually identified with Gr. -'� �#�-�, which reflects *WULD plus *G RPWD (Kortlandt 1983c: 100). Alternatively, it may represent the composition form *WUL� plus �DVXQ from the higher decades. But the form may actually be the phonetic reflex of *WUL+-G RPW-, see below.

40. NµD DVXQ ‘forty’ represents the composition form NµD � plus *G RPW-. It is attractive to regard �DVXQ as the phonetic reflex of *G RPW- after a syllabic resonant.

50. \LVXQ ‘fifty’ probably reflects *SHQN �-G RPW-, with irregular loss of

the medial vowel and \L� from pretonic *KLQ� (cf. Kortlandt 1983b: 11 [this vol., 41]). The loss of the medial vowel has a perfect analogue in Latin TX QJHQW ‘five hundred’ .43

60. YDWµVXQ ‘sixty’ has a reduced grade vowel in the initial syllable, which probably reflects the earlier composition form *YDFµ� beside the cardinal YHFµ and the ordinal *XFµ�, cf. NµD � beside þµRUNµ and *WµXU�.

70. HZWµDQDVXQ ‘seventy’ reflects *VHSWP plus �DVXQ from *G RPW- after a syllabic resonant. 80. XWµVXQ ‘eighty’ represents XWµ plus �VXQ.

4. The syncope in Latin TX QJHQW cannot be recent because the rise of �J� requires a preceding nasal at an early stage, cf. especially VHVFHQW ‘600’ versus VHSWLQJHQW ‘700’ , analogically RFWLQJHQW ‘800’ , similarly TXDGULQJHQW ‘400’ after TX QJHQW .

3URWR�$UPHQLDQ�QXPHUDOV 101

90. LQQVXQ ‘ninety’ represents LQQ plus �VXQ.

100. KDULZU ‘hundred’ has now been identified as a borrowing from Iranian (Bailey 1987).

Thus, I think that a characteristic feature of the Proto-Armenian system of numerals is the existence of three different ablaut grades in the root. The full grade is found in HUHNµ ‘3’ , þµRUNµ ‘4’ , KLQJ ‘5’ , YHFµ ‘6’ , HZWµQ ‘7’ , also in HUNX ‘2’ , HUNRWDVDQ ‘12’ , HUNHULZU�‘200’ . The zero grade is attested in HUNLU ‘2nd’ , HULU ‘3rd’ , XWµ ‘8’ , LQQ ‘9’ , NµVDQ ‘20’ . The reduced grade, which originated from a resyllabification of the zero grade on the pattern of the full grade, is found in NµD RUG ‘4th’ , WDVQ ‘10’ , NµD DVXQ ‘40’ , YDWµVXQ ‘60’ , and perhaps in *DZWµQ beside HZWµQ ‘7’ . The full grade was only preserved in the lower cardinals and largely replaced by the vocalism of the ordinals in the higher numerals. The influence of the original ordinals is indeed pervasive: it is reflected not only in the suffix of HUNLU ‘2nd’ and HULU ‘3rd’ from *WµXU� ‘4th’ , but also in the final consonant of KLQJ ‘5’ , the initial consonant of YHFµ ‘6’ , the initial vowel of XWµ ‘8’ , the vocalism of LQQ ‘9’ , and the rise of the reduced grade in WDVQ ‘10’ , NµD DVXQ ‘40’ , YDWµVXQ ‘60’ , and perhaps *DZWµQ ‘7’ . The heterogeneity of these reflexes strongly supports the theory presented here. The forms PHWDVDQ ‘11’ and HUHVXQ ‘30’ seem to suggest that Armenian shared the Greek vocalization of the final laryngeal in µ?� and -'?�, whereas the form PLQ ‘1’ points to loss of the laryngeal before the final nasal. The gen.sg. ending �L of the D�stems, e.g. DP ‘year’ , gen. DPL from *�L+RV (Kortlandt 1984a: 100 [this vol., 47]), rather suggests that the nom.sg. forms in *�L+ and *�D+ merged phonetically as a result of the apocope. It is therefore more probable that PHWDVDQ and HUHVXQ reflect an epenthetic vowel which developed between the final laryngeal of *VPL+ and *WUL+ and the initial cluster *G - of the second component. This epenthetic vowel is parallel to the one which developed after a syllabic resonant in NµD DVXQ ‘40’ and HZWµDQDVXQ ‘70’ . It appears that the initial glottalization of *G - merged with the reflex of the laryngeals both after a syllabic resonant and after the sequence *�L+�. While the reflex �D5D� remained unchanged, pretonic *�L+D� was contracted to �H� in the same way as *�LKD� from *� VD� in the aorist.

[On glottalization in Proto-Armenian see also Kortlandt 1998c [this vol., 126ff.]. On the initial consonant of KLQJ ‘five’ , \LVXQ ‘fifty’ , Autun KLRFW ‘seven’ , LVFKXQ ‘fifty’ see Kortlandt 1998a [this vol., 122ff.]. ]

102

[10]

Intervocalic *�Z� in Armenian* At the beginning of this century, Holger Pedersen wrote (1905a: 196 [= Kl.S., 58]): “ Man braucht nicht so weit auszuholen, um die doppelte gestalt des idg. Y im inlaut zu erklären. Das intervokalische Y erscheint als arm. Y, wo es auslautend geworden ist, sonst aber als J: NRY ‘kuh’ , aber NRJL ‘butter’ ; DUHY ‘sonne’ , aber DUHJDNQ ‘sonne’ . Die flexionsformen halten das Y des nominativs durch eine leicht verständliche analogiebildung fest: gen. NRYX, DUHYX.” This rule has often been challenged (e.g., Meillet 1936, Greppin 1972, Godel 1975, Eichner 1978, Schmitt 1981, Godel 1982, Olsen 1986), to my mind in an unjustified way (cf. Kortlandt 1980b). I shall briefly discuss the counterexamples. WµRYHP ‘enchant’ , Skt. VWiXWL ‘praises’ . Greppin points out correctly that the loss of the athematic conjugation was evidently later than the development of intervocalic *�Z� into a velar obstruent (stage 11 of Kortlandt 1980b: 101 [this vol., 29]). The further development of the latter into an occlusive �J� was much more recent (my stage 19). WHZHP ‘endure’ is evidently a derivative of WHZ ‘duration’ (Pedersen 1905a: 199 [= Kl.S., 61], Greppin 1972: 72). JRYHP ‘praise’ is similarly a derivative of JRY ‘praise’ (Pedersen l.c.). KRYLZ ‘shepherd’ “ ist als *RYL�S � ‘schafhüter’ gedeutet worden, wogegen sich jedoch Hübschmann ausspricht” (Pedersen 1905a: 198 [= Kl.S., 60]). The etymology was proposed by Bugge, who observed that the word “ bezeichnet zunächst ‘schafhirt’ (und ist dafür das gewöhnliche wort in der bibelübersetzung), dann ‘hirt’ im allgemeinen” (1893: 16). I see no objection to the view that *KRY ‘sheep’ was lost at a recent stage because of its homophony with KRY ‘cool(ness), fresh(ness)’ . Pedersen regarded KRYLZ as a derivative of the latter word. GHGHZHP, �LP ‘vacillate’ and KHZDP ‘pant’ are of little value (Eichner 1978: 155), while DZLZQ ‘fury’ and DZHWLNµ ‘good news’ (cf. Olsen 1986) are quite useless (see also Kortlandt 1987a [this vol., 75ff.]). No conclusions can be based on HáHZLQ ‘pine, fir, cedar’ , which Lidén compared with Russ. MDORYHF ‘juniper’ (1906: 491). The latter word is better derived from MDORY\M ‘barren’ (Vasmer 1958: 488). Medial *�Z� was lost before consonantal �U�, e.g. QRU ‘new’ , VRU ‘hole, cave’ , gen.sg. DáEHU, DOHU, QHUGL (with secondary �GL) of DáELZU ‘fountain’ , DOLZU ‘flour’ , QHDUG� ‘fibre, sinew’ (Pedersen 1906: 353 [= Kl.S., 131], Kortlandt


,QWHUYRFDOLF�*-w- LQ�$UPHQLDQ� 103

[11]

[12]

1980b: 102 [this vol., 29f.]).1 There is no reason to assume an intervocalic *�Z� in QRU and VRU, which evidently adopted the suffix *�UR� at an early stage. It is clear that *HZU did not merge with *HXU from *HV U in NµR\U ‘sister’ . The rise of the suffix in LZU ‘his, her’ may have been more recent than the apocope (Godel 1975: 111). There is no evidence for *�Z� in HUNDQ ‘millstone’ , Lith. JuUQD, or LQQ ‘nine’ , which represents the vocalism of the ordinal *HQXQR�. DUHJ is the original gen.sg. form of DUHZ ‘sun’ (Benveniste 1965: 9, Godel 1975: 83). It represents the phonetic reflex of *DUHZRV, with preservation of the velar obstruent before the reflex of final *�V after the apocope (cf. Kortlandt 1980b: 104 [this vol., 31]).2 \DJ ‘to satiety’ has been identified as \�DJ, with �DJ from *VDZ� (Meillet 1921). Vedic iVLQYDQ ‘insatiable’ actually points to a formation to be compared with Latin YHWXV ‘old’ from *XHW�XHV� beside Greek |-#+ ‘year’ from *XHW�HV� (cf. Beekes 1985: 61). I therefore think that we have to start from an V�stem *VH+ �HV� ‘satiety’ which is found with different hiatus fillers in Gr. � #+ and Arm. *VDZRV, cf. *�W�X� in Lith. VRW�V beside *�X�W� in Russ. V\W\M ‘sated’ , pointing to an original adjective in *�XHV� beside a formation in *�W�. The form *VDZRV yielded �DJ in the same way as *DUHZRV yielded DUHJ. VDJ ‘goose’ represents a metathesis of *J V from *JDQV� (cf. Pedersen 1906: 454 [= Kl.S., 232]), with depalatalization of the initial obstruent before a laryngeal (cf. Schrijver 1991: 113), and cannot be compared with Russ. VRYD ‘owl’ . No conclusions can be based on KRJ ‘care’ . Thus, I think that Benveniste's clever discovery (1965) and Meillet's ingenious comparison (1921) do not invalidate Pedersen's general rule. [See also Kortlandt 1996a [this vol., 117ff.]. ]

1. I am now inclined to identify development 17b with stage 13b of my chronology (1980b) because this offers a better structural picture. There is no reason to assume the rise of an epenthetic vowel before final �U at stage 14 and its subsequent loss at stage 16, especially in view of VLZQ ‘column’ < *VLZXQ < *VLZRQQ (Pedersen 1905a: 217 [= Kl.S., 79], Kortlandt 1985c: 21 [this vol., 64f.]), with loss of *Z before *X. We find delabialization before non-final �U� in QHDUG, as in OHDUG ‘liver’ . 2. I now think that the delabialization at stage 12c did not affect *Z before *R, perhaps because it had less friction than the other labialized consonants, and that the gen.sg. ending was always *�RV, cf. Kortlandt 1984a: 100 [this vol., 47]. The reflex of final *�V may have delabialized the reflex of *�Z� immediately after the apocope.

104

[28]

Palatalization of dentals in Armenian* In his contribution to the second international symposium on Armenian linguistics (1987), John Greppin lists six “ strong examples of *7\ that produce a palatal affricate” (1993: 16): J �‘moist’ < *JKHLGK�\�, P ‘middle’ < *PHGK�\�, D ‘right’ < *VH+ GK�\�, PDþDU ‘young wine’ < *PDG�\�, RURþHP ‘chew’ < *URG�\�, NRþµHP ‘call’ < *J � RW�\�. Though one may disagree with Greppin's reconstructions (cf. Kortlandt 1987a: 63 [this vol., 77] on RURþHP), these examples seem perfectly adequate to substantiate the regular outcome of palatalized dentals in Armenian. On the other hand, Greppin lists six “ strong examples of *7\ that produce a dental affricate” : JLQM ‘coriander’ < *ZHQGK�\�, [DFDQHP ‘bite’ < *NKDG�\�, DáF� ‘salty’ < *VDOG�\�, KHFDQLP ‘mount’ < *VHG�\�, PLF ‘mire’ < *PHLG�\�, DQLFDQHP ‘curse’ < *+ � QHLG�\�. It is clear that these examples are much weaker than those of the first list. As Greppin points out himself (1993: 18), “ Arm. JLQM, DáF� and PLF have loose semantic parameters” , which leaves us with the three nasal presents to which we may add PXFDQHP� ‘introduce’ < *PRXG�\H� (Godel 1975: 82). But nasal presents are never derived from \H�presents, so these reconstructions must be wrong. There simply is no good evidence for dental affricates from patatalized dentals in Armenian.1 * Reprinted from $QQXDO�RI�$UPHQLDQ�/LQJXLVWLFV 15 (1994), 27-31. 1. Birgit Olsen has tried, to my mind unsuccessfully, to find additional examples of dental affricates from palatalized dentals in Armenian (1988: 7f., 1993: 36ff.). This is not the place to discuss Olsen’s idiosyncratic theory of the subjunctive, which builds on an unmotivated and highly unnatural type of analogical development and does not explain the eventual distribution of the personal endings (on which cf. Kortlandt 1981c [this vol., 34ff.]). The classical doctrine developed by Meillet and Pedersen is certainly preferable. I

3DODWDOL]DWLRQ�RI�GHQWDOV�LQ�$UPHQLDQ 105

[29]

[30]

Nasal presents supply a present tense to an aorist stem. It follows that DQLF�, [DF�, KHF�, PXF� represent original aorist stems. Any theory which derives these aorists from \H�presents must explain why these \H�presents became aorists in prehistoric Armenian. It is much more probable that they reflect an original aorist formation. In fact, actual \H�presents usually have a derived Fµ�aorist in Armenian, e.g. SµD\OHP ‘I shine’ < *SKDO�\H�, SµD\OHFµL ‘I shone’ . Conversely, we even find \H�presents derived from earlier VNH�presents when the suffix �Fµ� became a subjunctive and aorist marker, e.g. þDQDþµHP ‘I know’ < * Q+ � �VN� plus �\H� (cf. Meillet 1936: 109). It follows that �\H� was a present marker SDU�H[FHOOHQFH. Some news travels slowly. As far as I am aware, the idea that the dental affricate of DQLF�, [DF�, KHF� was the result of a palatalization of dentals was first put forward more than a hundred years ago by Sophus Bugge (1893: 47). The correct derivation of �F� from *�GV� was first proposed shortly after the turn of the century by Holger Pedersen (1905a: 206 [= Kl.S., 68]), who added PXF�: “ HPXW ‘intravit’ und HPRLF ‘induxit’ verhalten sich wohl zu einander wie gr. |�� zu |��)� (präsens PWDQHP und PXFDQHP).” A year later Pedersen discussed the sigmatic formation in more detail (1906: 423-427 [= Kl.S., 201-205]). He added DáFHDO ‘salted’ and pointed out that beside �F� from *�GV� we have �]� (after liquids �M�, i.e. �G]�) from *�GKV� in HOX]DQHP ‘extract’ (Gr. y�JE)#µ��), VX]DQHP ‘plunge’ (Gr. �JEK7), KHáMDQHP� ‘suffocate’ (cf. KHáG), further �Fµ� from *�WV� in DQFµDQHP ‘pass, flow’ (Gr. x�-#µ��), �Fµ� from *- V- in OXFµDQHP ‘light’ (cf. OR\V), �F� from *� V� in OXFDQHP ‘untie’ (Gothic �OXNDQ), �]� from *� KV� in QNOX]DQHP ‘submerge’ , �V� from *�VV� in NRUXVDQHP ‘lose’ , �V� from *�QV� in VNVDQLP� ‘begin’ , �áF� from *�OV� in VWHáFDQHP ‘create’ (Gr. )-z��7). Any theory which purports to explain the stem-final affricate of DQLF�, [DF�, KHF�, PXF� must take this material into consideration. As Pedersen pointed out already (1906: 427-429 [= Kl.S., 205-207]), the reflexes �F�, �]� of *�GV�, *�GKV� and *� V�, *� KV� are not the phonetic outcome of original Indo-European clusters but represent “ in sämmtlichen fällen eine (allerdings in ziemlich alter zeit) erneuerte form” which arose in connection with the spread of the sigmatic aorist (on which cf. Kortlandt 1987b [this vol., 79ff.]). The crucial development was the loss of intervocalic *�V�� (Kortlandt 1980b: 99-101 [this vol., 27-29]). When *V became *K (my stage 5) and *VN became *F (my stage 6), the original clusters of root-final dental or palatal obstruent plus sigmatic aorist marker all merged with the Indo-European VN�suffix into a dental affricate.21 The stem-final affricate in the aorist was then disambiguated on the basis of the root-final obstruent which was found

hope to present my own views on the formation of the Armenian subjunctive elsewhere in the near future. 2. This is the origin of the Fµ�aorist, as I intend to argue elsewhere [this vol., 107ff.].


elsewhere in the paradigm, so that we end up with �Fµ�, �F�, �]� (�M�) reflecting *�WV�, *�GV�, *�GKV�. The correctness of this theory is nicely corroborated by the sigmatic aorist of roots in a labial obstruent (Pedersen 1906: 426 [= Kl.S., 204]). While original *�SV� is attested in HSµHP ‘I cook’ (Gr. {57), the labial consonant was lost in the sigmatic aorist FµDVHD\ ‘I lost my temper’ (cf. FµDZ ‘pain’ ), KLZVL ‘I wove’ (from *�XEK�, Gr. ÁQ�?�7). It appears that the sigmatic aorist marker was restored when *V had become *K and that the preceding labial obstruent was subsequently lost. I conclude that Pedersen's theory of the aorist stem formation remains unsurpassed and must be taken seriously in any treatment of Armenian historical phonology and morphology. [See further Kortlandt 1995a [this vol., 107ff.], 1996b [this vol., 110ff.], 1999 [this vol., 129ff.]. ]

107

[14]

The sigmatic forms of the Armenian verb*

“ In der Verwendungsweise besteht eine weitgehende Übereinstimmung der armenischen Aoriste auf �Fµ�H��, �Fµ�D�� mit den griechischen V-Aoristen” (Klingenschmitt 1982: 286). Since the two formations bear a stronger affinity than could possibly have been produced by accident, it is hard to examine them without believing that they have sprung from a common source. Moreover, if this identification is correct, “ ergibt sich für die armenischen Konjunktive WDFµH� und HNHFµH� die Möglichkeit einer Zurückführung auf mittels eines Suffixes *�VH�R� gebildete Aoristkonjunktive” (Klingenschmitt 1982: 43). The question now is: how did PIE. *�V� come to be represented by �Fµ� in Armenian? Klingenschmitt assumes that Arm. �Fµ� developed from *�VV� and compares Skt. DY WVDP ‘I dwelt’ of the root YDV�, with dissimilation of *�VV� to �WV� (1982: 287). This is too narrow a basis for the wide generalization of �Fµ� in Armenian, however. It may therefore be useful to have a look at what happened in other Indo-European languages before embarking on a discussion of the Armenian material. In Vedic, the aorist marker �V� was retracted to � � after L, X, U, N, including L from a vocalized laryngeal. The new suffix �L � < *�+V� then was a natural candidate for analogical spread after root-final consonants. Similarly, the future suffix �L \D� < *�+V\D� spread at the expense of original �V\D�. In Avestan, there is no Lã-aorist because the laryngeal was not vocalized. In Gathic, the aorist marker is �ã� after root-final L, X, U, [ (10 roots), �ã� or �V� replacing a root-final obstruent (8 roots), and �V��or �K� after a root-final nasal or vowel (7 roots); the attested number of different forms is 29 with �ã�, 9 with �V�, and 9 with �K� (cf. Beekes 1988a: 183-185). Thus, it appears that the conditioned variant �ã� would be a natural candidate for analogical spread.


In Old Church Slavic, the sigmatic aorist marker is �[��after L, X, U and replacing root-final velars and has spread analogically to the position after any vowel, while the original suffix �V� is still found replacing non-velar root-final


[15]

[16]

obstruents with 11 roots (cf. Leskien 1962: 141f.). The suffix �[� soon replaced �V� altogether in later Slavic. In Old Irish, things are more complicated. Elsewhere I have argued that the Old Irish V-subjunctive and D-subjunctive both represent the PIE. sigmatic aorist injunctive and correspond to the Vedic V-aorist and L -aorist, respectively (1984c). On the one hand, the suffixal �V� assimilated a preceding root-final obstruent, so that we have �V� < �VV� < *�7V�, *�.V� and loss of the root-final consonant. On the other hand, the �V� was lost in *�DV� < *�+V� before a vowel and the reflex of the vocalized laryngeal adopted the function of the lost �V�, resulting in an D-subjunctive. In the preterit, 3rd sg. *�V�W became �W after liquids and velars but �VV after the vocalized laryngeal �D�, thereby giving rise to a W-preterit and an DVV-preterit (cf. Thurneysen 1975: 417). What these developments have in common, is that the function of the suffixal �V� was transferred to the reflex of *�7V�, *�.V�, *�+V�, *�VW, *�+VW. In Greek, the sigmatic aorist “ ne pouvait pourtant se maintenir si l'analogie n'avait conservé le V caractéristique là où il subissait des altérations” (Chantraine 1967: 177). The suffixal �V� was preserved after obstruents, assimilating a preceding dental. When intervocalic �V��became �K� and �VV� was simplified to �V�, the geminate could be restored, e.g. y-z�J))� beside y-z�J)� ‘I fulfilled’ , similarly |)3�))� beside |)3�)� ‘I split’ of the root )3� -. Interestingly, both variants spread to the position after a vocalized laryngeal in Aeolic and Homeric Greek: ²�J()))� ‘I destroyed’ , 9µ�()))� ‘I tamed’ , �9�J()))� ‘I called’ . Thus, the sigmatic aorist marker represents *�7V�. In Doric and some other dialects, -!- spread at the expense of -)-, e.g. ��K?!�+ ‘seated’ , |%��!J ‘he played’ , with *�.V� for *�7V� (Chantraine 1967: 180). What is the expected sigmatic aorist marker in Armenian? As in Greek and Old Irish, intervocalic *�V� was lost in Armenian, so we are in need of a suitable model for an analogical replacement. If Armenian developed along similar lines as the other languages, we should consider *�7V�, *�.V�, and *�+V� as natural candidates for analogical spread. Holger Pedersen has pointed out “ dass in etymologisch isolirten wörtern velar + V immer þµ (þµRU ‘trocken’ , gr. !J'²+, !�'²+), palatal oder dental + V immer Fµ ergibt” (1906: 429 [= Kl.S., 207]), while *�+V� yielded �D� before a vowel. Since dentals and palatals were more frequent than velars, we expect �Fµ� for *�V� and �DFµ� for *�+V� if Armenian developed in the same way as Greek. In a previous study I argued that we have “ traces of an original root aorist in HNL, HNQ, HNDNµ, HG, HGDNµ, HW, HNHU, and evidence for a sigmatic formation in HGL, HWX, Há , OXD\, �HD\, Fµ-aorists, and nasal presents of stems in F, ], M, Fµ, V of roots in dentals and palatals” (1987b: 51 [this vol., 81]). The middle aorists in �HD\ are best derived from *� V� (cf. Godel 1975: 121), and the same holds for the subjunctive in �LFµ�, where the thematic flexion provoked an early

7KH�VLJPDWLF�IRUPV�RI�WKH�$UPHQLDQ�YHUE 109

introduction of the new consonant. The addition of the suffix �LFµ�� to present stems eventually yielded a present subjunctive and its addition to Fµ-aorist stems a new aorist subjunctive. The only active Fµ-aorists without a 3rd sg. form in ��H�DFµ appear to be HOLFµ ‘he filled’ << *�SO V�, H[LFµ ‘he stopped’ , which is unclear, and QN Fµ ‘he threw’ << *�J � HOV� (cf. Kortlandt 1987b: 52 [this vol., 81]). The only subjunctives without �L� preceding �Fµ� are WDFµ� ‘give’ << *WDV� < *G+ � �V� and HNHFµ� ‘come’ . The former represents a formation which is independent of the original root aorist and has a counterpart in Tocharian AB W V� ‘put’ < *GK+ � �V� (cf. Kortlandt 1994c: 64), while the latter, which has an analogical H-grade, probably arose from a reanalysis of the present stem which is attested in Gr. �9)�7 and Vedic JiFKD� ‘go’ . [See further Kortlandt 1996b [this vol., 110ff.] and 1999 [this vol., 129f.].]

110

[36]

The Proto-Armenian verbal system* The usual methodology of linguistic reconstruction is twofold. On the one hand, we may postulate features of the proto-language on the basis of what is attested in the daughter languages. This is the usual procedure in the case of older or major languages. On the other hand, we may look for traces of categories which have been reconstructed for the proto-language on the basis of other branches. This is the usual procedure in the case of more recently attested or minor languages, such as Armenian. Here we are dealing with a limited number of etymologies and a considerable gap between the proto-language and the earliest material. We may now wonder how these factors influence the reconstruction of the linguistic system as a whole. In the following I will argue that the number of etymologies is not very important and that the establishment of sound laws may even prevent scholars from arriving at a correct assessment of the reconstructed categories.

I�In order to examine the influence of the number of etymologies on the reconstruction of the verbal system I will reconsider the material assembled by Solta (1960). Solta divides the Armenian words into four classes, depending on the number of Indo-European language branches where they have correspondences: $ in most languages, % in many languages, & in a few languages, ' in a single language. I will add a fifth class ( containing words which Solta does not mention but which are adduced by Godel (1975). The number of verbal entries in the separate classes is as follows: $ 34, % 37, & 33, ' 12, ( 40. Note that I am not concerned with the correctness of the etymologies but only with the categories which they represent. A reconstruction of the verbal system on the basis of the class $

* Reprinted from 3URFHHGLQJV� RI� WKH� �WK� LQWHUQDWLRQDO� FRQIHUHQFH� RQ� $UPHQLDQ�OLQJXLVWLFV (Delmar, N.Y.: Caravan Books, 1996), 35-43.

etymologies now yields the following picture: � present in �HP�� LP, aorist in �HFµL�� HFµD\��DþHP ‘grow’ , JDQHP ‘beat’ , QHP ‘beat’ , Q HP ‘erase’ , JOHP ‘roll’ , GLHP ‘suck’ , OL]HP ‘lick’ , PDOHP ‘crush’ ,�

7KH�3URWR�$UPHQLDQ�YHUEDO�V\VWHP 111

[37]

POPOHP ‘rub’ , PL]HP ‘piss’ , P P LP ‘murmur’ , VLUHP ‘love’ , NµHUHP ‘scratch’ , NµRUHP ‘scratch’ , NµHUWµHP ‘strip’ . � present in �HP, aorist in �DFµL��JLWHP ‘know’ . � present in �DP, aorist in �DFµL��P P DP ‘murmur’ . � present in �DP, aorist in �FµL��NHDP ‘live’ . � present in �HP�� LP, aorist in �L�� D\�� DFHP ‘carry’ , EHUHP ‘carry’ , QVWLP ‘sit down’ . � present in �DP, aorist in �X��WDP ‘give’ . � present in �HP, suppletive aorist in �L��D\�� PSHP ‘drink’ , XWHP ‘eat’ , aor. DUEL��NHUD\. � present HP ‘am’ . � present in �VHP, aorist in �XD\��OVHP ‘hear’ . � present in �DþµHP, aorist in �HD\��þDQDþµHP ‘know’ , aor. FDQHD\. � present in �DQHP�� DQLP, aorist in �L�� D\��D RJDQHP ‘wet’ , R RJDQHP ‘wet’ , EXVDQLP ‘sprout’ , JWDQHP ‘find’ , GL]DQHP ‘amass’ , OL]DQHP ‘lick’ , ONµDQHP ‘leave’ , FQDQLP ‘give birth’ , KHFDQLP ‘mount’ . � present in �DQLP, aorist in � ��HáDQLP ‘become’ . � present in �DQDP, aorist in �DFµL��DFµD\��LPDQDP ‘understand’ , OXDQDP ‘wash’ , VWDQDP ‘acquire’ . � present in �QHP, aorist in �L��GQHP ‘put’ . � present in �QHP, aorist in �HD\��\D QHP ‘rise’ . � present in �QXP, aorist in �FµL��OQXP ‘fill’ . � present in �XP, aorist in �L��JHOXP ‘wring’ , OL]XP ‘lick’ , FµHOXP ‘split’ . It appears that most categories which were inherited from the proto-language are represented in this collection. We will see if the picture changes when we add the material of the class % etymologies: � present in �HP�� LP, aorist in �HFµL�� HFµD\�� D]D]HP ‘dry’ , GHGHZLP ‘waver’ , G QþµLP ‘ring’ , ODSµHP ‘lick’ , [DUHP ‘brand’ , [DUãHP ‘boil’ , FDPHP ‘chew’ , MXOHP ‘smelt’ , MDZQHP ‘dedicate’ , PDáWµHP ‘pray’ , VHUHP ‘procreate’ , VHULP ‘am born’ , WµH HP ‘rub off’ , FµWHP ‘scratch’ , FµWLP ‘scratch’ . � present in �DP, aorist in �DFµL�� DFµD\�� H DP ‘boil’ , ODP ‘weep’ , PQDP ‘remain’ , RUFDP ‘belch’ . � present in �DP��LP, suppletive aorist in �L��D\��JDP ‘come’ , XQLP ‘have’ , aor. HNL��NDOD\. � present in �DQHP��DQLP, aorist in �L��D\��DJDQLP ‘put on’ , DQLFDQHP ‘curse’ , DZFDQHP ‘anoint’ , HOX]DQHP ‘extract’ , HUJLFDQHP ‘tear’ , OXFDQHP ‘loosen’ , NODQHP ‘swallow’ , KDVDQHP ‘reach’ , PH DQLP ‘die’ , XVDQLP ‘learn’ , WHVDQHP ‘see’ , FµXFµDQHP ‘show’ . � present in �DQDP�� HQDP, aorist in �DFµD\�� HFµD\�� DUEHQDP ‘get drunk’ , ORJDQDP ‘bathe’ . � present in �NDQHP, aorist in �L��KDUNDQHP ‘strike’ .


[38]

� present in �QXP, aorist in �FµD\��]JHQXP ‘put on’ , QWµH QXP ‘read’ . � present in �QXP, aorist in �XD\��HUGQXP ‘swear’ . � present in �XP, aorist in �L��D\��KHQXP ‘weave’ , KDQXP ‘weave’ . It turns out that apart from the suppletive stems, the doubling of the material has yielded only two new types, viz. KDUNDQHP�� KDUL and HUGQXP��HUGXD\. We will now see if trebling the material by including the class &�etymologies changes the picture: � present in �HP�� LP, aorist in �HFµL�� HFµD\�� D\UHP ‘burn’ , D\FµHP ‘visit’ , KD\FµHP ‘request’ , JDUãLP ‘detest’ , GDGDUHP ‘cease’ , HUHZLP ‘appear’ , WµHNµHP ‘twist’ , WµRYHP ‘enchant’ , ODNHP ‘lick’ , NHáHP ‘hurt’ , NRþµHP ‘call’ , þµXHP ‘set out’ , VSµ HP ‘scatter’ , SµDUDWHP ‘scatter’ , WRSµHP ‘beat’ , FµD\WHP ‘splash’ , SµDUHP ‘embrace’ , SµDULP ‘embrace’ . � present in �HP, aorist in �DFµL��DVHP ‘say’ . � present in �DP, aorist in �DFµL�� DFµD\�� DáDP ‘grind’ , VSD QDP ‘threaten’ , Sµ QJDP ‘sneeze’ . � present in �DP, aorist in �FµL��DWHDP ‘hate’ . � present in �DP, suppletive aorist in �D\��HUWµDP ‘go’ , aor. þµRJD\. � aorist imperative JRJ ‘say’ . � present in �DQHP�� DQLP, aorist in �L�� D\�� DQNDQLP ‘fall’ , EHNDQHP ‘break’ , EXFDQHP ‘nourish’ , [DFDQHP ‘bite’ , PDFDQLP ‘stick’ , SµODQLP ‘fall’ . � present in �DQDP, aorist in �DFµL�� DFµD\�� WµDQDP ‘soak’ , PR DQDP ‘forget’ , XUDQDP ‘deny’ . � present in �QHP, aorist in �DUL��D QHP ‘make’ . � present in �QXP, aorist in �L��D QXP ‘take’ . � present in �XP, aorist in �L��KHáXP ‘pour’ , ]HáXP ‘pour’ . Disregarding the suppletive stems, we have found only two new types, viz. D QHP��DUDUL and D QXP��D L. Inclusion of the class ' etymologies adds yet another type to the picture, viz. �þµLP��HD\� � present in �HP��LP, aorist in �HFµL��HFµD\��HSµHP ‘cook’ , WµDWµDZHP ‘plunge’ , NUFHP ‘gnaw’ , V[DOHP ‘fail’ , V[DOLP ‘fail’ , FµUHP ‘violate’ , YD HP ‘fire’ , YD LP ‘burn’ . � present in �DP, aorist in �DFµL��NDUGDP ‘call’ . � present in �þµLP, aorist in �HD\��WµDNµþµLP ‘hide’ , PDWþµLP ‘approach’ . � present in �DQLP, aorist in �D\��DJDQLP ‘pass the night’ , ]HUFDQLP ‘escape’ . � present in �XP, aorist in �L��\DZHOXP ‘increase’ , ]HUFXP ‘strip’ . Thus, we established 17 categories on the basis of the 34 most widely attested etymological entries, then added two types on the basis of the next 37 entries, another two types for the next 33 entries, and another type for the last 12 entries. Though each step renders the picture more complete, it appears that the reconstruction of the morphological system is not strongly dependent on the number of etymologies. In order to see what we have missed, we will now


[39]

examine the material of class (, which is not included in Solta’s lists of verbal etyma but taken from Godel’s grammar: � present in �HP�� LP, aorist in �HFµL�� HFµD\�� DPDþµHP ‘am ashamed’ , JRþµHP ‘call’ , JRUFHP ‘work’ , KD\LP ‘look’ , PHFDUHP ‘honour’ , SµD\OHP ‘shine’ , VWLSHP ‘urge’ , WµD DPLP ‘wither’ , WµDUãDPLP ‘wither’ . � present in �HP, aorist in �DFµL��NDUHP ‘can’ . � present in �DP, aorist in �DFµL�� DFµD\�� NDP ‘stand’ , RUVDP ‘hunt’ , \XVDP ‘hope’ . � present in �HP, aorist in �L��KDQHP ‘draw’ , KHáXVHP ‘nail’ . � present JR\ ‘is’ . � present in �þµLP, aorist in �HD\��KDQJþµLP ‘rest’ , SµD[þµLP ‘flee’ . � present in �QþµLP, aorist in �HD\��HUNQþµLP ‘fear’ , PDUWQþµLP ‘fight’ . � present in �DQþµHP, aorist in �D\��PHáDQþµHP ‘sin’ . � present in �DQHP, aorist in �L�� KDUFµDQHP ‘ask’ , OXFµDQHP ‘light’ , PXFDQHP ‘introduce’ , VX]DQHP ‘plunge’ . � present in �DQHP, aorist in �HD\��\DQFµDQHP ‘trespass’ . � present in �DQLP, aorist in �DUD\��WDQLP ‘carry’ . � present in �LQLP, aorist in � ��OLQLP ‘become’ . � present in �DQDP�� HQDP, aorist in �DFµD\�� HFµD\�� QWµDQDP ‘run’ , PHUMHQDP ‘approach’ . � present in �QDP, aorist in �ML��MD\��ED QDP ‘lift’ , GD QDP ‘turn’ . � present in �QXP, aorist in �FµL�� QNHQXP ‘throw’ . � present in �QXP, aorist in �D\�� H QXP ‘get warm’ . � present in �QXP, aorist in �HD\�� D\WQXP ‘swell’ , NµDáFµQXP ‘am hungry’ , VDUWQXP ‘am shocked’ , ]DUWµQXP ‘awake’ . � present in �XP, aorist in �L��DUJHOXP ‘hinder’ , WµRáXP ‘let’ , ]HQXP ‘slaughter’ . It turns out that this additional collection of 40 etyma yields no less than eight new types: JR\;��QþµLP��HD\;��DQþµHP��D\;��DQHP��HD\;�WDQLP��WDUD\;�OLQLP�� Há ;� ED QDP�� EDUML and GD QDP�� GDUMD\;�and the common type of �QXP, �HD\, which is evidently productive, cf. WµDNµQXP beside WµDNµþµLP ‘hide’ and PDWQXP beside PDWþµLP ‘approach’ . Though these types may be recent, they are clearly built from inherited material. It follows that increasing the number of etymologies does not lead to a better picture of the original morphological system. We should rather start from an analysis of the attested categories and try to recover the separate elements and their distribution.

,,�The relation between present and aorist stems is the heart of the Armenian verbal system. If we disregard the suppletive stems including HP ‘am’ , JR\ ‘is’ and JRJ ‘say’ and discount HáDQLP ‘become’ (cf. Godel 1975: 54), we have found 26 types which can conveniently be classified into five major categories:


[40]

[41]

� present in �HP��LP��DP, aorist in �FµL��FµD\. � present in �HP��LP��DP, aorist in �L��D\. � present in �þµHP��þµLP, aorist in �D\. � present in �QHP��QLP��QDP��QXP��XP, aorist in �L��D\. � present in �QDP��QXP, aorist in �FµL��FµD\. It has long been recognized that the Fµ�aorist and the þµ�present correspond to the V�aorist and the VN�present of other Indo-European languages. The historical connections have remained unclear, however, because they seem to be at variance with the regular sound correspondences, cf. KDUFµDQHP ‘ask’ , aor. KDUFµL, which clearly represents PIE. *SU -VNH-, with reanalysis of the suffix �Fµ� as part of the stem. Like the imperfect *HEKHUHW which became an aorist HEHU ‘carried’ , the imperfect *HSU VNHW�became an aorist HKDUFµ, from which a nasal present was subsequently derived. I think that the reanalysis of suffixes and the derivation of nasal presents from aorist stems are the keys to a proper understanding of the Proto-Armenian verbal system. As the major types in �HP�� HFµL and �DP�� DFµL represent PIE. *�H�, *�\H�, *�H\H� and *� \H�, it seems probable that �LP�� HFµD\ reflects *� \H�, at least in part. The mixed type �HP�� DFµL seems to reflect the PIE. perfect. The Fµ�aorists of NHDP ‘live’ and DWHDP ‘hate’ can easily be analogical, and the same holds for the type in �DQDP�� DFµL�� DFµD\. Since none of these offers a starting-point for the analogical spread of the Fµ�aorist, we are left with the type in �QXP�� FµL�� FµD\ as the final possibility. Note that this is the only category where the 3rd sg. active form of the aorist does not end in ��H�DFµ, which is indeed a strong indication that we are dealing with an ancient type, cf. OQXP ‘fill’ , OFµL��HOLFµ, and QNHQXP ‘throw’ , QNHFµL, QN Fµ. While the former example clearly represents the PIE. root aorist *SO �, Klingenschmitt derives the latter from a root *JHV� (1982: 249). Be this as it may, it is clear that the few verbs of this type are too narrow a basis for the generalisation of the Fµ�aorist. We must therefore look for a suitable model among the other verb classes. Elsewhere (1987b [this vol., 79ff.]) I have argued that we find an original root aorist in 3rd sg. HNQ ‘came’ < *HJ � HPW, HG ‘put’ < *HGK W, HW ‘gave’ < *HG W, HNHU ‘ate’ < *HJ � HU+ � W, but sigmatic forms in 1st sg. HGL < *HGK VRP, HWX < *HG VRP, Há ‘became’ < *HNOHLVRP, OXD\ ‘heard’ < *NOHXV�. In the latter instance I submitted that the present tense OVHP ‘hear’ represents a sigmatic aorist stem with a secondary nasal infix *NOXQV�, in a similar way as PSHP ‘drink’ reflects the reduplicated present with an analogical nasal infix *SLPE�, cf. Vedic StEDWL ‘drinks’ . It appears that the sigmatic aorist spread at the expense of the root aorist at an early stage, when the infixed nasal present was still productive and *V had not yet become *K. Against this background it seems probable to me that the middle aorist in �HD\ represents *� V� (cf. Godel 1975: 121) and that the D�flexion originated in the athematic endings which began in a


[42]

nasal, as happened in the Greek sigmatic aorist. Starting from the expansion of the sigmatic aorist at an early stage and its eventual replacement by the Fµ�aorist, we may wonder what happened when intervocalic *�V� was lost. If there was any continuity between the two formations, the clue must be found in the aorists from which nasal presents were derived, between the creation of OVHP��OXD\ and that of QNHQXP, QNHFµL. I think that we have a sigmatic aorist in HUGQXP ‘swear’ , HUGXD\ < *GKUHXV�, and in D QXP ‘take’ , H QXP ‘get warm’ , aor. D L�� H D\ < *�UV�, as distinct from \D QHP ‘rise’ , \DUHD\ < *�U V�, imperative DUL. There is a clearly asigmatic, reduplicated aorist in D QHP ‘make’ , DUDUL. I agree with Klingenschmitt (1982: 201) that the aorist of WDQLP ‘carry’ , WDUD\ is built on the imperative WDU, as Gr. y)K?7 ‘eat’ is on the original athematic imperative *HVWKt, Vedic DGGKt. The type in �XP, which appears to have originated from �QXP after a root-final *O or *Q (cf. Pedersen 1906: 354 [= Kl.S., 132]), does not shed any light on the problem and can be left out of consideration. It follows that the clue must be found among the verbs with a stem-final obstruent and a derived nasal present. Limiting myself to the material adduced above, I list the following examples: � D RJDQHP��R RJDQHP ‘wet’ , JWDQHP ‘find’ , ONµDQHP ‘leave’ , DJDQLP ‘put on; pass the night’ , DQNDQLP ‘fall’ , EHNDQHP ‘break’ . � EXVDQLP ‘sprout’ , GL]DQHP ‘amass’ , OL]DQHP ‘lick’ , KHFDQLP ‘mount’ , DQLFDQHP ‘curse’ , DZFDQHP ‘anoint’ , HOX]DQHP ‘extract’ , HUJLFDQHP ‘tear’ , OXFDQHP ‘loosen’ , KDVDQHP ‘reach’ , PH DQLP ‘die’ , aor. PH D\ < *�UV�, XVDQLP ‘learn’ , WHVDQHP ‘see’ , FµXFµDQHP ‘show’ , EXFDQHP ‘nourish’ , [DFDQHP ‘bite’ , PDFDQLP ‘stick’ , ]HUFDQLP ‘escape’ , KDUFµDQHP ‘ask’ , OXFµDQHP ‘light’ , PXFDQHP ‘introduce’ , VX]DQHP ‘plunge’ . � KDUNDQHP ‘strike’ , ED QDP ‘lift’ , GD QDP ‘turn’ , aor. KDUL��EDUML��GDUMD\. While the first category apparently reflects thematic or root aorists, cf. Vedic iYLGDP ‘found’ , Gr. |��%#� ‘left’ , the second seems to represent a sigmatic formation, cf. PWDQHP ‘enter’ < *PXG�, aor. PWL, 3rd sg. HPXW beside PXFDQHP ‘introduce’ < *PHXGV�, and OR\V ‘light’ < *OHX - beside OXFµDQHP ‘light’ < *OHX V- (cf. Godel 1975: 126 and Pedersen 1906: 425 [= Kl.S., 203]). The mere fact that all these verb stems end in a dental affricate or fricative is a strong indication that the category reflects a sigmatic formation. Moreover, they often correspond to stems in dental plosives elsewhere, e.g. DQFµDQHP ‘pass’ , DQLFDQHP ‘curse’ , [DFDQHP ‘bite’ , aor. DQFµL��DQLFL�� [DFL, Gr. x�-#µ�� ‘meet’ , ²�J� #O ‘reproach’ , Vedic NK GDWL ‘chews’ . Thus, we arrive at the following scenario. When intervocalic *�V� was lost, the sigmatic aorist stem of most roots in an obstruent ended in �Fµ� < *�WV�, *�GV�, *�GKV�, *�NV�, *�JV�, *�JKV�. The new aorist marker then replaced the lost *�V� in productive categories, giving rise to the Fµ�aorist. The root-final obstruent was subsequently restored whenever possible (cf. Pedersen 1906: 429 [= Kl.S.,


[43]

207]), e.g. PXFDQHP�� [DFDQHP, with �F� < *�GV� replacing �Fµ� after the devoicing of the voiced plosives, similarly in OXFDQHP ‘loosen’ , OXFL, which was disambiguated from OXFµDQHP ‘light’ , OXFµL. The new aorist marker was evidently suppressed in KDUNDQHP ‘strike’ , KDUL, in order to disambiguate the verb from KDUFµDQHP ‘ask’ , KDUFµL. While the syllabic form of the nasal suffix �DQ� is regular after an original cluster in nasal presents derived from sigmatic aorists, the formation of ED QDP ‘lift’ and GD QDP ‘turn’ belongs to an earlier stage, when the root syllable *EUM�, *GUM� still contained a syllabic resonant: these verbs probably had thematic aorists. I hope to have shown that there is continuity between the Indo-European V�aorist and the Armenian Fµ�aorist and that there is no reason to postulate an unmotivated formation in *�VNH� for the latter, as earlier investigators have done. The rise of the new aorist marker provoked an enlargement of the Indo-European VN�present by the suffix *�\H�, which yielded the þµ�present, as Meillet has perceived (1936: 109). The only exception is the stem KDUFµ� ‘ask’ , which was reanalysed as an aorist. Though the derivation of Armenian �Fµ� and �þµ� from PIE. *�V� and *�VN�, respectively, is at variance with the well-known sound laws of the language, it is correct from a morphological point of view. Conversely, there is no reason to revise the sound laws on the basis of these correspondences, as is sometimes proposed. Sound laws must be established on the basis of isolated etymologies, such as KDUFµ� < *SU -VNH-, whereas morphological correspondences must be evaluated against the background of the morphological system as a whole. I claim that these two procedures are largely independent. [See also Kortlandt 1998b [this vol., 125] and 1999 [this vol., 129f.]. ]

117

[56]

Arm. FDáU ‘laughter’ * In his detailed and insightful analysis of the linguistic relationship between Armenian and Greek, James Clackson discusses Arm. FDáU (gen. FDáX) and Gr. �z�7+ ‘laughter’ at some length (1994: 126-132). He argues that “ if the form * HO+ V [thus Klingenschmitt 1982: 147] had been inherited into Armenian then one would expect the pre-apocope reflex *FHOXK which would have been assimilated, one assumes, into the normal X�declension” (127). But FDáU belongs to the small group of Armenian X�stem nouns with �U in the nominative which “ can all be derived from Proto-Indo-European *X�stem neuters” (126), e.g. FXQU ‘knee’ , DUWDZVU ‘tear’ , Gr. �²�/, 9�'/. Clackson identifies Arm. DVU ‘fleece’ (gen. DVX) with Latin SHF ‘cattle’ rather than Gr. %²�#+ ‘fleece’ and derives FDáU from * HOX, which is formally identical with Latin JHO ‘ice’ (131). As the discussion touches upon a series of controversial issues, I would like to clarify my position here. It is noteworthy that FXQU and DUWDZVU are not inflected in the singular and thereby differ from FDáU, DVU and PHáU ‘honey’ (gen. PHáX). This lack of inflection must have resulted from the apocope in view of the ease with which Armenian created new case forms, e.g. gen. FDáHU and FDáUX beside FDáX, also DVUX and DVUR\ beside DVX. Elsewhere I have derived DUWDZVU from a paradigm *GUD XU, obl. *GUDNUX�, which yielded an alternation *�DVXU, *�DZUX� (1985a: 61 [this vol., 61f.]). The analogy with *PD��LU ‘mother’ , *KZH�K�XU ‘sister’ , obl. *PDZU�, *KZHKU�, may have given rise to a pre-apocope paradigm *DUWD�Z�VXU, obl. *DUWDZ�V�U�, pl. DUWDVX�. In the case of FXQU, the comparison with Gr. �²�/, gen. �#/�²+, pl. �#º�� suggests a pre-apocope paradigm *FXQXU, obl. *FXQJ� (which can easily have been replaced by *FXQU�), pl. *FXQJD, distinct from the regular paradigm of *PHáXU, obl. *PHáX�, cf. Gr. µzK/, gen. µzK/#+. This explains the existence of two different flexion types of X�stem neuters in Armenian. Arm. DVU ‘fleece’ must be identified with Gr. %#�/- ‘small livestock’ (cf. Ruijgh 1992), with R�grade as in Arm. FXQU, Gr. �²�/, unlike Latin SHF , JHQ . In Armenian, the initial *S� developed to *K� which was lost before *R and


this vowel subsequently yielded D� in unstressed open syllables (cf. Kortlandt


[57]

[58]

1983b: 10 [this vol., 40]). Clackson objects that “ the absence of K� in this word does not prove that the vocalism was originally R” because “ there are clear cases of unexplainable aspiration or lack of it” (1994: 160). He mentions three instances, viz. D\F ‘goat’ , DFHP ‘I bring’ , and RJL beside KRJL ‘spirit’ (218). However, Arm. D\F and Gr. �©! < *DL � beside *D � in Indo-Iranian and Balto-Slavic suggest an early borrowing, Arm. DFHP must be identified with Latin JHU < *+ HV� rather than DJ < *+ H �, and Arm. KRJL beside RJL can easily have restored the initial K� on the basis of the verb which is attested in KHZDP ‘I pant’ (cf. Kortlandt 1983b: 14 [this vol., 43]). The actual alternation between initial K� and zero turns out to fit the expected distribution rather nicely (cf. also Kortlandt 1984b [this vol., 54ff.]). Following de Lamberterie (1978: 271), Clackson assumes “ lowering of *�H� to �D� before �X� in the following syllable” in order to explain the root vowel of FDáU < * HOX and DVU < *SH X (1994: 159 and 126f.). Neither of them mentions Pedersen's superior treatment (1906: 414-422 [= Kl.S., 192-200]). It seems to me that the isolated form KHUX ‘last year’ < *SHUXWL suffices to disprove the rule. The hackneyed example YDWµVXQ ‘sixty’ beside YHFµ ‘six’ must be explained differently, as is clear from NµD DVXQ ‘forty’ beside þµRUNµ ‘four’ and *DZWµQ beside HZWµQ ‘seven’ , also NµD RUG ‘fourth’ replacing *WµXU� and WDVQ ‘ten’ , �WDVDQ ‘-teen’ with �D� for *�H� from the ordinal (cf. Kortlandt 1994b [this vol., 99ff.]). The root vowel of JDUXQ ‘spring’ < *ZHVU and DULZQ ‘blood’ < *HVDU shows the regular development of *�HVD�, as distinct from *�L�K�D� < *� VD� in the aorist ending �HDFµ (cf. Kortlandt 1995a [this vol., 107ff.]) and from the anteconsonantal reflex of *�HVU� in NµHUE ‘by sister’ < *VZHVUEKL and MHUE��DNDO� ‘(taken) by hand’ < * KHVUEKL. While de Lamberterie points to Meillet as the 8UKHEHU�GHV�/DXWJHVHW]HV, the latter had evidently given up the idea by the time he wrote the (VTXLVVH (1936: 38, 39, 42, 55, 82). The initial vowel of DQXQ ‘name’ represents *R� < *+ � -, not *H� (cf. Kortlandt 1987a: 63 [this vol., 77]). Thus, I think that we must identify Arm. FDáU with Gr. �z�7+ < * HO+ V, obl. * O+ V�, cf. Arm. JHZá ‘village’ < *�O �L�, gen. JHá < *�O\� (Klingenschmitt 1982: 61, 147, 154). Clackson tries to derive Gr. �z�7+ from *� X�V�, acc. �z�7 (Î')J) for *�RZ�D� (Od. 20.346), dat. �z�Ç (|�K��#�) for *�RZL (Od. 18.100), which does not account for the stem *JHODV� but “ would match the reconstruction of a neuter X�stem for Armenian FDáU” (1994: 129f.). This leaves both the formation of Gr. �z�7+ and the root vowel of Arm. FDáU unexplained. The assumption that the two words represent different morphological formations in spite of their semantic identity must be rejected in favor of the hypothesis that the Armenian neuter X�stem paradigm reflects the “ old and unproductive type of noun inflection” (131) which is found in Greek. This does not affect the possibility of deriving the word from the same root as Latin JHO and English FKLOO.

$UP��FDáU µODXJKWHU¶ 119

Looking for traces of words in *� V, Clackson proposes to derive Arm. D\J ‘dawn’ from the locative *+ HXVL of *+ HXV V, Gr. �I+ (223). I find this etymology highly attractive. Clackson attributes the generalization of the vocalism to “ the widespread use of locutions such as QG� D\J ‘at dawn’ ” , to which I would attribute the loss of initial *K�, as I would in DQDJDQ ‘late’ , D DZDZW ‘morning’ , which continue other case forms, with loss of *K� as in D DJDVW ‘curtain’ , cf. �K�DJDQLP ‘I put on clothes’ (Kortlandt 1983b: 13 [this vol., 42f.]). Thus, we can reconstruct a paradigm *KDZ, obl. *KDZ�, acc. *KDJQ, loc. *�K�D\J. We now expect a pre-apocope paradigm *FHOX�K� < * HO+ V, acc. *FHORQ or *FHOXQ < * HO+ RVP or *� VP, obl. *FDO�D�� < * O+ V�, and contraction of the stem-final *�D� with the following ending. This pattern was evidently closer to *FXQX�U� < * RQX, obl. *FXQ�J�� < * RQZ� than to the regular X�stem flexion, which explains the transition of FDáU to the neuter paradigm. In the case of Gr. ¬ 'I+ ‘sweat’ , acc. ¬ 'Æ < -²A (cf. Il. 10.574), Arm. NµLUWQ < *VZLGURVP or *� VP, the oblique stem *NµLWD � < *VZLGUV� was replaced by regular *NµLUWDQ�. These words provide evidence for a rather close relationship between Armenian and Greek. [Note that the accent of Albanian GLUVs ‘sweat’ < *VZLGU� points to an original disyllabic form which may have been identical with its Greek and Armenian cognates.]

120

[8]

Arm. Q U ‘sister-in-law’ * Rüdiger Schmitt derives Arm. Q U ‘wife of the husband's (younger) brother’ from PIE. *�+�\HQ+ WHU�, Vedic \ WDU�, Greek J(¨)�9-J'-, Latin LDQLWU� (1996). His view calls for several comments. First of all, the reconstruction of an initial laryngeal is highly improbable. If it were correct, the Vedic reflex of the zero grade would be **LQL�, not \ �, and also the Latin reflex would probably be **LQL�, not LDQL�. The reason for reconstructing an initial laryngeal in this word is that PIE. *\� allegedly yielded � in Armenian. However, the latter hypothesis cannot be upheld. The classic comparison of Arm. XU ‘water’ with Lith. M UD ‘sea’ is certainly false because the latter word cannot be separated from Old Prussian ZXUV ‘pond’ and Vedic Y U� ‘water’ . The expected reflex of PIE. *\� is zero in Armenian, as is clear from the pronoun RU ‘who, which’ , where the relative replaced the interrogative, as in Polish MDN ‘how’ beside Russian NDN� (cf. Kortlandt 1983b: 11 [this vol., 41] and Clackson 1994: 52). Thus, we must start from *\HQ� yielding *LQ� and subsequent reduction and loss of the initial vowel in pretonic position. Secondly, Schmitt assumes vocalization of the medial laryngeal resulting in a stem *\HQDWHU�, followed by a “ spontaneous sound change” which assimilated the medial vowel to the surrounding vowels so as to yield *\HQHWHU�. This is an DG� KRF assumption because it is not supported by other instances. Since a medial *�D� is difficult to get rid of, as Schmitt has made quite clear, the inescapable conclusion is that the laryngeal was not vocalized. In his survey of laryngeal developments in the Indo-European languages, Robert Beekes proposes that in Armenian medial laryngeals were vocalized before clusters but not before single consonants (1988b: 77). Thus, we find vocalization in FQDZá ‘parent’ < *�+WO�, HUDVWDQNµ ‘buttocks’ < *-+ W-, DáDZUL ‘mill’ < *�+WU�, but not in GXVWU ‘daughter’ < *�+W U, JHáPQ ‘wool’ < *�+PHQ�, DUPXNQ ‘elbow’ < *�+PR�. If this is correct, we expect the following paradigm of the word *\HQ+WHU� before the apocope:


$UP� Q U µVLVWHU�LQ�ODZ¶� 121

sg. nom. *LQGLU pl. nom. *LQGHUHK acc. *LQGHUDQ acc. *LQGHUDV gen. *DQDZUR gen. *DQDUFµX inst. *DQDUEL inst. *DQDUELK It is clear that this paradigm could not survive. In fact, it seems probable to me that it never arose in the first place because the loss of *W before syllabic *U (at stage 13a of Kortlandt 1980b: 102 [this vol., 30]) provided a good motivation for eliminating the dental obstruent from the paradigm altogether. I therefore think that we can reconstruct a word *LQLU, *LQHU��‘sister-in-law’ beside *PD\LU, *PD�Z�U� ‘mother’ and *[ � HXU, *[ � H�K�U��‘sister’ . Regularization of the paradigm then produced the pre-apocope nom.sg. form *LQH\LU which underlies the attested forms of the word Q U. The analysis proposed here now confirms Beekes' rule for the vocalization of medial laryngeals in Armenian. [See also Kortlandt 1998a [this vol., 122ff.]. ]

122

[16]

The development of *\��in Armenian* There is no agreement on the development of initial *\�� in Armenian (cf. Clackson 1994: 52, with references): (1) XU�‘water’ , Lith. M UD�‘sea’ , (2) MHU�‘you (pl.)’ , Skt. \ \iP�‘id.’ , (3a) OHDUG�‘liver’ , Skt. \iN W�‘id.’ , (3b) OXF�‘yoke’ , Skt. \XJiP�‘id.’ , (4) RU�‘who’ , Skt. \i��‘id.’ . These developments are phonetically regular according to Meillet (1), Pedersen (2), Hamp (3) and Pisani (4), who cannot all be right. We must therefore reconsider the evidence. Lithuanian M UD� cannot be separated from Old Prussian ZXUV� ‘pond’ , Skt. Y U�‘water’ , Toch. A ZlU, B ZDU�‘id.’ , and must therefore have an initial M��of secondary origin, so that the connection with Arm. XU�must be abandoned. The latter word may represent original *G � \��(cf. Pedersen 1906: 429 [= Kl.S., 207]). Pronominal MH��can only represent *\X��plus *�H�. I think that we have to start from an accusative *\ZH, with the initial semivowel of French KXLW�‘eight’ , which was created on the analogy of the singular pronoun *WZH� that developed into NµH�. The phonetic merger of the labialized palatal semivowel with the reflex of original *

� Z�, e.g. in MD\Q�‘voice’ , Russ. ]YRQ�‘chime’ , can be compared with the merger of *Z�� with *J � � �� into J�� and with the merger of voiceless *VZ��> *KZ��with *N � ��into Nµ�. As the development of *\��to O��(3) is phonetically improbable and has no structural parallel in Armenian, the initial consonant of OHDUG�and OXF�must be analogical. The motivation for an analogical addition of O�� was perhaps the phonetic loss of *\��(4), which shortened the root. The model for the additional O�� was provided by the word for ‘fat’ *OLS�, as in the Germanic cognate for ‘liver’ , and by the verb OXFDQHP� ‘unyoke’ (e.g., Pokorny 1959: 504, 509). Indeed, pronominal R�, X��can only represent earlier *\R��(Kortlandt 1983b: 11 [this vol., 41], Mawet 1986: 86), cf. Polish MDN�‘how’ . While initial *\��was lost in the classical language, the dialects show a * Reprinted from $QQXDO�RI�$UPHQLDQ�/LQJXLVWLFV 19 (1998), 15-18.

7KH�GHYHORSPHQW�RI *y-�LQ�$UPHQLDQ 123

[17]

more complicated picture. As Jos Weitenberg has convincingly argued (1986), original *\��can be reflected as the voiced counterpart of K�, for which I shall write +�. Voiced +��always represents *\��and does not correspond to classical K�, e.g. Muš +XP� ‘to whom?’ versus KXP� ‘raw’ , also +XU� ‘where?’ , +LVXQ�‘fifty’ . The only instance of classical K�� for *\�� is found in KLP�beside LP� ‘for what?’ . It appears that +��often reflects the prefix \��< *LQ�. We may therefore have a look at what happened to the preposition L�< *LQ, which is written \��before vowels. Interestingly, the prefix *LQ�� always appears as \�� (or Q�, cf. Meillet 1936: 96) before a vowel, never as *L��before a consonant. It is therefore highly probable that initial *L�� was lost phonetically in pretonic syllables (cf. also Kortlandt 1997 [this vol., 120f.]), as a result of which the prefix was eliminated before consonants. The loss of pretonic *L�� can be identified chronologically with the syncope (stage 22 of Kortlandt 1980b: 104 [this vol., 31]). It can also be identified with the loss of the palatal element in the genitives RUGZR\, DQXR\�of RUGL�‘son’ , DQLZ�‘wheel’ , where the segment carrying the palatal feature adopted the rounding and back articulation of the following segment. At this stage, there was no phonemic opposition between /i/ and /y/ in the language (cf. Kortlandt 1976: 99). While initial *L�� lost its palatal feature and was reduced to shwa, which was subsequently lost before single consonants, the loss of the palatal feature in an initial *\��yielded voiced +�, which was eliminated in the majority of the dialects. In the meantime there was a complementary distribution between initial shwa before consonants and initial +��before vowels. When initial L��was restored on the basis of monosyllables, including the preposition L, the phonemic status of +��was subject to change in the separate dialects. Initial *+L�� yielded KL�� in the monosyllable KLP� ‘why’ , whereas pretonic *KL�� became voiced in +LVXQ, written \LVXQ� ‘fifty’ , as opposed to monosyllabic KLQJ� ‘five’ . A similar distribution appears in the Autun glossary, where we find KL�� in monosyllabic KLRFW�< HZWµQ� ‘seven’ versus pretonic L�� in LVFKXQ�< \LVXQ�‘fifty’ , cf. also KXQF�< \DZQNµ�‘eyebrows’ . When the preposition L�was restored before words with initial shwa plus consonant or initial +�� plus vowel, this shwa or +��lost its phonemic status after the vocalic preposition and was suppressed or reduced to an automatic glide. Once again, there was no phonemic opposition between /i/ and /y/, or between shwa and /H/, but there was an opposition between /i/ and shwa, or between /y/ and /H/, at this stage. This system is attested in the oldest manuscript of the Armenian Gospel (Matenadaran 6200, dated to 887), as Weitenberg has recently shown (1997). Here we find the preposition L�written L�\��before vowels and unwritten before word-initial +�, which is written \�. The reason for this remarkable orthography is evidently that +��was automatically lost after the preposition L, so that the preposition could not be restored without destroying the phonemic make-up of a


word beginning with +�� and was therefore omitted. The spelling \VXQ� ‘fifty’ may represent initial shwa plus yod, phonemically /Hisun/, cf. L� ãWHPDUDQV� for \ãWHPDUDQV� ‘in the secret chambers’ with initial yod plus shwa (Weitenberg 1997: 47, fn.3). It appears that the syllabification of these forms was different in the respective dialects.

125

[20]

Arm. JRP�‘am’ * Arm. JRP� ‘I am, exist, subsist’ is usually compared with Gothic ZDV�‘was’ (e.g. Meillet 1936: 112) and identified as an original perfect of the root *+ � ZHV�, Vedic YiVDWL� ‘dwells’ , Hittite KXLã]L� ‘lives’ (cf. Klingenschmitt 1982: 260, Clackson 1994: 105). It remains unclear why the perfect should have replaced the original present tense in this verb. Elsewhere I have suggested that the Germanic preterit *ZDV�, pl. *Z ]� ‘was, were’ is better derived from a compound of the root *+ � HV�, viz. *XE V�, *XE ]�, which can be compared with Greek Â%J�µ� ‘I subsist’ , Latin VXEVXP�‘I am present’ (Kortlandt 1992b: 105). This etymology explains not only the alternation of sg. *ZDV�, pl. *Z ]�, but also the present tense forms Old English HDUW�‘thou art’ , pl. Anglian �H�DURQ�and Old Swedish DUX�‘are’ , which can now be regarded as a back formation from * V��on the analogy of the preterits * N� ‘traveled’ , * O��‘nourished’ , * Q� ‘breathed’ of the verbs *DN�, *DO�, *DQ�.

The previous existence of a perfect 3sg. * VH, 3pl. * VU� beside an imperfect 3sg. * VW, 3pl. * VQW� is supported not only by Vedic VD, V~U�beside V, VDQ, but also by evidence from Greek, Slavic, and Celtic (cf. Kortlandt

1986b). The perfect endings in the Homeric imperfect 1sg. ��, 2sg. �)K�, 3sg. �J-� are totally unmotivated if we start from an original imperfect, but quite natural if we start from an R�grade perfect, where sg. * V�� alternated with pl. * V�. The Slavic imperfect is a compound formation of a deverbal noun followed by the original perfect of the verb ‘to be’ , e.g. 3sg. �DãH� reflecting * VH. The perfect stem * V� of the root *HV��evidently served as a model for the creation of the Celtic �preterit, e.g. Old Irish WiFK� < *W N�, UiWK� < *U W��of WHFKLG�‘flees’ < *WHN�, UHWKLG�‘runs’ < *UHW�. We must therefore consider the possibility that the paradigm of Arm. JRP, like the English form DUH, represents a back formation from the original perfect which underlies the English preterit ZDV, ZHUH. Thus, I propose to derive Arm. JR�� from *ZR� < *XSRV�, which was created to supply a present tense on the basis of the preterit 3sg. *XS VH� < *XS�R��H�RVH� ‘Á%��, suberat’ . Another example of a prefix which has been incorporated into a verbal root is provided by XWHP� ‘I eat’ < *VRP� plus *HG�, Vedic ViP�DG�, Spanish FRPR�< FRP�HGR� ‘I eat’ (cf. Kortlandt 1986a: 40 [this


vol., 70]).

�

126

[12]

Armenian glottalization revisited* Andrzej Pisowicz has recently (1997) commented on my account of the Armenian second consonant shift (1978a [this vol., 20ff.]). As he has evidently misunderstood my argumentation, I shall try to clarify the matter here. Pisowicz agrees with my view that the so-called voiced aspirates, e.g. Gµ, are a secondary development from plain voiced stops such as G�in the Central Armenian dialects. They subsequently tended to lose the voicing in the initial part of their duration and to shift the aspiration as breathy voice (murmur) onto the following vowel, which received a lower pitch (cf. Allen 1950: 200). I shall write [H] for this breathy voice (or voiced aspiration), as distinct from [h] for voiceless breath, e.g. in the voiceless aspirate Wµ = [th]. Thus, we are dealing with a development of G� > Gµ� = [dH] > [tH], distinct from [th], in the Central Armenian dialects from Erevan to Sebastia. This development has a perfect analogue in Panjabi, where the Indic voiced aspirates yielded plain voiceless stops plus a low tone on the following vowel, e.g. JKR > Nz � ‘horse’ (cf. Kortlandt 1985d: 193). It follows that the merger of voiced aspirates with voiceless stops requires two distinct steps: first the devoicing of the initial part of the consonant (as in the Central Armenian dialects) and then the absorption of breathy voice by the following vowel (or devoicing so as to yield a merger with the voiceless aspirates). The main disagreement between Pisowicz and myself concerns the primary or secondary character of glottalization in East Armenian S, W, N. Pisowicz thinks that the glottalized sounds “ are rather secondary as their frequency in OA was the lowest” and that this “ points to them as unmarked members of the existing oppositions” (1997: 217). This argument is entirely obscure to me. If a feature is infrequent, why should it be secondary? On the contrary, I would rather expect it to be a precious archaism. Is an infrequent feature characteristic of unmarked members of existing oppositions? I would rather claim the opposite! Pisowicz adduces the presence of glottalized consonants in Hemšin and Zeyt‘un in words with original E, G, J�as an argument in favor of “ viewing all glottalized consonants in Armenian as secondary”


(ibidem). This is of course nonsense, as Pisowicz knows perfectly well because

$UPHQLDQ�JORWWDOL]DWLRQ�UHYLVLWHG 127

[13]

he does not consider the presence of voiced consonants in West Armenian in words with original S, W, N� as an argument in favor of viewing all voiced consonants in Armenian as secondary. The real question is whether the Armenian glottalization arose at Pisowicz's “ first stage of the Eastern development” (1997: 226) or was already present in Proto-Armenian (as I claim). The crucial issue is the development in the Southern dialects, where S, W, N� and E, G, J� changed places. According to Pisowicz's account, we have to assume the following sequence of events (1997: 222f.): W1. voiced E�> murmured [pH], W2. voiceless S�> fortis S�> voiced E, W3. murmured [pH] > voiceless [p]. Both S�and E�remained distinct from the voiceless aspirate Sµ�= [ph], where the voice onset time is further delayed. The difficulty in this account is that it remains unclear how S�and E�can have changed places. If we denote fortis S� by [pb], the combination of lengthening of the voice onset time in E�> [pH] > S�and shortening of the voice onset time in S�> [pb] > E�evidently did not result in a phonemic merger. The hold and release in the non-aspirated stop is no longer than in the voiced aspirate (cf. Khachaturian 1983: 61) and there is no reason to assume that it ever was. We must therefore ask what the fortition and subsequent lenition in S�> [pb] > E�really stand for. If consonantal length did not play a role, the fortis character must be identified with a laryngeal feature, viz. glottalization. In fact, the presence of a glottal closure explains why the voice onset time was shortened in S�> [pb] > E�while it was lengthened in E�> [pH] > S. The next question now is: how old was the glottalization? For Eastern Armenian Pisowicz assumes that the plain voiceless stops S, W, N� “ acquired glottalization as the distinctive feature. This process may be regarded as a continuation of the fortition” (1997: 225f.): E1. voiced E�> murmured [pH], E2. voiceless S�> fortis S�> glottalized [p'], E3. rise of new voiced E�in loan-words. For the dialect of Muš Pisowicz adds for medial and final positions: E4. glottalized [p'] > voiced E, E5. murmured [pH] > voiceless [ph]. Since the developments E2 plus E4 are equivalent to W2, it is much simpler to identify the fortition with the rise of glottalization. Since the glottalization must also be assumed for the Southern dialects, as was pointed out above, it can be reconstructed for Proto-Armenian. This eliminates the “ fortition” as an early Armenian development and relegates the rise of glottalization to prehistoric times.


[14]

Deriving the Van dialect from the Central group, Pisowicz assumes the following development (1997: 229): E6. murmured [pH] > glottalized [p’]. This is a highly unnatural development because glottalization and breathy voice are opposite laryngeal features. It is much more probable that glottalized [p’] derives from plain [p] which arose from the devoicing of E. In fact, there seems to be an original complementary distribution between the dialects with voiced aspirates and the dialects where Adjarian’s law operated, which points to an early isogloss between the Central dialects (including Ararat) on the one hand and the dialects of Van and Karabagh on the other. The geographical distribution of the latter dialectal type, which is also found in isolated pockets in the Ararat region, suggests that the absence of voiced aspirates is an archaism here. I therefore stick to the relative chronology which I proposed twenty years ago (1978a: 13 [this vol., 23ff.]): A1. voiced G�> Gµ�> [tH] in the Central dialects (from Erevan to Sebastia), A2. devoicing of G�> [t] in the Southeastern dialects (Sasun, Van, Karabagh), A3. West Armenian consonant shift: W�= [t’] > G, A4. local developments: W�> G, W�> Wµ, [tH] > [th]. The essential point in my account is that the Southern dialects, e.g. Sasun, must be derived from the Southeastern (Van) type, not from the Central dialects.

129

[48]

The Armenian causative* The regular formation of the Armenian causative is a nasal present derived from a sigmatic aorist of a stem in an X-diphthong, viz. �XFµDQHP, aor. �XFµL, 3rd sg. �R\Fµ� (e.g. Meillet 1936: 116). Though it is reminiscent of the Greek verbs in -JE7, the Slavic verbs in �RYDWL, and the Lithuanian verbs in �iXWL and �~RWL (cf. Kortlandt 1995b), there can be no historical connection because the latter types are denominatives and comparatively recent. The formation in �XFµDQHP has variants in �XVDQHP and �X]DQHP in NRUXVDQHP ‘lose’ and HOX]DQHP ‘extract’ , QNOX]DQHP ‘submerge’ , SµOX]DQHP ‘cause to fall’ (Meillet 1936: 117, Klingenschmitt 1982: 263). This corroborates the historical connection with the sigmatic aorist (cf. Kortlandt 1987b: 51 [this vol., 80f.] and 1996b: 42 [this vol., 115f.]). The suffix �XFµ�, �XV�, �X]� was evidently extracted from the sigmatic aorist of roots ending in �X� plus obstruent, cf. especially PXFDQHP ‘introduce’ , aor. PXFL, 3rd sg. HPR\F beside PWDQHP ‘enter’ , aor. PWL, 3rd sg. HPXW. It thus appears that the causative is a further development of the factitive type attested in OXFµDQHP ‘light’ , OXFDQHP ‘loosen’ , VX]DQHP ‘plunge’ (cf. Pedersen 1906: 425 [= Kl.S., 203]). The verb HOX]DQHP ‘extract’ played a pivotal role in the development of the formation because the vocalization of the initial laryngeal rendered the root disyllabic, cf. Greek y�JE)#µ�� ‘shall come’ . The verb HODQHP ‘go out/up’ , aor. HOL, 3rd sg. HO probably reflects an original root aorist *HOXG�, which was thematicized in Greek y�/K- and Old Irish OXLG ‘went’ . Perhaps the most remarkable formation is the causative of ]JDP ‘feel, hear’ , viz. ]HNXFµDQHP ‘warn’ , which is built on the suppletive aorist stem HN� ‘came’ of JD�, where H� reflects the augment and the root� �N� is from Indo-European *J � HP� (cf. Klingenschmitt 1982: 263).

The new formation in �XFµDQHP replaced the Indo-European causative in *�HLH�, which had merged with the primary thematic flexion at an early stage, as is clear from the development of HUHNµ� ‘three’ < *WUHLHV. The rise of active sigmatic aorists with factitive meaning can be compared with the same development in Greek |)-�)� ‘placed’ , | )� ‘plunged’ , |Q )� ‘produced’ beside the intransitive root aorists |)-��, | �, |Q � (cf. Chantraine 1967: 181). The generalization of X-vocalism in the causative suffix �XFµ� is * Reprinted from $QQXDO�RI�$UPHQLDQ�/LQJXLVWLFV 20 (1999), 47-49.


reminiscent of the spread of �\Y� in the Slavic secondary imperfectives, which originated with roots in an X-diphthong (cf. Vaillant 1966: 484). We may therefore surmise that the predecessors of the verbs EXVDQLP ‘sprout’ < *EKRX� and FµXFµDQHP ‘show’ < *VNHX� played some part in the Armenian development. The latter verb may have been pivotal in the sense that the root vowel could be reanalysed as part of the suffix. The middle present EXVDQLP ‘grow, take root’ is evidently built on the aorist EXVD\, 3rd sg. EXVDZ, perhaps on the analogy of XVDQLP ‘learn’ , aor. XVD\ < *HX �. The formation of EXVD\ is an unsolved problem. Since the only other verb with the same suffix is OVHP ‘hear’ , aor. OXD\, for which I have proposed a present with a nasal infix derived from a sigmatic aorist stem * OXQV� (Kortlandt 1987b: 50 [this vol., 80]), I am inclined to assume a sigmatic derivative stem meaning ‘to grow’ beside the common Indo-European root with the meaning ‘to be’ . A present tense *EK+XHVPL may have given rise to Armenian *EXQV�, Greek QE#µ�� ‘grow’ , and the Old Irish subjunctive EH�. The elimination of the suffix in the Greek intransitive aorist |QP� ‘grew’ can be compared with the elimination of the final consonant of the root *J � HV� in |)�� ‘(the fire) went out’ . This would account for the particular meaning of the Greek verb.

131

Arm. DULZQ�‘blood’ * “ The root vowel of JDUXQ ‘spring’ < *ZHVU and DULZQ ‘blood’ < *HVDU shows the regular development of *�HVD�, as distinct from *�L�K�D� < *� VD� in the aorist ending �HDFµ (cf. Kortlandt 1995a [this vol., 107ff.]) and from the anteconsonantal reflex of *�HVU� in NµHUE ‘by sister’ < *VZHVUEKL and MHUE�DNDO� ‘(taken) by hand’ < * KHVUEKL” (Kortlandt 1996a: 57 [this vol., 118]). As Job has pointed out (1986: 24), a development of *HV � > *HKDU > DU� is at variance with my relative chronology (1980b), according to which the loss of intervocalic *�K� (my stage 10) preceded the elimination of the syllabic resonants (my stage 14). We must therefore assume vocalization of the laryngeal in *HV+U > *HVDU > DU�. It follows that the epenthetic vowel in *ZHVDU must be of analogical origin. Indeed, the phonetically regular development of *ZHVU yields **JH �, not JDU�, cf. MH � ‘hand’ < * KHVU. Clackson mistakenly states (2000: 28, fn. 6) that “ the expected outcome of *HV - is *HU�” if my chronology is correct. He evidently missed my “ assimilation to a following *U” (Kortlandt 1980b: 101 [this vol., 29]). I conclude that there is no evidence for Clackson’ s view (2000: 29) that “ the sequence *-HV - seems to have developed to �HU� via *�HKDU�” : in fact, �HU� is the direct phonetic reflex of *�HKU� before consonant plus vowel, where the *�K� was lost without lengthening the following *U. My view is now endorsed by Viredaz (2000: 304), who does not mention my publications. Clackson tries to find supporting evidence for his view in the verb (2000: 32): “ In some instances it is possible that the �HD� of Classical Armenian derives, not from Pre-Armenian *L�D (as is probably the case for QHDUG < *VQ Z W), but from *H�D. A case in point is the aorist conjugation of the H-conjugation verbs such as JRUFHDFµ ‘he worked’ from the verb JRUFHP ‘I work’ ” . This is highly improbable because there is no reason to assume that the ending �DFµ was ever added to the thematic vowel. On the other hand, there are clear indications that D�endings were added to an aorist stem in *� V� (cf. Godel 1975: 121 and Kortlandt 1996b: 40-42 [this vol., 114ff.]), which probably was the only source for the type in �HDFµ�

Clackson subscribes to the alleged lowering of *H to D before X in the


following syllable, which was first proposed and later abandoned by Meillet (cf.


[12]

Kortlandt 1996a: 57 [this vol., 118]). The isolated form KHUX ‘last year’ < *SHUXWL suffices to show that it is mistaken. Clackson’ s suggestion (2000: 34) to derive the initial vowel of DULZQ from lowering of *H� to D� before �LZ� in the following syllable therefore seems quite useless to me. In the dialects we find D� beside H� in a number of nouns, e.g. in HáED\U ‘brother’ < *EKU�, Há LZU ‘horn’ , HáWLZU ‘marsh’ , cf. also DáELZU ‘source’ < *EKU�, DUWDZVU ‘tear’ < *GU�, D X ‘stream’ < *VU�, where the expected phonetic reflex is H�, as in HUHNµ ‘three’ < *WU�, HUHZLP ‘appear’ , HU Fµ ‘elder’ , HUHV ‘face’ < *SU�, HUNDQ ‘mill-stone’ < *J � U�, etc. Since D� is limited to nouns and occurs side by side with H�, I think that it represents an original preposition after which the prothetic vowel did not arise, in particular the expected cognate *D of Slavic SR (cf. Pedersen 1906: 434-436 [= Kl.S., 222-224]). No such explanation is possible for the initial vowel of DULZQ. Clackson’ s suggestion (2000: 36) that DULZQ has DU� < *DKDU� < *+ � V+ � U� cannot be correct because vocalized *+ � � yielded H� in HODQHP ‘go out/up’ (Kortlandt 1999: 48 [this vol., 129]), HOX]DQHP ‘extract’ , HUHN ‘evening’ , with raising before a nasal L� in LQQ ‘nine’ , LP ‘mine’ , LQHZ ‘with me’ , with rounding before a (syncopated) rounded vowel R� in RUFDP ‘vomit’ (cf. Kortlandt 1987a: 62f. [this vol., 76f.]). Both DQXQ ‘name’ and DWDPQ ‘tooth’ have *+ � �, not *+ � �, as is clear from Greek ²�#µ�, ²�/µ�, ± I�, ± #E+, and especially �I�/µ#+ ‘nameless’ , �7 ²+ ‘toothless’ < * -+ � -, which are archaic formations, cf. w�I�/µ#+, w�² 7�, also �¬µ7 z7 ‘(the teeth) set on edge’ , and Phrygian RQRPDQ ‘name’ (cf. Kortlandt 1987a: 63 [this vol., 77]). The Aeolic form | #�-J+ can easily be explained as secondary on the basis of the verb | 7 ‘eat’ . The root *+ � HG� ‘bite, sting’ is also attested in Lith. ~RGDV, Latv. X{GV ‘gnat’ and in Gr. ± E�� ‘pain’ , �I /�#+ ‘painless’ . The verb \�D QHP ‘rise’ , aor. \�DUHD\ represents *+ � HU�, as do Gr. ²'�/µ�� and Latin RULRU, while D Q ‘ram’ must be compared with Gr. w'�J�²+ (cf. Meillet 1936: 46), which points to an original *+ � -. There is no evidence for Arm. D� < *+ � -.

133

Û3�4�1��#FFLFT�

Û)JTUPSJDBMÛ1IPOPMPHZÛPGÛ$MBTTJDBMÛ

"SNFOJBOÛÛÛÛÛÛÛ 1SFGBDFÛÛ 8IFOÛ,PSUMBOEUÛUPMEÛNFÛUIBUÛIJTÛBSUJDMFTÛPOÛ"SNFOJBOÛXPVMEÛCFÛDPMMFDUFEÛ BOEÛ FEJUFE Û *Û TVHHFTUFEÛ UIBUÛ IFÛ TIPVMEÛ BEEÛ BÛ TZTUFNBUJDÛQSFTFOUBUJPOÛPGÛUIFÛIJTUPSJDBMÛQIPOPMPHZ ÛTPÛUIBUÛJUÛXPVMEÛCFÛFBTZÛUPÛGJOEÛIJTÛJEFBTÛBOEÛUPÛTFFÛUIFÛDPIFSFODF�Û)FÛBHSFFE ÛPOÛDPOEJUJPOÛUIBUÛ*ÛXPVMEÛXSJUFÛJU�Û"GUFSÛTPNFÛDPOTJEFSBUJPOÛ*ÛBDDFQUFE�Û*ÛBMXBZTÛTBXÛIJTÛBSUJDMFTÛJOÛNBOVTDSJQUÛ CFGPSFÛ UIFZÛXFSFÛ QVCMJTIFE Û BOEÛ PGUFOÛ VSHFEÛ IJNÛ UPÛ BEEÛ BÛGFXÛXPSETÛUPÛNBLFÛUIFNÛNPSFÛFBTJMZÛVOEFSTUBOEBCMF�Û)FÛBMNPTUÛBMXBZTÛEFOJFEÛ UPÛEPÛ TP ÛQPJOUJOHÛPVUÛ UIBUÛ FWFSZUIJOHÛIBEÛCFFOÛ TBJEÛBMSFBEZÛ JOÛUIFÛ UFYU�Û 0GÛ DPVSTFÛ IFÛXBTÛ SJHIU Û CVUÛ UPÛ SFBEÛ JUÛ JOÛ UIFÛXBZÛ SFRVJSFEÛ JTÛPGUFOÛSBUIFSÛEJGGJDVMU�ÛÛ 5IFÛ PUIFSÛ SFBTPOÛ JT Û PGÛ DPVSTF Û UIBUÛ "SNFOJBOÛ IJTUPSJDBMÛQIPOPMPHZÛJTÛOPUÛSFBMMZÛFBTZ ÛBOEÛUIBUÛUIFÛFYJTUJOHÛJOUSPEVDUJPOTÛBSFÛWFSZÛTIPSUÛ BOEÛ CZÛ OPXÛ TPNFÛ UXFOUZÛ ZFBSTÛ PME�Û 'PSÛ FYBNQMFÛ UIFÛ MBSZOHFBMT ÛXIJDIÛ BSFÛ OPXÛ TVDIÛ BOÛ JNQPSUBOUÛ TVCKFDU Û BSFÛ OPUÛ USFBUFEÛ JOÛ UIFNÛTJNQMZÛCFDBVTFÛBUÛUIBUÛUJNFÛJUÛXBTÛOPUÛZFUÛQPTTJCMFÛUPÛTBZÛNVDIÛBCPVUÛJU�Û1FSTPOBMMZ Û *Û BMTPÛ GPVOEÛ UIBUÛ JUÛ JTÛ PGUFOÛ EJGGJDVMUÛ UPÛ GJOEÛ UIJOHTÛ JOÛ UIFÛFYJTUJOHÛ JOUSPEVDUJPOT�Û0OFÛ PGÛNZÛ BJNT Û UIFO ÛXBTÛ UPÛ QSPWJEFÛ BÛ TVSWFZÛXIFSFÛFWFSZUIJOHÛXBTÛEJTDVTTFEÛJOÛBÛQMBDFÛXIFSFÛPOFÛDBOÛFBTJMZÛGJOEÛJU�Û5IBUÛJTÛF�H�ÛUIFÛSFBTPOÛXIZÛ*ÛBEEFEÛBÛEFUBJMFEÛUBCMFÛPGÛDPOUFOUT�Û*ÛGFFMÛUIBUÛ*Û IBWFÛ OPUÛ BMXBZTÛ TVDDFFEFEÛ JOÛ DMFBSMZÛ QSFTFOUJOHÛ UIFÛ JTTVFT Û TJNQMZÛCFDBVTFÛUIFÛQSPCMFNTÛBSFÛPGUFOÛTPÛDPNQMJDBUFE�ÛÛ *ÛIBWFÛOPUÛ FBSMJFSÛQVCMJTIFEÛPOÛ"SNFOJBO Û BOEÛ UIFÛPOMZÛBJNÛPGÛNZÛXSJUJOHÛ UIJTÛ JOUSPEVDUJPOÛ JTÛ UPÛ HJWFÛ BÛ TVSWFZÛ DPOUBJOJOHÛ,PSUMBOEU�TÛWJFXT ÛXIPTFÛXPSLÛIBTÛTPÛNVDIÛDPOUSJCVUFEÛUPÛ"SNFOJBOÛTUVEJFT�Û0OMZÛIFSFÛBOEÛUIFSFÛPOFÛNBZÛGJOEÛBÛSFNBSLÛXIJDIÛJTÛNZÛPXO�Û*ÛNBEFÛHSBUFGVMÛVTFÛ PGÛ UIFÛ FYJTUJOHÛ JOUSPEVDUJPOT Û BOEÛ PGÛ $MBDLTPO�TÛ FYDFMMFOUÛ SFDFOUÛTUVEZ ÛXIJDIÛEJTDVTTFTÛNBOZÛQSPCMFNTÛJOÛBÛWFSZÛDMFBSÛXBZÛUIPVHIÛ*ÛEPÛOPUÛBHSFFÛXJUIÛIJNÛPOÛBMMÛQPJOUT�ÛÛÛÛÛÛÛ

134� 5��6��3��%HHNHV��Û 'PSÛ UIFÛ TBLFÛ PGÛ UIFÛ VOJUZÛ PGÛ UIFÛ CPPL Û *Û VTFEÛ ,PSUMBOEU�TÛOPUBUJPOT ÛUIPVHIÛJOÛTPNFÛDBTFTÛ*ÛXPVMEÛQSFGFSÛBOPUIFSÛPOF�ÛÛÛÛÛÛÛÛÛ *ÛDPVMEÛOPUÛIBWFÛXSJUUFOÛUIJTÛXJUIPVUÛUIFÛDPOUJOVPVTÛTVQQPSUÛPGÛ,PSUMBOEU ÛXJUIÛXIPNÛ*ÛEJTDVTTFEÛNBOZÛQPJOUT�Û0GÛDPVSTF Û*ÛSFNBJOÛTPMFÛSFTQPOTJCMFÛGPSÛUIJTÛQSFTFOUBUJPO�Û*UÛXBTÛBÛHSFBUÛJOUFMMFDUVBMÛBEWFOUVSFÛUPÛHPÛ JOUPÛ UIFÛ EFUBJMTÛ PGÛ"SNFOJBOÛ IJTUPSJDBMÛ QIPOPMPHZ�Û.ZÛ QSFTFOUBUJPOÛNVTUÛBMTPÛCFÛTFFOÛBTÛBÛUPLFOÛPGÛNZÛHSBUJUVEFÛGPSÛ,PSUMBOEUnTÛDPOUJOVJOHÛTUJNVMBUJPOÛBOEÛGSJFOETIJQÛJOÛUIFÛZFBSTÛXFÛXPSLFEÛJOÛUIFÛ6OJWFSTJUZÛPGÛ-FJEFO�Û-FJEFO ÛTFQUFNCFSÛ��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ3PCFSUÛ4�1�Û#FFLFTÛÛÛ 5"#-&Û0'Û$0/5&/54ÛÛÛ1SFGBDFÛÛÛÛ��ÛÛ*/530%6$5*0/ÛÛ ��ÛÛ"SNFOJBOÛÛ Û ��ÛÛ5IFÛFBSMJFTUÛ"SNFOJBOTÛÛÛ Û ��ÛÛ4PVSDFTÛÛ Û ��ÛÛ%JBMFDUTÛÛÛ Û ��ÛÛ5IFÛBMQIBCFUÛÛ Û ��ÛÛ5IFÛQIPOFNJDÛTZTUFNÛÛ Û ��ÛÛ.PEFSOÛQSPOVODJBUJPOÛÛ ��ÛÛ*OEP�&VSPQFBOÛÛÛ Û ��ÛÛ5IFÛ1*&ÛQIPOFNJDÛTZTUFNÛÛ Û ��ÛÛ/PUFÛPOÛUIFÛMBSZOHFBMTÛÛÛ ��ÛÛ"SNFOJBOÛBOEÛ*OEP�&VSPQFBOÛÛ Û ��ÛÛ'SPNÛ1*&ÛUPÛ"SNFOJBO�ÛBÛTVSWFZÛÛ Û ��ÛÛ'SPNÛ"SNFOJBOÛUPÛ1*&�ÛBÛTVSWFZÛÛ Û ��ÛÛ5IFÛQPTJUJPOÛPGÛ"SNFOJBOÛÛÛ��Û"$$&/5ÛÛ ��ÛÛ5IFÛMPTTÛPGÛGJOBMÛTZMMBCMFTÛÛ Û ��Û3FUBJOFEÛGJOBMÛWPXFMTÛÛ ��ÛÛ1SFUPOJDÛWPXFMÛSFEVDUJPO�ÛUIFÛ"SNFOJBOÛWPXFMÛBMUFSOBUJPOTÛÛ��Û708&-4ÛÛ ��ÛÛ�FÛÛ Û BÛ�F��F�� C��FÛ�ÛJ�CFGPSFÛOBTBMÛÛ Û DÛ�F��BÛ

+LVWRULFDO�3KRQRORJ\�RI�&ODVVLFDO�$UPHQLDQ�� 135

Û ��ÛÛ�PÛÛ Û BÛ�P��P�� CÛ�P��V�CFGPSFÛOBTBMÛÛÛÛÛ Û DÛ�P��BÛÛÛ ��ÛÛ� �� ÛÛ� ÛÛ��Û%*1)5)0/(4ÛÛ ��ÛÛ�FJÛÛ ��ÛÛ�PJ�ÛÛÛÛÛÛÛÛÛÛÛÛ Û BÛ�PJ�� CÛ�PJ��BZÛÛ ��ÛÛ�FVÛÛ ��ÛÛ�PV�ÛÛÛÛÛÛÛÛÛÛÛÛ ��ÛÛ5IFÛ1*&ÛMPOHÛEJQIUIPOHTÛÛ��ÛÛ5)&Û1305)&5*$Û708&-ÛÛ��ÛÛ708&-Û$0/53"$5*0/Û"/%Û708&-Û"44*.*-"5*0/ÛÛ ��Û7PXFMÛDPOUSBDUJPOÛÛ ��ÛÛ7PXFMÛBTTJNJMBUJPOÛÛ��ÛÛ4&.*708&-4ÛÛ ��ÛÛ4FNJWPXFMTÛBTÛWPXFMTÛÛ Û ��ÛÛ1*&Û�JÛÛ Û ��ÛÛ1*&Û�VÛÛ ��ÛÛ4FNJWPXFMTÛBTÛDPOTPOBOUTÛÛ Û ��ÛÛ1*&ÛÛ�ZÛÛ Û Û BÛ�ZÛ�Û ÛÛ Û Û CÛ��B3Z� Û��P3Z �Û�Û�BZ3� Û�PZ3 �ÛÛ Û Û DÛ�Z��[FSPÛEÛ�ZJÛÛ�ÛTUSFTTFEÛIJ ÛBOEÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ1"SN�Û�IJÛ�ÛVOTUSFTTFEÛ ��ÛZJTVOÛÛÛ Û Û FÛÛ$MVTUFSTÛXJUIÛ�ZÛTFFÛ��ÛÛ Û ��ÛÛ1*&ÛÛ�XÛÛ Û Û BÛ�XÛ�ÛHÛÛ Û Û CÛ�XÛ�Û�X�ÛÛ Û Û DÛ�XÛ�ÛXÛJOÛPUIFSÛQPTJUJPOT ÛÛ Û Û EÛ�XÛ�Û[FSPÛÛ Û Û FÛ$MVTUFSTÛXJUIÛ�X�TFFÛ��ÛÛ

136� 5��6��3��%HHNHV��

Û��ÛÛ3&40/"/54ÛÛ ��ÛÛ4ZMMBCJDÛSFTPOBOUTÛÛ Û ��Û� ÛÛÛÛÛÛÛ Û ��Û� ÛÛ Û ��Û� ÛÛ Û ��Û� ÛÛ ��ÛÛ$POTPOBOUBMÛSFTPOBOUTÛÛ Û ��Û�SÛÛÛÛÛÛÛÛÛÛÛ��ÛÛ$MVTUFSÛÛ�ST�TFFÛ��ÛÛ Û ��Û�MÛÛ Û ��Û�OÛÛ Û Û BÛ�O��O�� CÛ�O��[FSPÛÛ Û ��Û�N�� BÛ�N��N�� CÛ�N��[FSPÛCFGPSFÛTÛÛÛÛÛ Û Û DÛ�N��O�� EÛ��N��[FSPÛÛ Û Û FÛ�NO��XOÛÛÛÛÛ Û Û GÛ�N��X�CFGPSF�BGUFSÛVÛÛÛÛÛÛÛ��Û1*&Û�TÛÛ ��Û�TÛÛ Û BÛ�TÛ�ÛT�� CÛ�T��I�� DÛ��T��L ÛÛ ��ÛÛ$MVTUFSTÛÛCFHJOOJOHÛXJUIÛ�TÛTFFÛ��ÛÛÛ��ÛÛ45014ÛÛ ��ÛÛ5IFÛ"SNFOJBOÛDPOTPOBOUÛTIJGUÛÛ ��ÛÛ-BCJBMTÛÛ Û ��ÛÛ�Q�� BÛ�Q��I��ÛÛ Û Û CÛ�Q��[FSPÛÛÛÛÛÛ Û Û DÛ�Q��X�� ÛÛ�CÛÛÛ Û ��ÛÛ�CI�� BÛ�CI��C�� CÛ�CI��X�CFUXFFOÛWPXFMTÛÛ Û ��ÛÛ$MVTUFSÛ�QT�TFFÛ��ÛÛ ��ÛÛ%FOUBMTÛÛ Û ��ÛÛ�U�� BÛ�U��[FSPÛÛ


Û Û Û CÛ�U��U�� DÛ�U��E�� EÛ��U��[FSPÛBGUFSÛO�� FÛ�U��U �� GÛ�U��X�CFGPSFÛSFTPOBOUÛHÛ�U��Z�CFUXFFOÛWPXFMT ÛXÛCFUXFFOÛCBDLÛWPXFMTÛÛÛÛÛ Û ��ÛÛ�EÛÛÛ Û ��ÛÛ�EIÛÛÛ Û ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛEFOUBMÛTFFÛ��ÛÛ ��ÛÛ1BMBUBMTÛÛ Û ��Û� ÛÛÛ Û Û BÛ� ��[FSPÛÛÛÛÛ Û Û CÛ� ��X�ÛÛÛÛÛ Û Û DÛ� ��TÛÛ Û ��ÛÛ� ÛÛÛ Û ��ÛÛ� I�ÛÛÛÛÛÛ Û ��ÛÛ%FQBMBUBMJ[BUJPOÛÛÛ Û ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛQBMBUBMÛTFFÛ��ÛÛÛ ��ÛÛ-BCJPWFMBSTÛBOEÛQVSFÛWFMBSTÛÛ Û ��ÛÛ�L ÛÛÛÛÛÛÛ Û Û BÛ�L Û�Û[FSPÛ ÛÛ Û Û CÛ�L Û�ÛHÛ ÛÛ Û Û DÛ�L ��ÛL ÛÛ Û Û EÛ�L Û�ÛTÛÛÛ Û ��ÛÛ�H ÛÛÛ Û ��ÛÛ�H I�� ÛÛ1BMBUBMJ[BUJPOÛÛÛÛÛÛÛ Û Û BÛ�L Û�Û �� CÛ�H ��Û �� DÛ�H I��Û �� CFUXFFOÛWPXFMTÛÛÛÛ Û ��Û$MVTUFSTÛCFHJOOJOHÛXJUIÛWFMBS�Û��ÛÛ ��Û$MVTUFSTÛTUPQÛ�ÛMBSZOHFBMÛTFFÛ��ÛÛ��ÛÛ-"3:/(&"-4ÛÛ ��ÛÛ8PSE�JOJUJBMÛÛ Û ��ÛÛCFGPSFÛWPXFM Û)7�ÛÛ Û ��ÛÛCFGPSFÛDPOTPOBOU Û)$�ÛÛ Û ��ÛÛCFGPSFÛSFTPOBOU Û)3�ÛÛ Û ��ÛÛCFGPSFÛTFNJWPXFM Û)8�ÛÛ ��ÛÛ8PSE�GJOBMÛÛ Û ��ÛÛBGUFSÛWPXFM Û�7)Û

138� 5��6��3��%HHNHV��Û Û ��ÛÛBGUFSÛDPOTPOBOU Û�$)ÛÛ Û ��ÛÛBGUFSÛSFTPOBOU Û�3)ÛÛ Û ��ÛÛBGUFSÛTFNJWPXFM Û�8)ÛÛ ��ÛÛ*OÛJOMBVUÛÛ Û ��ÛÛBGUFSÛWPXFM Û7)ÛÛ Û Û ��ÛÛCFGPSFÛWPXFM Û7)7�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ Û Û ��ÛÛCFGPSFÛDPOTPOBOU Û7)$ÛÛ Û Û ��ÛÛCFGPSFÛSFTPOBOU ÛÛ7)3ÛÛÛÛÛÛÛÛÛÛ Û Û ��ÛÛCFGPSFÛTFNJWPXFM Û7)8ÛÛ Û ��ÛÛBGUFSÛDPOTPOBOU Û$)ÛÛ Û Û ��ÛÛCFGPSFÛWPXFM Û$)7ÛÛ Û Û ��ÛÛCFGPSFÛDPOTPOBOU Û$)$ÛÛ Û Û ��ÛÛCFGPSFÛSFTPOBOU Û$)3ÛÛ Û Û ��ÛÛCFGPSFÛTFNJWPXFM Û$)8ÛÛ Û ��ÛÛBGUFSÛSFTPOBOU Û3)ÛÛ Û Û ��ÛÛCFGPSFÛWPXFM Û3)7�� ÛÛCFGPSFÛDPOTPOBOU Û3)$ÛÛ Û Û ��ÛÛCFGPSFÛSFTPOBOU Û3)3ÛÛ Û Û ��ÛÛCFGPSFÛTFNJWPXFM Û3)8ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ��ÛÛBGUFSÛTFNJWPXFM Û8)ÛÛ Û Û ��ÛÛCFGPSFÛWPXFM Û8)7�ÛÛ Û Û ��ÛÛCFGPSFÛDPOTPOBOU Û8)$ÛÛ Û Û ��ÛÛCFGPSFÛSFTPOBOU Û8)3ÛÛ Û Û ��ÛÛCFGPSFÛTFNJWPXFM Û8)8ÛÛ ��ÛÛ5XPÛMBSZOHFBMTÛÛ Û ��Û))ÛÛ Û ��Û)3)ÛÛ Û ��Û)8)ÛÛ��ÛÛ$-645&34Û

�� $MVTUFSTÛCFHJOOJOHÛXJUIÛ�S�Û�STÛ�Û �� $MVTUFSTÛCFHJOOJOHÛXJUIÛTÛÛ Û BÛ�TZÛ�ÛZÛÛÛÛÛÛÛ Û CÛ�TXÛ�ÛL ÛÛÛÛÛ Û DÛÛ�TSÛ�Û ÛÛ Û EÛ�TQ��TQÛÛÛÛÛ Û FÛÛ�TU��TUÛÛÛ Û GÛÛ�T Û�TL ��TLÛ�ÛD ÛÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛMBCJBM�ÛÛÛ�QT��T�Û�QTU��TU� Û�Q �ÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛEFOUBMÛÛ Û BÛÛ�EIZÛ�Û ÛÛÛ Û CÛ�UXÛ�ÛL ÛÛÛÛ


Û Û DÛÛ�EXÛ�ÛLÛÛÛ Û EÛ�5 ÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛWFMBSÛÛ Û BÛ�LZÛ�Û ÛÛÛÛÛÛÛ Û CÛ� Û�ÛuÛÛÛÛÛÛÛ Û DÛÛ� TÛ�ÛD ÛuÛÛÛÛÛ Û EÛÛ�LT �L T��D ÛuÛÛÛ ��ÛÛTUPQÛ�ÛMBSZOHFBMÛÛÛ Û BÛ�Q)Û�ÛQ � ÛÛÛÛÛ Û CÛ�U)Û�ÛU � ��ÛÛÛ Û DÛ�L ) �L)��ÛYÛÛ��ÛÛ05)&3Û%&7&-01.&/54ÛÛ ��ÛJÛ�ÛBÛ�ÛFBÛÛÛ ��ÛF��FJ�CFGPSFÛu � � ÛÛÛÛÛÛÛ ��Û�JPÛ�ÛXPÛBOEÛJXÛ�ÛVÛ�ÛUV O FBO�� Û�PQ$��PX$��ÛV$ÛÛ ��Û5IFÛTFRVFODFÛ��O, �ÛÛÛ ��ÛÛ�BVH��BH ��BL�� ÛZ�FQFOUIFTJTÛÛÛ ��ÛX�FQFOUIFTJTÛÛ ��ÛÛF �J Û��FX �JX ÛÛÛ ��D ��T�CFGPSFÛD Û ÛÛÛÛ ��Û'JOBMÛ�ZÛBGUFSÛWPXFMÛÛ ��Û-PTTÛPGÛTUPQTÛCFUXFFOÛDPOTPOBOUTÛÛ ��Û.FUBUIFTJTÛÛ��ÛÛ3&-"5*7&Û$)30/0-0(:ÛÛ ��Û*OUFSNFEJBUFÛTUBHFTÛÛ ��Û3FMBUJWFÛDISPOPMPHZÛÛ ��Û&YBNQMFTÛÛÛÛÛÛ ��-2863(9'8-32�ÛÛÛÛÛÛ ÛÛ ��Û"SNFOJBOÛÛÛ ��Û5IFÛFBSMJFTUÛ"SNFOJBOTÛÛ 5SBEJUJPOBMMZÛ UIFÛ "SNFOJBOÛ BMQIBCFUÛ XBTÛ EFWJTFEÛ JOÛ ��Û#FGPSFÛUIBUÛUJNFÛUIFÛIJTUPSZÛPGÛUIFÛDPVOUSZÛJTÛQSFIJTUPSZ ÛFYDFQUÛUIBUÛJUÛJTÛNFOUJPOFEÛ CZÛ PUIFSÛ QFPQMFT�Û 5IFÛ "SNFOJBOTÛ DBMMÛ UIFNTFMWFTÛ)BZ Û QM�Û)BZL Û PGÛXIJDIÛ UIFÛ PSJHJOÛ JTÛ OPUÛ LOPXOÛ GPSÛ DFSUBJO�Û 5IFÛ FUZNPMPHZÛ CZÛ1*&Û�QPUJTÛ�NJHIUZ ÛMPSE�ÛDG�ÛCBZÛ�XPSE�Û�Û�FL, � XMWÛTFFNTÛJNQPTTJCMF ÛBTÛIÛ�Û�QÛXBTÛMPTUÛCFGPSFÛ�P�Û*OÛUIFÛ)JUUJUFÛBSDIJWFTÛBÛMBOEÛ)BZBTBÛJTÛNFOUJPOFE Û

140� 5��6��3��%HHNHV��XIJDIÛIBTÛCFFOÛBTTVNFEÛUPÛCFÛUIFÛTPVSDFÛPGÛUIFÛOBNF�Û#VUÛJOÛ"SNFOJBOÛUIFÛ�Z�ÛXPVMEÛQSPCBCMZÛIBWFÛCFFOÛMPTU�Û5IFÛDPOOFDUJPOÛXJUIÛUIFÛOBNFÛPGÛUIFÛ QFPQMFÛ PGÛ UIFÛ )BUUJÛ GSPNÛ XIJDIÛ UIFÛ )JUUJUFTÛ HPUÛ UIFJSÛ OBNFÛ GJUTÛ"SNFOJBOÛTPVOEÛMBXTÛBTTVNJOHÛUIBUÛUIFÛ)�ÛPGÛ)BUUJÛDPVMEÛCFDPNFÛ)�ÛJOÛ"SNFOJBO�Û5IFÛRVFTUJPOÛJTÛXIFUIFSÛUIFÛ"SNFOJBOTÛ MJWFEÛJOÛUIFÛ MBOEÛPGÛUIFÛ )BUUJ�Û 8IFOÛ UIFZÛ DBNFÛ GSPNÛ UIFÛ XFTUÛ BOEÛ PWFSSBOÛ UIFÛ )JUUJUFÛ&NQJSF Û UIFZÛNBZÛXFMMÛ IBWFÛ MJWFEÛ GPSÛ TPNFÛ UJNFÛXFTUÛ PGÛ UIFÛ6SBSUJBOÛUFSSJUPSZ�Û5IJTÛEFSJWBUJPOÛUIFSFGPSFÛNVTUÛCFÛTFSJPVTMZÛDPOTJEFSFE�Û5IFÛGBDUÛUIBUÛ UIFÛ OBNFÛ DBOÛ IBWFÛ EFWFMPQFEÛ GSPNÛ )BUUJÛ DBOÛ IBSEMZÛ CFÛ BÛDPJODJEFODF�Û*UÛNBZÛCFÛOPUFEÛUIBUÛUIFÛOBNFÛDPVMEÛIBWFÛ�ZÛGSPNÛ�TZ��ÛBOEÛUIFÛ)�ÛDPVMEÛEFSJWFÛGSPNÛBÛQ�ÛJGÛOPUÛGPMMPXFEÛCZÛ�P��ÛÛ 8FÛGJSTUÛGJOEÛUIFÛ"SNFOJBOTÛJOÛUIFÛBSFBÛPGÛUIFÛSFBMNÛPGÛ6SBSUV�Û4PÛJUÛ TFFNTÛ WFSZÛ QSPCBCMFÛ UIBUÛ UIFÛ "SNFOJBOTÛ DBNFÛ JOUPÛ UIFJSÛ IJTUPSJDBMÛMBOETÛ XIFOÛ UIFÛ SFBMNÛ PGÛ 6SBSUVÛ DPMMBQTFE Û CFUXFFOÛ ��Û BOEÛ ��Û UIFÛ"SNFOJBOTÛNBZÛUIFNTFMWFTÛIBWFÛQMBZFEÛBÛQBSUÛJOÛUIFÛFOEÛPGÛUIFÛ6SBSUJBOÛXPSME�Û 4PÛ UIFZÛ DPVMEÛ IBWFÛ MJWFEÛ OFBSCZ Û CVUÛ XFÛ IBWFÛ OPÛ NFBOTÛ PGÛLOPXJOHÛBUÛXIBUÛUJNFÛUIFZÛDBNFÛJOÛUIJTÛEJSFDUJPO�Û*UÛJTÛQPTTJCMFÛUIBUÛUIFZÛDBNFÛUPÛUIFÛBSFBÛBSPVOEÛ��Û#$ ÛXIFOÛUIFÛ)JUUJUFÛ&NQJSFÛEJTBQQFBSFE�Û/PSÛEPÛXFÛIBWFÛ BOZÛIJTUPSJDBMÛ JOEJDBUJPOÛ GSPNÛXIFSFÛ UIFZÛ DBNF Û GSPNÛUIFÛ&BTUÛPSÛGSPNÛUIFÛ8FTU�Û0OÛWFSZÛHFOFSBMÛHSPVOETÛsÛDPNQBSJTPOÛXJUIÛXIBUÛXFÛLOPXÛUPÛIBWFÛIBQQFOFEÛUIFÛBSSJWBMÛPGÛ UIFÛ)JUUJUFTÛBOEÛPGÛ UIFÛ1ISZHJBOTÛ sÛ JUÛ JTÛ NPSFÛ QSPCBCMFÛ UIBUÛ UIFZÛ DBNFÛ GSPNÛ UIFÛ 8FTU�Û)FSPEPUVT�ÛTUBUFNFOUÛ��ÛUIBUÛUIFZÛXFSFÛ&RUGÎN��POIKOI�Û�DPMPOJTUTÛPGÛUIFÛ1ISZHJBOT�ÛJTÛOPUÛWFSZÛSFMJBCMF�ÛTVDIÛTUBUFNFOUTÛCZÛDMBTTJDBMÛBVUIPSTÛBSFÛ PGUFOÛ XSPOH Û BOEÛ )FSPEPUVTÛ CBTFEÛ IJNTFMGÛ POÛ UIFJSÛ FRVJQNFOU�Û&VEPYPTÛ BQ�Û 4U�Û #Z[�Û T�W�Û RMENeAÛ TBZTÛ UIBUÛ UIFÛ "SNFOJBOTÛ T6Û FWN6ÛPOLL�ÛFRUGeZOUSIN�ÛCVUÛJUÛJTÛOPUÛDMFBSÛXIBUÛUIJTÛFYBDUMZÛNFBOT�Û4UJMM ÛJGÛUIFÛNFOUJPOÛPGÛUIFÛ1ISZHJBOTÛDPOUBJOTÛBÛIJTUPSJDBMÛFMFNFOU ÛJUÛNBZÛJNQMZÛUIBUÛUIFÛ"SNFOJBOTÛPODFÛMJWFEÛOFBSÛFBTUÛPGÛUIFÛ1ISZHJBOT�Û"ÛDPOGJSNBUJPOÛPGÛUIFJSÛXFTUFSOÛPSJHJOÛIBTÛPGUFOÛCFFOÛ TFFOÛ JOÛ UIFÛ TVQQPTFEÛDPOOFDUJPOÛPGÛ"SNFOJBOÛ XJUIÛ (SFFL�Û )PXFWFS Û UIJTÛ DPOOFDUJPOÛ JTÛ OPXÛ TFWFSFMZÛBUUBDLFEÛPOÛHPPEÛHSPVOET�Û#VUÛUIFSFÛBSFÛTPNFÛJOEJDBUJPOTÛUIBUÛQPJOUÛUPÛBÛDPOOFDUJPOÛ XJUIÛ 5ISBDJBOÛ BOEÛ "MCBOJBOÛ sÛ IBSEMZÛ XJUIÛ 1ISZHJBOÛ TFFÛCFMPXÛ��ÛÛ 8FÛBSFÛUPMEÛUIBUÛUIFÛ.FEFTÛVOEFSÛ,VBYBSFT Û��Û�Û��ÛCFBUÛUIFÛ"SNFOJBOT�Û"Û MJUUMFÛ MBUFSÛ UIFJSÛ MBOEÛCFDBNFÛQBSUÛ PGÛ UIFÛ1FSTJBOÛ&NQJSF�Û5IFÛOBNFÛ"SNFOJB ÛUIFÛPSJHJOÛPGÛXIJDIÛBMTPÛJTÛOPUÛFYBDUMZÛLOPXO ÛJTÛGJSTUÛNFOUJPOFEÛJOÛ%BSJVT�ÛJOTDSJQUJPOÛJOÛ#JTVUVO�Û8IJMFÛUIFÛ"LLBEJBOÛWFSTJPOÛTUJMMÛ TQFBLTÛPGÛµS¼v VÛ BOEÛ UIFÛ&MBNJUFÛ WFSTJPOÛ XIJDIÛ JTÛ PMEFSÛ UIBOÛ UIFÛ1FSTJBOÛIBTÛ)BSNJO� ÛUIFÛ0MEÛ1FSTJBOÛUFYUÛIBTÛ"SNJOB�Û5IFÛXPSEÛGPSÛUIFÛQFPQMFÛ "SNJOJZB�Û &MBN�Û )BSNJOVZBÛ QSPCBCMZÛ IBTÛ UIFÛ 1FSTJBOÛ TVGGJYÛ


BEEFE�Û)BSNJOVÛ�"SNFOJB�ÛPOMZÛ91IÛ��ÛJTÛEFSJWFEÛGSPNÛUIFÛOBNFÛPGÛUIFÛMBOE�Û*UÛQSPCBCMZÛJTÛUIFÛMBOEÛXIFSFÛ.FEFTÛPSÛ&MBNJUFTÛGJSTUÛFODPVOUFSFEÛUIFÛ "SNFOJBOT�Û 5IFÛ JOTDSJQUJPOÛ NFOUJPOTÛ BOÛ "SNFOJBOÛ DBMMFEÛ)BMEJUB ÛXIJDIÛ JTÛ BOÛ6SBSUJBOÛOBNF�Û)JTÛTPOÛSBJTFTÛBÛ SFCFMMJPOÛBHBJOTUÛ%BSJVTÛ JOÛ#BCZMPO�Û"OPUIFSÛ"SNFOJBOÛMFBETÛBOÛBSNZÛPGÛ%BSJVTÛBHBJOTUÛBÛSFCFMMJPOÛJOÛ"SNFOJB�Û4PÛBUÛ UIBUÛ UJNFÛUIFÛ"SNFOJBOTÛXFSFÛXFMMÛFTUBCMJTIFEÛJOÛUIFÛ1FSTJBOÛXPSME�ÛÛ *OÛ UIFÛ WPDBCVMBSZÛ 6SBSUJBOÛ XPSETÛ IBWFÛ CFFOÛ JEFOUJGJFE�Û 'SPNÛ*SBOJBO ÛIPXFWFS ÛBÛWFSZÛMBSHFÛOVNCFSÛPGÛCPSSPXJOHTÛIBWFÛCFFOÛQPJOUFEÛPVU�Û5IJTÛJTÛOPÛTVSQSJTF ÛPGÛDPVSTF ÛXIFOÛ"SNFOJBOÛXBTÛGPSÛTPÛMPOHÛVOEFSÛUIFÛ JOGMVFODFÛ PGÛ .FEFTÛ BOEÛ 1FSTJBOT�Û 5IFÛ PMEFTUÛ BOEÛ MBSHFTUÛ MBZFS ÛIPXFWFS Û DPNFTÛ GSPNÛ 1BSUIJBO�Û (FPHSBQIJDBMMZÛ UIJTÛ JTÛ POMZÛ UPÛ CFÛFYQFDUFE ÛCVUÛJUÛJNQMJFTÛUIBUÛUIFÛMBSHFTUÛJOGMVYÛEBUFTÛGSPNÛBGUFSÛUIFÛGBMMÛPGÛUIFÛ"DIBFNFOJEÛ&NQJSF�Û5IFÛRVFTUJPOÛXIFUIFSÛUIFÛPMEFTUÛ *SBOJBOÛMPBOTÛQSFEBUFÛUIFÛ"SNFOJBOÛMPTTÛPGÛGJOBMÛTZMMBCMFTÛIBTÛCFFOÛNVDIÛEFCBUFE�ÛUIFÛRVFTUJPOÛ JTÛ XIFUIFSÛ UIFÛ MPBOXPSETÛ IBWFÛ UIFÛ TBNFÛ TUFNWPXFMÛ BTÛ UIFÛPSJHJOBMÛ*SBOJBOÛXPSET�ÛÛ 'SPNÛ MBUFSÛ UJNFTÛ EBUFÛ UIFÛ CPSSPXJOHTÛ GSPNÛ (SFFLÛ BOEÛ 4ZSJBO ÛXIJDIÛBSFÛNVDIÛNPSFÛMJNJUFEÛJOÛOVNCFS�Û0MTFOÛ�� Û��Û�Û��ÛOPUFTÛJOÛUIFÛWPDBCVMBSZÛPGÛ UIFÛ#JCMFÛTPNFÛ��Û *SBOJBOÛ MPBOTÛBHBJOTUÛ�Û��Û(SFFLÛBOEÛ��Û4ZSJBOÛMPBOT�Û5IFÛOVNCFSÛPGÛJOIFSJUFEÛXPSET ÛJ�F�ÛXPSETÛXJUIÛBOÛ*OEP�&VSPQFBOÛ FUZNPMPHZ Û JTÛ OPUÛ MBSHF�Û POFÛ FTUJNBUFÛ HJWFTÛ ��Û XPSET�Û1PLPSOZÛ IBTÛ POMZÛ ��Û"SNFOJBOÛXPSET�Û 4PÛ UIFSFÛ SFNBJOTÛ BÛ WFSZÛ MBSHFÛBNPVOUÛPGÛXPSETÛPGÛVOLOPXOÛPSJHJO�ÛÛ $ISJTUJBOJUZÛ CFDBNFÛ UIFÛ TUBUFÛ SFMJHJPOÛ JOÛ ��Û5IFÛ BMQIBCFUÛXBTÛEFTJHOFEÛJOÛ��ÛCZÛBÛNPOLÛPSÛNJTTJPOBSZÛDBMMFEÛ.BTUPD ÛMBUFSÛ.FTSPQ ÛXIPÛTUBSUFEÛBÛ#JCMFÛUSBOTMBUJPOÛGSPNÛ(SFFL ÛXIJDIÛXBTÛGJOJTIFEÛJOÛ��Û*UÛ XBTÛ GJSTUÛ QSJOUFEÛ JOÛ �� Û JOÛ "NTUFSEBN�Û 5IFÛ QFSJPEÛ VOUJMÛ ��Û JTÛDPOTJEFSFEÛ UIFÛ $MBTTJDBMÛ MBOHVBHF Û UIFÛ MBOHVBHFÛ DBMMFEÛ PTLFEBS Û �UIFÛHPMEFOÛ MBOHVBHF�Û BÛ UFSNÛ DPJOFEÛ CZÛ UIFÛ .FLIJUIBSJTUÛ TDIPMBSTÛ JOÛ UIFÛ97**UIÛ DFOUVSZ�Û5IJTÛ MBOHVBHFÛXBTÛBSUJGJDJBMMZÛ SFUBJOFEÛ JOÛVTFÛVOUJMÛ UIFÛ9*9UIÛDFOUVSZ�Û0OFÛOPXÛEJTUJOHVJTIFTÛBÛ�QPTU�$MBTTJDBM�ÛQFSJPEÛ7*UIÛBOEÛ7**UIÛDFOUVSZÛBOEÛBÛQSF�NJEEMFÛ"SNFOJBOÛTUBHFÛ7***UIÛ�Û9*UIÛDFOUVSZ�Û.JEEMFÛ"SNFOJBOÛJTÛUIFÛQFSJPEÛPGÛUIFÛ9**UIÛ�Û97**UIÛDFOUVSJFTÛFTQFDJBMMZÛLOPXOÛ GSPNÛ $JMJDJB Û BGUFSÛ XIJDIÛ POFÛ TQFBLTÛ PGÛ .PEFSOÛ "SNFOJBO ÛXIJDIÛ PDDVSTÛ JOÛ UXPÛ GPSNT Û &BTUFSOÛ BOEÛ8FTUFSOÛ "SNFOJBO Û BOEÛ JOÛ BÛOVNCFSÛPGÛMPDBMÛEJBMFDUT�ÛÛ 5IFÛ QSPPGÛ UIBUÛ"SNFOJBOÛXBTÛ BOÛ JOEFQFOEFOUÛ CSBODIÛ PGÛ *OEP�&VSPQFBOÛXBTÛQSPWJEFEÛCZÛ)×CTDINBOOÛ��Û)JTÛFUZNPMPHZÛPGÛ��ÛJTÛOPUÛZFUÛSFQMBDFEÛJOÛBÛ&VSPQFBOÛMBOHVBHF�ÛÛÛÛÛÛ Û

142� 5��6��3��%HHNHV��Û ��Û4PVSDFTÛÛ 8IJMFÛXFÛIBWFÛJOTDSJQUJPOTÛTJODFÛUIFÛGJGUIÛDFOUVSZ Û MJUFSBSZÛUFYUTÛBSFÛ POMZÛ LOPXOÛ GSPNÛ MBUFSÛ NBOVTDSJQUT�Û 5IFÛ PMEFTUÛ NBOVTDSJQUÛ JTÛ UIFÛ.PTDPXÛ(PTQFMÛGSPNÛ��ÛUIPVHIÛUIJTÛJTÛOPUÛUIFÛCFTUÛNBOVTDSJQU�Û.PTUÛNBOVTDSJQUTÛ EBUFÛ GSPNÛ UIFÛ FOEÛ PGÛ UIFÛ9**UIÛ DFOUVSZÛ POXBSET�Û 5IFZ Û PGÛDPVSTF ÛDPOUBJOÛZPVOHFSÛFMFNFOUTÛBOEÛNVTUÛCFÛVTFEÛXJUIÛDBSF�ÛÛÛ 5IFÛPMEFTUÛUFYUT ÛXIJDIÛBSFÛVTFEÛUPÛFTUBCMJTIÛUIFÛFBSMJFTUÛTUBHFÛPGÛUIFÛ MBOHVBHF Û BSFÛ GJSTUÛ PGÛ BMMÛ UIFÛ #JCMFÛ USBOTMBUJPO�Û BOEÛ UIFÛ �"HBJOTUÛ)FSFTJFT�Û & D� B BOEPD Û CZÛ &[OJLÛ PGÛ ,PMC�Û 5IFOÛ UIFÛ �-JGFÛ PGÛ.FTSPQ�Û CZÛ,PSJXO Û�5IFÛ)JTUPSZÛPGÛ"SNFOJB�Û1BUNVU JXO�)BZPD ÛCZÛBOÛBVUIPSÛXJUIÛUIFÛ QTFVEPOZNÛ"HBUIBOHFMPT Û BOEÛ UIFÛ �)JTUPSZÛ PGÛ "SNFOJB�Û CZÛ.PTFTÛGSPNÛ$IPSFO�Û5IFÛGSBHNFOUTÛPGÛPMEÛQPFUSZÛDJUFEÛCZÛ.PTFTÛUIFÛ�4POHTÛPGÛ(P UO� Û(P U BO� FSHL Û UFMMÛ BCPVUÛ QSF�$ISJTUJBOÛ UJNFTÛ CVUÛ BSFÛ OPUÛ BSDIBJDÛMJOHVJTUJDBMMZ�ÛÛÛÛÛÛÛÛ ��Û%JBMFDUTÛÛÛÛÛÛÛ BÛ0OFÛIBTÛXPOEFSFEÛJGÛUIFÛ$MBTTJDBMÛEJBMFDUÛBTÛXFÛIBWFÛJUÛTIPXTÛFMFNFOUTÛGSPNÛEJGGFSFOUÛEJBMFDUT�Û*OÛGBDUÛPOMZÛPOFÛUIJOHÛTFFNTÛUPÛQPJOUÛUPÛUIJT�ÛUIFÛXPSEÛU BSuBNJNÛ�ESZ�ÛCFTJEFÛU B BNJNÛXIJDIÛIBTÛFYBDUMZÛUIFÛTBNFÛNFBOJOH�Û 5IFÛ EFWFMPQNFOUÛ PGÛ �STÛ JTÛ JOÛ BMMÛ GVSUIFSÛ JOTUBODFTÛ Û BOEÛ OPÛPUIFSÛFYQMBOBUJPOÛPGÛUIFÛPUIFSÛGPSNÛIBTÛCFFOÛGPVOE�ÛÛ 0UIFSÛWBSJBOUTÛNPTUMZÛDPODFSOÛUIFÛQSFTFODFÛWFSTVTÛBCTFODFÛPGÛI� ÛBTÛJOÛIBSCFOBMÛ�CFÛESVOL� ÛGSPNÛUIFÛ*&ÛSPPUÛ�TFSCI��Û5IFÛWBSJBUJPOÛNBZÛCFÛFYQMBJOFEÛ UISPVHIÛ UIFÛ EFWFMPQNFOUÛ PGÛ EJGGFSFOUÛ BCMBVUÛ HSBEFT Û GVMMÛHSBEFTÛ�TF�PSCI�ÛMPTJOHÛUIFÛ�T�ÛXJUIPVUÛUSBDF ÛXIJMFÛUIFÛGVMMÛHSBEFÛ�TSF�PCI�ÛBOEÛQFSIBQTÛUIFÛ[FSPÛHSBEFÛLFQUÛUIFÛI�ÛMPOHFS ÛCFGPSFÛJUÛXBTÛMPTUÛBOEÛUIFÛJOJUJBMÛS�ÛHPUÛBÛQSPUIFUJDÛWPXFMÛ,PSUMBOEUÛ��C Û��Û<�ÛUIJTÛWPM� Û��>�ÛÛ "CTFODFÛWFSTVTÛQSFTFODFÛPGÛI�Û GSPNÛBÛ MBSZOHFBMÛ JTÛ GPVOEÛF�H�Û JOÛIBZD Û �JOTQFDUJPO��Û *UÛ JTÛ QSPQPTFEÛ UIBUÛ UIFÛ GPSNTÛ EFSJWFÛ GSPNÛ �, � FJT�Û �ÛIBZ�ÛWFSTVTÛ�, � PJT�Û�ÛBZ��Û/PX ÛXIJMFÛJUÛJTÛQPTTJCMFÛUIBUÛEJGGFSFOUÛEJBMFDUTÛHFOFSBMJ[FEÛ FJUIFSÛ UIFÛ I�Û PSÛ JUTÛ BCTFODFÛ JOÛ GPSNTÛ PGÛ UIJTÛ SPPU Û JUÛ TFFNTÛRVJUFÛQSPCBCMFÛUIBUÛXFÛNBZÛGJOEÛIFTJUBUJPOÛXJUIJOÛBÛTJOHMFÛEJBMFDU�Û5IFTFÛDBTFTÛ DBOOPUÛ UIFSFGPSFÛ CFÛ VTFEÛ BTÛ FWJEFODFÛ GPSÛ EJGGFSFOUÛ EJBMFDUTÛ JOÛ$MBTTJDBMÛ"SNFOJBO�ÛÛÛ CÛ 5IFÛ TFDPOEÛ RVFTUJPOÛ UIBUÛ NBZÛ CFÛ BTLFEÛ JTÛ JGÛ MBUFSÛ NBUFSJBMÛHJWFTÛ JOGPSNBUJPOÛBCPVUÛ UIFÛFBSMZÛEFWFMPQNFOUTÛPGÛ UIFÛ MBOHVBHFÛXIJDIÛDBOOPUÛCFÛGPVOEÛJOÛUIFÛ$MBTTJDBMÛMBOHVBHF�Û5IBUÛJT ÛJTÛUIFSFÛMBUFSÛNBUFSJBMÛXIJDIÛ EPFTÛ OPUÛ HPÛ CBDLÛ POÛ UIFÛ $MBTTJDBMÛ MBOHVBHFÛ BOEÛ UIFSFGPSFÛ IBTÛSFGMFYFTÛ PGÛ BOÛ PMEFSÛ TJUVBUJPO Û "Û DMFBSÛ JOTUBODFÛ JTÛ UIFÛ GBDUÛ UIBUÛ TPNFÛEJBMFDUTÛIBWFÛOPNJOBUJWFÛ GPSNTÛ JOÛ �OÛ UIBUÛHPÛCBDLÛ UPÛ UIFÛPMEÛBDDVTBUJWFÛMJLFÛ PUOÛ �GPPU�Û JOÛ UIFÛ$MBTTJDBMÛ MBOHVBHFÛXIFSFÛ BÛEJBMFDUÛEPFTÛOPUÛIBWFÛ


UIJTÛ�OÛBOEÛWJDF�WFSTB�Û5IVT ÛCFTJEFÛ$MBTT�ÛBTU Û�TUBS�ÛPOFÛEJBMFDUÛIBTÛBTUF O ÛXIJDIÛ DPOUJOVFTÛ �, � TUFM�N�Û *UÛ JTÛ DMFBSÛ UIBUÛ UIJTÛ GPSNÛ DBOOPUÛ IBWFÛ CFFOÛUBLFOÛGPSNÛUIFÛ$MBTTJDBMÛEJBMFDU�Û JUÛNVTUÛIBWFÛCFFOÛTFMFDUFEÛBUÛBOÛFBSMJFSÛTUBHFÛDG�Û,PSUMBOEUÛ��D Û��Û<�ÛUIJTÛWPM� Û��>�Û/PUFÛUIBUÛXFÛBSFÛIFSFÛDPODFSOFEÛXJUIÛNPSQIPMPHZ�Û$G�ÛBMTPÛUIFÛXPSEÛGPSÛ�NJML�ÛJOÛ��ÛÛ "ÛRVJUFÛEJGGFSFOUÛ GBDUÛ JTÛ UIBUÛ TPNFÛEJBMFDUTÛEJTUJOHVJTIÛ CFUXFFOÛBOÛ JOJUJBMÛWPJDFEÛ)�ÛBOEÛBÛWPJDFMFTTÛPOFÛI��Û5IFÛWPJDFEÛ)�Û SFQSFTFOUTÛ1*&Û Z��Û 5IFÛ BTTVNQUJPOÛ JTÛ UIBUÛ UIFÛ $MBTTJDBMÛ MBOHVBHFÛ BMTPÛ IBEÛ UIJTÛEFWFMPQNFOUÛ BUÛ BOÛ FBSMJFSÛ TUBHF Û CVUÛ IFSFÛ UIFÛ)�ÛXBTÛ MPTU�Û 5IJTÛ JOTJHIUÛIFMQTÛDMBSJGZÛEFUBJMTÛPGÛUIFÛ$MBTTJDBMÛMBOHVBHF ÛTFFÛ��E�ÛÛ 5IFÛ NPTUÛ JNQPSUBOUÛ DPOUSJCVUJPOÛ GSPNÛ MBUFSÛ EJBMFDUTÛ JTÛ UIFÛPCTFSWBUJPOÛ UIBUÛ TPNFÛ TPVOETÛ BSFÛ HMPUUBMJ[FE Û BOEÛ UIFÛ DPODMVTJPOÛ UIBUÛUIJTÛQIFOPNFOPOÛJTÛJOIFSJUFEÛGSPNÛ1SPUP�*OEP�&VSPQFBOÛBOEÛSFGMFDUTÛUIFÛHMPUUBMJ[BUJPOÛPGÛUIFÛWPJDFEÛVOBTQJSBUFEÛTUPQT�ÛTFFÛ��ÛÛ DÛ"ÛEJGGFSFOUÛRVFTUJPOÛJTÛXIFUIFSÛXFÛDBOÛJEFOUJGZÛUIFÛQPTJUJPOÛPGÛUIFÛ $MBTTJDBMÛ MBOHVBHFÛ BNPOHÛ PUIFSÛ EJBMFDUT Û BOEÛ TPÛ EFUFSNJOFÛ JUTÛHFPHSBQIJDBMÛQPTJUJPOÛXIJDIÛJTÛOPUÛBÛMJOHVJTUJDÛRVFTUJPO�Û4PNFÛGBDUTÛBSFÛEJGGJDVMUÛUPÛFYQMBJO�Û5IVTÛ$MBTT�ÛF CBZSÛ�CSPUIFS�ÛTUBOETÛBHBJOTUÛBYQBSÛPGÛBMMÛNPEFSOÛEJBMFDUT�Û5IJTÛRVFTUJPOÛIBTÛOPUÛZFUÛCFFOÛTBUJTGBDUPSJMZÛBOTXFSFE�Û0GÛ DPVSTF Û JUÛ JTÛ RVJUFÛ QPTTJCMFÛ UIBUÛ EFUBJMTÛ PGÛ BÛ HJWFOÛ EJBMFDU Û F�H�Û UIFÛ$MBTTJDBMÛ MBOHVBHF ÛNBZÛIBWFÛCFFOÛ MPTUÛ FWFSZXIFSFÛ FMTF�Û5IJTÛ JTÛOPUÛ BOÛVOLOPXOÛQIFOPNFOPOÛ JOÛ UIFÛIJTUPSZÛPGÛ MBOHVBHFT�Û0OFÛNJHIUÛ UIJOLÛPGÛUIFÛQPTJUJPOÛPGÛ.ZDFOBFBOÛDPNQBSFEÛXJUIÛUIFÛEJBMFDUTÛLOPXOÛGSPNÛMBUFSÛUJNFT�ÛÛ 5IFÛ PWFSBMMÛ DPODMVTJPOÛ JTÛ UIBUÛ UIFÛ $MBTTJDBMÛ MBOHVBHFÛ JTÛ POFÛEJBMFDUÛ HSPVQ ÛQFSIBQTÛPGÛBÛTNBMMÛOVNCFSÛPGÛTQFBLFST Û UIBUÛ UIFSFÛXFSFÛTFWFSBMÛEJBMFDUTÛ UIPVHIÛQFSIBQTÛEJGGFSJOHÛPOMZÛPOÛBÛ MJNJUFEÛ TDBMF Û BOEÛUIBUÛ UIFÛ NPEFSOÛ EJBMFDUTÛ NBZÛ QSFTFSWFÛ JNQPSUBOUÛ EBUBÛ GPSÛ UIFÛSFDPOTUSVDUJPOÛPGÛUIFÛPMEFTUÛIJTUPSZÛPGÛUIFÛMBOHVBHF�ÛÛÛÛÛÛ ��ÛÛ5IFÛBMQIBCFUÛÛÛÛ 5IFÛ "SNFOJBOÛ BMQIBCFUÛ XBTÛ USBEJUJPOBMMZÛ EFWJTFEÛ CZÛ CJTIPQÛ.FTSPQ Û DBMMFEÛ.BvUPD Û JOÛ��Û5IFÛ(SFFLÛBMQIBCFUÛXBTÛBQQBSFOUMZÛBÛTPVSDFÛPGÛJOTQJSBUJPO ÛBTÛUIFÛPSEFSÛPGÛUIFÛMFUUFSTÛPGÛUIFÛ(SFFLÛBMQIBCFUÛJTÛSFUBJOFE Û CVUÛ UIFÛ TIBQFÛ PGÛ UIFÛ TJHOTÛ DBOÛ POMZÛ CFÛ EFSJWFEÛ GSPNÛ JUÛ XJUIÛNVDIÛ EJGGJDVMUZ�Û "MTPÛ UIFÛ MFUUFSTÛ UIBUÛ XFSFÛ BEEFEÛ BSFÛ JOTFSUFEÛ JOÛVOQSFEJDUBCMFÛ QMBDFT�Û OPÛ TZTUFNÛ IBTÛ CFFOÛ EJTDPWFSFE�Û 'PSÛ MJOHVJTUJDÛQVSQPTFTÛ JUÛ JTÛ UIFSFGPSFÛNPSFÛ FGGJDJFOUÛ UPÛ GPMMPXÛ UIFÛ PSEFSÛ PGÛ UIFÛ -BUJOÛBMQIBCFU�ÛÛÛ 5IFÛ BMQIBCFUÛ JOÛ UIFÛ USBEJUJPOBMÛ PSEFSÛ UIFÛ DPSSFTQPOEJOHÛ(SFFLÛMFUUFSTÛBSFÛHJWFOÛCFGPSFÛUIFNÛJT�Û

144� 5��6��3��%HHNHV��

A��x��B��B��zÛÛÛÛÛC��G��|��H�D��~��E�E��F�Z��[�H�� s�� ÛÛÛÛÛ��U �s�� I��J�s��M�s��Y�

s��D�K��L�s��I�s��K�L�� s��

M��N��

s�� Z��N��¢��O��X��¤��u��O��¦��P��s��¨�� P��ª��Q��

s��¬�� R��®�� S��°��T��s��²��W�T��´��U�s��¶��S��s��¸��D �U��º��X��F��¼��Q �C��¾��L �s��À�� s��Â��G��ÛÛ *OÛUIFÛUXFMGUIÛDFOUVSZÛUIFÛTJHOÛÀÛXBTÛBEEFE ÛUSBOTDSJCFEÛ ��5IJTÛTPVOEÛXBTÛ UIFÛSFTVMUÛPGÛDPOUSBDUJPOÛPGÛ UIFÛPMEFS Û$MBTTJDBMÛEJQIUPOHÛBX�Û$G�Û �Û"MTPÛBÛTJHOÛGPSÛGÛXBTÛBEEFE ÛXIJDIÛPOMZÛPDDVSTÛJOÛGPSFJHOÛOBNFT�ÛÛ 5IFÛ "SNFOJBOÛ BMQIBCFUÛ JTÛ BMNPTUÛ DPNQMFUFMZÛ QIPOFNJD Û J�F�ÛFWFSZÛTJHOÛJOEJDBUFTÛBÛTJOHMFÛQIPOFNF�Û0OMZÛWÛJTÛBOÛBMMPQIPOFÛPGÛX�Û5IFÛQIPOPMPHJDBMÛTUBUVTÛPGÛ Û ÛBOEÛ ÛSFRVJSFTÛEJTDVTTJPOÛTFFÛ��Û5IFÛPUIFSÛFYDFQUJPOÛ JTÛ UIBUÛ �V�Û JTÛXSJUUFOÛXJUIÛ UXPÛTJHOT ÛPX�Û5IJTÛXBTÛ UBLFOÛPWFSÛGSPNÛ(SFFL ÛXIFSFÛPVÛDBNFÛUPÛJOEJDBUFÛ�V�ÛJOÛUIFÛ�UIÛDFOUVSZÛ#�$�ÛBMSFBEZÛ*OÛBÛMJOHVJTUJDÛDPOUFYUÛUIFÛOPUBUJPOÛPXÛTIPVMEÛOPÛMPOHFSÛCFÛVTFE�ÛJUÛXPVMEÛCFÛ BXLXBSEÛ UPÛ SFUBJOÛ UIFÛ POMZÛ �NJTUBLF�Û JOÛ UIFÛ "SNFOJBOÛ BMQIBCFUJDÛTZTUFNÛ XIJDIÛIBTÛ DBVTFEÛBOPUIFSÛQSPCMFN�Û UIFÛVTFÛPGÛWÛ BGUFSÛPÛ JGÛ <PX> ÛOPUÛ <V> Û JTÛNFBOU�Û5IFÛWBMVFÛPGÛ UIFÛ TJHOTÛ JTÛHJWFOÛXIFOÛXFÛEJTDVTTÛ UIFÛQIPOFNJDÛTZTUFN�ÛÛ &BSMJFSÛTPNFÛTJHOTÛXFSFÛUSBOTDSJCFEÛEJGGFSFOUMZ�ÛÛÛÛÛÛDÛÛÛXBTÛVTFEÛGPSÛÛ D �

�� ÛÛÛÛÛÛÛÛÛÛÛ ÛÛÛ ��S�ÛÛ�ÛÛÛÛÛÛÛ ÛÛ �ÛÛÛÛÛÛÛÛÛÛÃ��ÛÛÛÛÛÛÛ Û ÛÛÛÛÛPÛÛÛÛÛÛÛÛÛÛ Û ÛÛÛ5IFÛMBUUFSÛUXPÛBSFÛTUJMMÛTPNFUJNFTÛVTFE ÛQSFTVNBCMZÛCFDBVTFÛUIFÛTJHOÛJTÛNPSFÛSFBEJMZÛBWBJMBCMF�ÛÛ ÛÛ ��ÛÛ5IFÛQIPOFNJDÛTZTUFNÛÛÛ 5IFÛQIPOFNJDÛTZTUFNÛDBOÛCFÛQSFTFOUFEÛBTÛGPMMPXT�ÛÛÛÛ


ÛÛTUPQTÛÛÛÛÛÛMBCJBMTÛÛÛ Û Q�� Q �� C��ÛÛEFOUBMT� � U� U � E��ÛÛWFMBSTÛ Û L� L � H��ÛÛÛÛÛÛBGGSJDBUFTÛ�� D� D � K�� ÛÛÛGSJDBUJWFTÛ Û T� �� [�� u� � �ÛÛÛÛÛÛÛÛÛÛÛÛÛ Û Û YÛ I��ÛSFTPOBOUTÛÛÛÛÛMJRVJETÛ Û S� �� M� � ��ÛÛÛOBTBMTÛÛ Û O� N��ÛÛÛÛÛTFNJWPXFMTÛ Û Z� � X��ÛÛÛÛWPXFMTÛ Û J�� V�� F� Û P�� BÛÛÛÛÛ5IFSFÛBSFÛUIFÛGPMMPXJOHÛEJQIUPOHT�Û BZÛ Û PZ�� BXÛ FX� � JX�� FBÛÛÛÛÛ5IFSFÛJTÛBÛUSJQIUPOH�Û Û Û Û FBZÛÛÛÛÛ 5IFÛ WBMVFÛ PGÛ UIFÛ BCPWFÛ TJHOTÛ XJMMÛ JOÛ NPTUÛ DBTFTÛ CFÛ DMFBSÛ GSPNÛUIFJSÛQPTJUJPOÛJOÛUIFÛQIPOFNJDÛTZTUFN�Û ÛJOEJDBUFTÛBTQJSBUJPO�ÛÛÛ -FTTÛDMFBSÛBSFÛ �BOEÛ �Û5IFÛ ÛXBTÛDBMMFEÛBÛ �EPVCMFÛS�ÛCVUÛJUÛXBTÛBÛTJNQMFÛ QIPOFNF�Û JUÛ XBTÛ TUSPOHMZÛ SPMMFEÛ BTÛ JOÛ 4QBOJTI�Û *UÛ PDDVSSFEÛCFGPSFÛO ÛXIFSFÛSÛ JTÛOPUÛ GPVOE�Û)PXFWFS Û UIFÛ �XBTÛ MBUFSÛHFOFSBMJ[FEÛ JOÛPUIFSÛQPTJUJPOTÛJOÛQBSBEJHNT�ÛDG�ÛKF OÛ�IBOE� ÛHFO�ÛKF JO��"MTPÛS�DBNFÛCFGPSFÛOÛUISPVHIÛHFOFSBMJ[BUJPO�ÛWFSJOÛ�VQQFS ÛTVQSFNF� ÛHFO�ÛWFSOPZ�Û5IVTÛUIFTFÛTPVOETÛBSFÛQIPOFNFT�ÛÛÛ 5IFÛ �JTÛUSBEJUJPOBMMZÛQSPOPVODFEÛBTÛ< >�Û1SPCBCMZÛJUÛXBTÛBÛ�EBSL� ÛWFMBSÛ M� JOÛPSJHJO�Û *UÛ TFFNTÛ UPÛIBWFÛCFFOÛ JOÛPSJHJOÛBÛWBSJBOUÛPGÛ M�� Û TUBOETÛFTQFDJBMMZÛ CFGPSFÛ DPOTPOBOUÛ XIJMFÛ MÛ JTÛ GPVOEÛ JOÛ BOMBVUÛ BOEÛ BGUFSÛ Z�Û *OÛPUIFSÛ QPTJUJPOT Û IPXFWFS Û UIFSFÛ JTÛ BOÛ PQQPTJUJPOÛ CFUXFFOÛ UIFNÛ HPMÛ�IFBU�Û�ÛHP Û�UIJFG��ÛÛ� W� JTÛBOÛBMMPQIPOFÛPGÛX�Û *UÛTUBOETÛJOÛBOMBVUÛBMTPÛPGÛDPNQPVOET ÛDG�ÛWBTOÛ �CFDBVTF��Û'VSUIFSÛ JUÛ JTÛ GPVOEÛBGUFSÛP ÛDG�ÛIPWJXÛ �TIFQIFSE��ÛDG�Û UIFÛJOTUS�Û FOEJOHÛ �BX Û ÞJXÛ XJUIÛ B�Û BOEÛ J�TUFNTÛ CVUÛ -PWÛ XJUIÛ P�TUFNT�Û 5IJTÛ JTÛEPOFÛCFDBVTFÛPXÛXBTÛVTFEÛGPSÛ�V��Û4PÛUIJTÛJTÛPOMZÛBÛHSBQIJDÛNBUUFS�Û5IFÛWBMVFÛ PGÛXÛ JTÛ OPUÛ DFSUBJO�Û 5IFÛ USBEJUJPOBMÛ QSPOVODJBUJPOÛ JTÛ <W> Û CVUÛ UIJTÛ

146� 5��6��3��%HHNHV��

NBZÛOPUÛCFÛPME�Û*UÛDPVMEÛIBWFÛCFFOÛBÛCJMBCJBMÛGSJDBUJWF ÛPSÛBÛTFNJWPXFM�ÛÛÛ "SNFOJBOÛEJEÛOPUÛIBWFÛHFNJOBUFT�Û'PSNTÛMJLFÛFSSPSEÛ�UIJSE� �IBWFÛFFOÛ �CFUXFFOÛÛUIFÛUXPÛS�T�Û<ZFS SPSE>�ÛÛ 5IFÛ Û EFSJWFTÛ GSPNÛ 1*&Û FJ Û CVUÛ JUÛ JTÛ OPUÛ EJTUJODUJWFMZÛ MPOH Û BTÛ"SNFOJBOÛEPFTÛOPUÛIBWFÛBOÛPQQPTJUJPOÛCFUXFFOÛMPOHÛBOEÛTIPSUÛWPXFMT�Û*UTÛWBMVFÛJTÛJOÛGBDUÛ<FJ>�Û�Û 5IFSFÛ BSFÛ TFWFSBMÛ EJGGJDVMUJFTÛ XJUIÛ UIJTÛ TPVOE Û UIPVHIÛ JOÛQSBDUJDFÛUIFTFÛBSFÛOPUÛTPÛTFSJPVT�ÛsÛ5IFÛTPVOEÛBSPTFÛUISPVHIÛSFEVDUJPOÛPGÛJÛBOEÛVÛJOÛQSFUPOJDÛQPTJUJPO ÛJ�F�ÛJOÛBMMÛTZMMBCMFTÛFYDFQUÛUIFÛMBTU�ÛUIJTÛTPVOEÛIBTÛ QIPOFNJDÛ WBMVF�Û )PXFWFS Û UIFÛ TPVOEÛ BMTPÛ PDDVSTÛ BVUPNBUJDBMMZÛCFUXFFOÛDPOTPOBOUT�Û)FSFÛ JUÛ JTÛPGÛDPVSTFÛOPO�QIPOFNJD ÛCVUÛJUTÛVTFÛXBTÛFYUFOEFEÛJOÛDFSUBJOÛNPSQIPMPHJDBMÛDBUFHPSJFT�Û5ISPVHIÛUIJTÛEFWFMPQNFOUÛJUÛCFDBNFÛQIPOFNJDÛPOÛBÛNVDIÛXJEFSÛTDBMF ÛCVUÛUIJTÛEPFTÛOPUÛDPOTJEFSÛVTÛJOÛBÛIJTUPSJDBMÛEJTDVTTJPO�ÛÛ 5IFÛTFDPOEÛQSPCMFNÛJTÛUIBUÛUIFÛTPVOEÛJTÛOPUÛOPSNBMMZÛJOEJDBUFEÛJOÛXSJUJOH�Û0OÛUIJTÛQPJOU ÛUIFÛXSJUJOHÛTZTUFNÛJTÛEFGJDJFOU�Û*UÛJTÛXSJUUFOÛPOMZÛJOJUJBMMZ ÛCFGPSFÛDMVTUFSTÛCFHJOOJOHÛXJUIÛN �O � ÛBOEÛJOÛUIFÛQSFQPTJUJPOÛ TUÛ�BDDPSEJOH ÛBGUFS�Û OEÛ�BGUFS ÛVOEFS ÛXJUI�ÛBHSFFTÛXJUIÛUIFÛSVMF�ÛF�H�Û NQFNÛ�*Û ESJOL� Û OUJSÛ �TFMFDU��Û 5IFÛ GFXÛ PGÛ UIFTFÛXPSETÛ UIBUÛ IBWFÛ BOÛ FUZNPMPHZÛQPJOUÛ UPÛBOÛPSJHJOBMÛ GVMMÛWPXFM�Û OEFSL Û �JOUFTUJOFT�Û �Û(S�Û|NTERA Û OE� �Û(S�Û�NTI� Û TUÛ�Û�QPTU� Û K�BNÛ �UPÛMPOHÛGPS�ÛCFTJEFÛ J KÛ �EFTJSF��Û4PÛIFSFÛXFÛIBWFÛBÛOPSNBMÛWPXFMÛUIBUÛXBTÛSFEVDFEÛJOÛQSFUPOJDÛQPTJUJPO�ÛUIFÛ ÛJTÛBÛQIPOFNFÛIFSF�Û*UÛJTÛVOLOPXOÛXIZÛ ÛJTÛXSJUUFOÛPOMZÛIFSF�ÛÛ0OMZÛSBSFMZÛJTÛ ÛXSJUUFOÛJOÛPUIFSÛQPTJUJPOT�ÛTFFÛ��ÛÛ 5IVT ÛXFÛEPÛOPUÛLOPXÛXJUIÛDFSUBJOUZÛXIFSFÛFMTFÛ ÛPDDVSTÛJOÛUIFÛ$MBTTJDBMÛ MBOHVBHF�Û 5IFÛ HFOFSBMÛ BTTVNQUJPOÛ JTÛ UIBUÛ JUÛ PDDVSSFEÛ JOÛ UIFÛTBNFÛ QMBDFTÛ BTÛ JOÛ NPEFSOÛ "SNFOJBO�Û Û *OÛ NBOZÛ DBTFTÛ UIFÛ TPVOEÛ JTÛBVUPNBUJD�Û5IFÛNBJOÛ SVMFTÛ TFFNÛ UPÛ CF�Û BOÛ FWFSZÛ JOJUJBMÛ DMVTUFSÛIBTÛ UIFÛWPXFMÛCFUXFFOÛUIFÛGJSTUÛBOEÛUIFÛTFDPOEÛDPOTPOBOUÛHSFN� �XSJUF��<H SFN> ÛFYDFQUÛ CFGPSFÛXPSEÛ JOJUJBMÛTÛ �Û DPOTPOBOUÛ TQJUBL� �XIJUF�Û < TQJUBL>�Û'VSIFSÛSVMFT ÛXIJDIÛBSFÛOPUÛUIFÛSFTVMUÛPGÛIJTUPSJDBMÛFWFOUT ÛEPÛOPUÛDPODFSOÛVT�ÛÛ 5IBUÛ UIJTÛ TZTUFNÛ FYJTUFEÛ JOÛ $MBTTJDBMÛ "SNFOJBOÛ BMSFBEZÛ DBOÛTPNFUJNFTÛCFÛEFNPOTUSBUFE�Û0OFÛJTÛUIFÛSFOEFSJOHÛPGÛ"SNFOJBOÛOBNFTÛJOÛPUIFSÛMBOHVBHFT�ÛÛ'VSUIFS ÛNPOPTZMMBCJDÛBPSJTUÛGPSNTÛHPUÛUIFÛBVHNFOU ÛTPÛLMJÛ �*Û TXBMMPXFE�ÛXBTÛ <L MJ> Û TQBOÛ �IFÛ LJMMFE�Û XBTÛ < TQBO> Û PUIFSXJTFÛ UIFZÛXPVMEÛIBWFÛHPUÛF��ÛÛÛÛ *OÛTPNFÛDBTFTÛPOFÛDBOOPUÛCFÛTVSF ÛJGÛPOFÛIBTÛBÛXSJUUFOÛUFYUÛPOMZ�Û5IVT Û XFÛ IBWFÛ TUBOBNÛ < TUBOBN>Û CVUÛ TUPZ Û UIFÛ HFOJUJWFÛ PGÛ TVUÛ �MJF� Û JTÛQSPOPVODFEÛ<T UP>ÛUIBUÛUIFÛGJOBMÛ�ZÛJTÛOPUÛQSPOPVODFEÛJTÛJSSFMFWBOUÛIFSF�Û*GÛ UIFÛ FUZNPMPHZÛ JTÛ VOLOPXO Û BOEÛ JGÛ UIFÛXPSEÛ EPFTÛ OPÛ MPOHFSÛ FYJTUÛ JOÛNPEFSOÛ "SNFOJBO Û UIFSFÛ JTÛ BÛ QSPCMFN�Û *UÛ JT Û PGÛ DPVSTF Û BÛ QSPCMFNÛ PGÛ


TQFMMJOH ÛOPUÛBÛQSPCMFNÛPGÛUIFÛMBOHVBHF�ÛÛÛÛÛÛ ��ÛÛÛ.PEFSOÛQSPOVODJBUJPOÛÛ %FUBJMTÛ PGÛ UIFÛ QSPOVODJBUJPOÛ BSFÛ OPUÛ LOPXO�Û 8FÛ IBWFÛ SVMFTÛXIJDIÛ BSFÛNPTUMZÛ GPMMPXFEÛ BMTPÛ GPSÛ$MBTTJDBMÛ"SNFOJBO Û CVUÛ UIFTFÛ EBUFÛGSPNÛ UIFÛ ��UIÛ PSÛ ��UIÛ DFOUVSZ Û BOEÛ TPNFÛ PGÛ UIFNÛ BSFÛ OPUÛ WBMJEÛ GPSÛ UIFÛ$MBTTJDBMÛMBOHVBHF�ÛÛÛÛÛÛÛ5IFÛQSPOVODJBUJPOÛPGÛJOJUJBMÛZ�ÛBTÛÛJTÛOPUÛPME�ÛÛÛÛÛÛ8PSE�JOJUJBMÛF�ÛJTÛQSPOPVODFEÛ<ZF> ÛCVUÛUIJTÛJTÛQPTU�$MBTTJDBM�ÛÛÛÛÛ5IFÛTBNFÛIPMETÛGPSÛUIFÛQSPOVODJBUJPOÛ<XP>ÛGPSÛP��ÛÛÛÛÛPZÛCFGPSFÛDPOTPOBOUÛJTÛQSPOPVODFEÛ<VZ>�ÛUIJTÛNBZÛCFÛOPUÛ$MBTTJDBM�ÛÛÛÛÛJXÛ JTÛQSPOPVODFEÛBTÛ <ZV>�Û/PÛEPVCUÛ UIFÛPMEÛQSPOVODJBUJPOÛXBTÛ <JV>�ÛrÛ5IFÛNBOVTDSJQUTÛTPNFUJNFTÛXSJUFÛFX�ÛÛÛÛÛÛFBÛ JTÛ QSPOPVODFEÛ BTÛ <ZB>�Û UIJTÛ JTÛ FBTJMZÛ VOEFSTUPPEÛ BTÛ BÛ MBUFSÛEFWFMPQNFOU�Û 5IFÛ OPUBUJPOÛ FBÛ XBTÛ OPÛ EPVCUÛ VTFEÛ CFDBVTFÛ JUÛ XBTÛ BÛEJQIUPOHÛFB��ÛÛÛ ��ÛÛ*OEP�&VSPQFBOÛÛ ��ÛÛ5IFÛQIPOFNJDÛTZTUFNÛPGÛ1SPUP�*OEP�&VSPQFBOÛÛ 5IFÛ QIPOFNJDÛ TZTUFNÛ PGÛ 1SPUP�*OEP�&VSPQFBOÛ IBTÛ CFFOÛSFDPOTUSVDUFEÛBTÛGPMMPXT�ÛÛÛÛMBCJBMTÛÛÛÛÛÛÛÛÛÛQÛ��CÛÛÛÛÛÛCI ÛÛÛEFOUBMTÛÛÛÛÛÛÛÛÛU��E��EI��ÛÛÛQBMBUBMTÛÛÛÛÛÛÛÛ �� IÛÛÛWFMBSTÛÛÛÛÛÛÛÛÛÛL��H��HIÛÛÛMBCJPWFMBSTÛÛÛL ��H ��H IÛÛÛÛTJCJMBOUÛÛÛ T��ÛÛÛÛMBSZOHFBMTÛÛÛÛ, � ��, � ��, � ��MJRVJETÛÛ��S��M��OBTBMT��N��OÛÛÛÛTFNJWPXFMTÛÛÛÛÛÛJ��VÛÛÛÛÛWPXFMTÛÛÛÛÛÛÛÛÛÛÛÛÛÛF�ÛÛ��PÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ ��Û�� Û Û5IFÛ SFTPOBOUTÛ IBWFÛ DPOTPOBOUBMÛ BOEÛ WPDBMJDÛ BMMPQIPOFT�Û UIFÛ MBUUFSÛ BSFÛJOEJDBUFEÛBTÛ �� JOÛUIFÛDBTFÛPGÛUIFÛGJSTUÛGPVS ÛUIFÛDPOTPOBOUBMÛPOFTÛPGÛJ��V�ÛBTÛZ��X��"TÛUIFÛBMMPQIPOFTÛGPSNÛPOFÛQIPOFNF ÛJ�F�ÛBTÛBMMPQIPOFTÛBSFÛBVUPNBUJD Û JUÛ JTÛ OPUÛ OFDFTTBSZÛ UPÛ JOEJDBUFÛ XIJDIÛ BMMPQIPOFÛ JTÛ VOEFSÛEJTDVTTJPO�Û*ÛUIFSFGPSFÛTIBMMÛOPUÛJOEJDBUFÛUIFÛBMMPQIPOFT ÛFYDFQUÛXIFOÛUIFÛDPOUFYUÛ SFRVJSFTÛ JU�Û Û rÛ 5IFÛ SFTPOBOUTÛ BSFÛ WPDBMJDÛ CFUXFFOÛ DPOTPOBOUT�Û

148� 5��6��3��%HHNHV��

)PXFWFS Û UIFSFÛ JTÛ BÛ QSPCMFNÛXIFOÛ XFÛ IBWFÛ UXPÛ PSÛ NPSFÛ DPOTFDVUJWFÛSFTPOBOUT ÛBTÛFBDIÛPGÛUIFNÛDBOÛCFÛWPDBMJDÛPSÛDPOTPOBOUBM ÛEFQFOEJOHÛPOÛUIFÛ GVSUIFSÛ FOWJSPONFOU�Û Û &�H�Û �SJ�ÛNBZÛCFÛ <SJ>ÛPSÛ < Z>�Û8IBUÛIBQQFOTÛ JOÛUIFTFÛ DBTFTÛ EFQFOETÛ POÛ UIFÛ MBOHVBHFÛ BOEÛ UIFÛ UJNFÛ DPODFSOFE�Û 5IFÛTJUVBUJPOÛCFDPNFTÛFWFOÛNPSFÛEJGGJDVMUÛJGÛUIFSFÛBSFÛBEKPJOJOHÛMBSZOHFBMT ÛBTÛUIFTFÛ DBOÛ CFDPNFÛ WPXFMTÛ JOÛ DFSUBJOÛ QPTJUJPOT�Û8FÛ BSFÛ OPUÛ BMXBZTÛ TVSFÛXIBUÛIBQQFOFE ÛCVUÛXFÛTIBMMÛIBSEMZÛCFÛDPODFSOFEÛXJUIÛ UIFTFÛQSPCMFNTÛIFSF�ÛÛÛ ��ÛÛÛ/PUFÛPOÛUIFÛMBSZOHFBMTÛÛ 5IFÛ NBJOÛ GBDUTÛ BCPVUÛ UIFÛ MBSZOHFBMTÛ DBOÛ CFÛ TVNNBSJ[FEÛ BTÛGPMMPXT�ÛÛÛÛ��Û1*&ÛIBEÛUISFFÛMBSZOHFBMT�ÛÛÛÛÛ, � ��, � ��, � ��ÛÛÛÛ4JOHMFÛ) ÛXJUIPVUÛJOEFY ÛJTÛVTFEÛUPÛJOEJDBUFÛBOZÛMBSZOHFBM ÛPSÛXIFOÛJUÛJTÛOPUÛLOPXOÛXIJDIÛMBSZOHFBMÛJTÛDPODFSOFE�Û5IFÛPMEFSÛOPUBUJPOÛJTÛ � ÛFUD�ÛÛÛÛ��Û *OÛ )JUUJUF Û , � �� , � Û CFGPSFÛ �FÛ BSFÛ SFQSFTFOUFEÛ BTÛ I�Û CFGPSFÛ �PÛ UIFZÛEJTBQQFBS�Û, � � BMXBZTÛ HJWFTÛ [FSP�Û 5IFÛ EFWFMPQNFOUÛ JOÛ "SNFOJBOÛ JTÛ UIFÛTBNFÛBTÛJOÛ)JUUJUF�ÛÛÛÛ��Û"OÛBEKBDFOUÛFÛJTÛDPMPVSFEÛUPÛBÛCZÛ, � ÛUPÛPÛCZÛ, � �ÛÛÛÛ��Û, � F��F��, � F��BÛÛ1*&ÛEJEÛOPUÛIBWFÛBÛQIPOFNFÛB��, � F��P��)P��P�ÛJ�F�ÛBOÛPÛXBTÛOFWFSÛDIBOHFEÛÛÛÛ��ÛF, � F��BB��F, � P�"�BPÛFUD�ÛÛÛÛ#PUIÛTFRVFODFTÛ, � FJ�BOEÛF, � JÛSFTVMUFEÛJOÛBJÛ�ÛÛÛÛÛÛÛÛF, � J��BJÛ��, � FJ��BJ�ÛÛÛÛÛÛÛÛÛ*OÛUIFÛTBNFÛXBZÛBV�PSJHJOBUFE�ÛÛÛÛÛÛ1*&ÛEJEÛOPUÛIBWFÛEJQIUIPOHTÛBJ ÛBV�ÛÛÛÛ��ÛF, � $Û�Û $ÛÛÛÛÛÛÛÛF, � $�"� $Û�1*&ÛEJEÛOPUÛIBWFÛBÛQIPOFNFÛB�Û��F, � $�� 'ÛÛÛÛÛÛÛÛP)$Û�� $ÛÛÛÛÛ��ÛJ)$ �V)$�� $ � $Û1*&ÛEJEÛOPUÛIBWFÛQIPOFNFTÛ � �ÛÛÛÛ��Û', � $��(S�Û$F$��', � $��(S�Û$B$��', � $��(S�Û$P$ÛÛÛÛÛÛÛÛ*OÛUIFÛPUIFSÛMBOHVBHFT ÛUIFÛUISFFÛMBSZOHFBMTÛIBEÛUIFÛTBNFÛSFGMFY�Û**S�ÛJÛPSÛ[FSP Û#BMUP�4MBWJDÛ[FSP ÛPUIFSÛMBOHVBHFTÛBÛPSÛ[FSP�Û"SNFOJBOÛIBT�B�ÛÛ


ÛÛÛ��ÛÛ, � �, � �, � '��(S�ÛF�B�P$� �"SN�ÛF�B�P$� Û1ISZHJBOÛBMTP ÛPUIFSÛMBOHVBHFTÛ[FSP�ÛÛÛÛ��Û ,'�ÛÛÛÛÛÛÛÛÛÛ , � $��Û(S�ÛS $�� , � $��Û(S�ÛS $�� , � $��Û(S�ÛS $ÛÛÛÛÛÛÛÛÛ*OÛUIFÛPUIFSÛMBOHVBHFTÛUIFÛUISFFÛMBSZOHFBMTÛIBEÛUIFÛTBNFÛSFGMFY�Û4LU�ÛS� S�BGUFSÛMBCJBM Û"W�ÛBS �*U��$M�ÛS Û(N�ÛVS Û#4Û�Û� ÛCVUÛXJUIÛBDVUFÛBDDFOU�Û"SN�ÛBSB�ÛÛÛÛÛS �M �O �N�CFIBWFÛQBSBMMFM ÛCVUÛ ,'��**S�Û �ÛÛÛÛ�� ,:��Û4LU�ÛJS�VS �"W�ÛBS �"SN�ÛBS �(S�ÛBS Û(N�ÛVS��#4Û�JS��VSÛÛÛÛÛÛÛÛÛÛÛS �M �O �N�CFIBWFÛQBSBMMFMÛCVUÛ**S�ÛBO �BN�ÛÛÛÛ��ÛQ)�U)�L)��4LU�ÛQI ÛUI�LIÛ�1*&ÛEJEÛOPUÛIBWFÛWPJDFMFTTÛBTQJSBUFT�ÛÛÛÛÛÛÛÛÛÛC�E�H�H ��Û)��4LU�ÛCI�EI�HIÛ�ÛÛÛ ��Û"SNFOJBOÛBOEÛ*OEP�&VSPQFBOÛÛ ��Û'SPNÛ1*&ÛUPÛ"SNFOJBO�ÛBÛTVSWFZÛÛÛÛÛÛ1*&Û ÛÛÛÛÛÛ"SNFOJBOÛÛÛÛÛQ� � I� Û[FSP�ÛX��I��Û[FSPÛBOMBVUÛCFGPSFÛP�ÛXÛBGUFSÛWPXFMÛÛ��C� � Q��CI� � C �XÛÛÛÛÛÛÛÛÛÛÛÛÛÛXÛCFUXFFOÛWPXFMT��ÛÛÛ��U� �U Û U � E �X � Z Û [FSP� Û UÛ BGUFSÛ TJCJMBOU�ÛEÛ BGUFSÛ3�ÛXÛ CFGPSFÛ3ÛBOEÛ CFUXFFOÛ CBDL�WPXFMT�Û ZÛ CFUXFFOÛ WPXFMT�Û [FSPÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛBOMBVUÛCFGPSFÛDPOTPOBOU�Û�OU��OÛÛÛÛÛE� � U��EI� � E��ÛÛÛ � � T �X Û[FSPÛÛÛÛÛÛÛÛÛÛXÛCFGPSFÛS�Û[FSPÛBOMBVUÛCFGPSFÛDPOTPOBOU�� D�� I� � K �[�� [�JOUFSWPDBMJDÛÛÛÛÛL �Û ÛL � H � T Û [FSPÛ HÛ BGUFSÛ 3�Û TÛ BGUFSÛ V�Û [FSPÛ BOMBVUÛ CFGPSFÛDPOTPOBOUÛÛÛÛÛH �Û Û L��H I� � HÛÛÛ��L ��H ��H IÛÛÛ �� ÛÛÛ Û �ÛÛÛÛÛÛÛÛÛÛ CFGPSFÛF � �J � ÛÛ ÛJOUFSWPDBMJDÛÛÛÛÛT� �T �I �[FSP Û�L Û TÛ CFGPSFÛ TUPQTÛ BOEÛ Y�Û I�Û CFGPSFÛ J�Û [FSPÛÛÛFMTFXIFSFÛÛÛÛÛÛÛS� � S � Û Û �CFGPSFÛOÛÛÛÛÛMÛ Û M � ÛÛ Û ÛCFGPSFÛDPOTPOBOUÛÛÛÛÛOÛ Û O Û[FSPÛ Û [FSPÛCFGPSFÛT�ÛÛÛÛÛÛNÛ Û N ÛX Û[FSPÛ XÛCFGPSFÛO ÛCFGPSF�BGUFSÛV�Û[FSPÛCFGPSFÛTÛ

150� 5��6��3��%HHNHV��

ÛÛÛÛ ��M�� ÛÛÛÛÛ BSÛÛÛBM��BO��BNÛÛÛÛÛÛZ� � [FSP Û ÛZ ÛI� �BGUFSÛ3�ÛZÛBGUFSÛBÛCFGPSFÛ3�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛIÛXIFOÛTUSFTTFEÛCFGPSFÛJ�ÛÛÛÛÛÛXÛ Û H ��X Û[FSP� [FSPÛCFGPSFÛS ÛCFGPSF�BGUFSÛVÛÛÛÛÛJ��V� � J��V��F��P� � F J�BÛÛÛP V�BÛÛÛÛÛJ �V�CFGPSFÛOBTBM�ÛBÛTFFÛUIFÛUFYUÛÛÛÛÛ �� J��V��FJ��PJ� � �� ÛÛÛFV��PV�� PZ��PZ�ÛÛÛ, � FJ��, � FV� BZ��BXÛÛÛÛÛÛ JÛÛÛ Û J�ÛÛ-BSZOHFBMTÛÛÛ3Û�ÛS �M �O �N�Û8Û�ÛJ�Z �V�X�ÛÛsÛ�ÛOPÛFWJEFODFÛÛÛÛÛ)��)7�ÛÛ , � �F �, � F��IB �IP�ÛPUIFSXJTFÛ)Û�Û[FSP�� )$�ÛÛÛÛÛÛ F�B�P$�� )3��ÛÛ F�B�P3ÛÛ )8�� 8��Û�)� �7)ÛÛÛÛ ÛÛÛÛÛÛÛ�� $)ÛÛ �$B �� 3)�� s�� 6)��Û ÛÛÛÛÛÛ�7)�Û �7)7�� 77�� 7)$�ÛÛ $Û�� 7)3�ÛÛ 3Û�� 7)8�� 8��ÛÛÛÛÛÛ�$)�� $)7�� L)��YÛÛÛÛÛÛÛÛÛ��ÛÛÛÛÛÛÛÛÛÛ �$)$�Û $$ �$B$$ÛÛÛÛÛ �$)3�ÛÛ s�� $)8Û sÛÛÛÛ��3)�� 3)7�Û B37ÛÛ�� 3)$�ÛÛ B3B$�� 3)3�� s�� 3)8�� s�ÛÛÛÛÛ�8)��8)7�� s��ÛÛÛÛÛÛÛÛÛÛÛÛ��8)$�Û $ÛÛ��ÛÛÛÛÛÛÛÛÛÛÛ�8)3�Û OPÛTQFDJBMÛEFWFMPQNFOU�ÛÛÛÛÛÛÛÛÛÛÛÛÛ��8)8��88�ÛÛÛÛÛ))�ÛÛÛÛÛÛÛÛÛ sÛÛÛ)3)�ÛÛÛÛÛÛ B3BÛÛÛ)8)�� sÛÛÛÛÛ


ÛÛÛÛ$MVTUFSTÛÛÛÛÛ*Û EJTDVTTÛ BTÛ DMVTUFSTÛ HSPVQTÛ PGÛ DPOTPOBOUTÛ UIBUÛ BSFÛ SFGMFDUFEÛ CZÛ POFÛQIPOFNFÛ XIJDIÛ JTÛ OPUÛ JEFOUJDBMÛ XJUIÛ POFÛ PGÛ UIFÛ DPOTPOBOUTÛ PGÛ UIFÛPSJHJOBMÛDMVTUFS�ÛCVUÛ*ÛBEEFEÛ�TQ ��TUÛBOEÛ�QTÛ UIFÛGJSTUÛBOEÛ UIFÛ MBTUÛCFJOHÛEJTQVUFE Û UIFÛ TFDPOEÛ GPSÛ UIFÛ TBLFÛ PGÛ DPIFSFODF�Û 4PÛ *Û EPÛ OPUÛ USFBUÛ BTÛ BÛDMVTUFSÛ�QO��XOÛTFFÛPOÛ�QÛPSÛ�US��S�ÛTFFÛPOÛ�U�Û*OÛDBTFÛPGÛEPVCU Û*ÛUPPLÛUIFÛHSPVQÛBTÛBÛDMVTUFS�ÛÛÛÛÛ�STÛÛ�Û ÛÛÛÛÛÛ�TZÛ�ÛZÛÛÛÛÛÛÛÛ�TXÛ�ÛL ÛÛÛÛÛÛÛÛÛ�TSÛÛ�Û �ÛÛÛÛ�TQ��TQÛÛÛÛÛÛÛ�TU��TUÛÛÛÛÛÛÛÛÛÛ�T Û�TL ��TLÛ�ÛD ÛÛÛÛÛÛÛ�QT��T�Û�QTU��TU� Û�Q �ÛÛÛÛÛÛ�EIZÛ�Û ÛÛÛÛÛÛÛ�UXÛ�ÛL ÛÛÛÛÛÛ�EXÛ�ÛLÛÛÛÛÛÛÛÛÛÛ�5 ÛÛTFFÛUFYUÛÛÛÛÛ�LZÛ�Û ÛÛÛÛÛÛÛ� XÛ�ÛuÛÛÛÛÛÛÛÛÛÛÛ� TÛ�ÛD ÛuÛÛÛÛÛ�LT �L T��D ÛuÛÛÛÛÛÛTUPQÛ�ÛMBSZOHFBMÛÛÛÛÛ��Q)��Q � ÛÛÛÛÛÛÛÛÛ�U)��U � ÛÛÛÛÛÛÛ�L)��YÛÛÛÛÛ ÛÛ ��ÛÛ'SPNÛ"SNFOJBOÛUPÛ1*&�ÛBÛTVSWFZÛÛÛÛÛ"SN�ÛÛÛÛÛ�ÛÛ 1*&ÛÛÛÛÛBÛÛÛÛÛÛÛÛÛÛ Û , � ��, � F��F, � �P�ÛÛÛÛÛÛÛÛBS�BM�BO�BNÛ Û �� Û )�� )�� )�� )ÛCFGPSFÛWPXFM�ÛÛÛÛÛÛÛÛBSB�BMB�BOB�BNBÛÛ�� )�� )�� )� )ÛÛCFGPSFÛDPOTPOBOUÛÛÛÛÛBZÛ Û , � FJ��F, � J��PJÛÛÛÛÛFÛ Û F��, � ÛÛÛÛÛ Û Û FJ��PJÛÛÛÛÛFBÛ Û J�BÛÛÛÛÛPÛ Û PÛÛ, � ÛÛÛÛÛPZÛ Û FV��PV�� ÛÛÛÛÛJÛ Û J��J)�� F, � ��F�/�ÛÛÛÛÛÛÛÛÛJBÛ Û TBÛÛÛÛÛÛVÛ Û V��V)�� F, � ��P)��P�/ÛÛJXÛCFGPSFÛWPXFM�ÛÛÛÛÛÛÛÛÛC� � CI��E� � EI��3U�ÛÛÛÛÛÛ�H� � HI��3LÛÛÛX�ÛÛÛÛÛ�K� � I�ÛÛÛÛÛÛÛ Û Û HIÛCFGPSF�F �J�ÛÛEIZÛÛ3ZÛÛÛÛ��Q� � C�ÛÛÛÛÛÛÛÛU� � EÛÛÛÛ��L� � HÛÛEXÛÛ

152� 5��6��3��%HHNHV��

ÛÛÛ�D� � ��ÛÛÛÛ� � � HZÛÛEZÛÛÛÛ��Q � � Q)� ÛÛÛÛ�U � � U��U)� �ÛÛÛÛÛÛÛ�L � � L��UX��TXÛÛÛÛÛÛD �� TL��LT�ÛÛÛÛÛÛÛ Û Û TLZÛÛÛÛ��I� � QÛÛFYDFQUÛÛCFGPSFÛP��, � ��, � ��CFGPSFÛF�ÛÛZÛCFGPSFÛJ ÛTUSFTTFE�ÛÛÛÛÛ��Y� � L)ÛÛÛÛÛ�T� � ��/TÛÛTÛCFGPSFÛU �Q�ÛÛÛÛÛÛ[� � IÛJOUFSWPDBMJDÛÛÛÛÛ�u� � X�ÛÛTL��LTÛCFGPSF�DPOTPOBOUÛÛÛÛÛ� � � HIÛCFGPSF�F �J ÛJOUFSWPDBMJDÛÛÛÛÛZÛ Û TZÛÛÛÛ�X� CIÛ JOUFSWPDBMJD�Û QÛ CFGPSFÛ DPOT��Û UÛ CFGPSFÛ S�Û [FSPÛ CFGPSFÛ ÛBGUFSÛF ÛJ�ÛÛÛÛÛÛXPÛ Û JP�VOTUSFTTFE�ÛÛÛÛÛS �M �O �N� S ÛM �O �NÛÛÛÛ� � � ST�TS�ÛSÛCFGPSFÛO�ÛÛÛÛÛÛ Û Û MÛCFGPSFÛDPOT��ÛÛÛÛ ��ÛÛ5IFÛQPTJUJPOÛPGÛ"SNFOJBOÛÛ *UÛ XBTÛ BMXBZTÛ CFMJFWFEÛ UIBUÛ "SNFOJBOÛ IBEÛ BÛ DMPTFÛ DPOOFDUJPOÛXJUIÛ (SFFL�Û 5IJTÛ XBTÛ SFDFOUMZÛ EFOJFEÛ CZÛ $MBDLTPOÛ ��Û )FÛEFNPOTUSBUFTÛ UIBUÛ UIFSFÛ BSFÛ OPÛ BHSFFNFOUTÛ JOÛ QIPOPMPHZÛ CBTJDÛEFWFMPQNFOUTÛTVDIÛBTÛ�T��IÛEPÛOPUÛQSPWFÛBOZUIJOHÛPSÛNPSQIPMPHZÛF�H�ÛUIFÛ MPDBUJWFÛ JOÛ �P Û JTÛEFSJWFEÛCZÛ,PSUMBOEU Û ��B Û TFDUJPOTÛ �Û BOEÛ�Û <UIJTÛWPM� Û��G�> Û GSPNÛBÛHFOJUJWFÛFOEJOHÛ�JPT�Û5IFSFÛSFNBJOTÛUIFÛBHSFFNFOUÛ JOÛUIFÛ WPDBCVMBSZ Û XIJDIÛ IBTÛ BMXBZTÛ CFFOÛ UIFÛ SFBMÛ BSHVNFOU�Û $MBDLTPOÛNFOUJPOTÛ ��Û JOTUBODFT Û PGÛXIJDIÛIFÛEJTDVTTFTÛ ��Û JOÛEFUBJM�Û *Û BHSFFÛ UIBUÛTFWFSBMÛBHSFFNFOUTÛDBOOPUÛCFÛNBJOUBJOFEÛXPSETÛUIBUÛIBEÛ�EX�ÛBOEÛIBWFÛFSL�ÛJOÛ"SNFOJBO�ÛTFFÛ��DÛBOEÛUIBUÛTFWFSBMÛBSFÛKVTUÛSFUBJOFEÛBSDIBJTNTÛBXSÛ �Û £MARÛ �EBZ� Û BTÛ NBZÛ CFÛ GPVOEÛ CFUXFFOÛ BOZÛ UXPÛ *OEP�&VSPQFBOÛMBOHVBHFT�Û )FÛ SFUBJOTÛ POMZÛ UISFFÛ XPSETÛ UIBUÛ PDDVSÛ JOÛ "SNFOJBOÛ BOEÛ(SFFLÛ POMZÛ TJOÛ �Û KEN²J��FNQUZ��Û Z�BXFMVNÛ �Û ±FzLLWÛ �JODSFBTF�Û BOEÛ TVUÛ �ÛYEÀDOJÛ YEºDOMAIÛ �MJF��Û UIFÛ MBTUÛ IBSEMZÛ *&�Û *OÛ NZÛ WJFXÛ UIFSFÛ JTÛ NPSF�ÛOBXU JÛ �GBTUJOH�Û �Û(S�ÛN�FWÛ TJNQMZÛQPJOUTÛ UPÛ �OF, � CI��Û FSBTUBOL Û �ÛPRWKT²JÛ�CVUUPDLT�Û DBOÛ CFÛ EFSJWFEÛ GSPNÛ �QSF, � U�Û �Û �QSP, � U�Û CVUÛNBZÛ CFÛ KVTUÛ BÛSFUBJOFEÛ BSDIBJTN�Û POÛ P CÛ �Û ±LOFºROMAIÛ �XBJM�Û TFFÛ ��Û Û .PTUÛJOUFSFTUJOHÛJTÛTJXOÛ�ÛK WNÛ �QJMMBS� ÛXIJDIÛIBTÛWFSZÛSFDFOUMZÛCFFOÛTIPXOÛUPÛCFÛ *OEP�&VSPQFBOÛ -VCPUTLZ�1SBVTU Û JOÛ -VCPUTLZ�TÛ JOBVHVSBMÛ MFDUVSFÛ


GSPNÛBÛXPSEÛNFBOJOHÛ�TIJO�CPOF��Û)FSFÛXFÛIBWFÛrÛXIBUÛXFÛIBSEMZÛGJOEÛBOZXIFSFÛFMTFÛrÛBÛSFNBSLBCMFÛBHSFFNFOUÛJOÛNFBOJOHÛGPVOEÛPOMZÛJOÛUIFTFÛUXPÛMBOHVBHFT�ÛPOÛUIFÛPUIFSÛIBOE ÛUIJTÛNBZÛPOMZÛTFFNÛSFNBSLBCMFÛUPÛVT ÛCVUÛ CFÛ BÛ OPSNBMÛ EFWFMPQNFOU�Û 'PS Û TJODFÛ JUÛ JTÛ DMFBSÛ UIBUÛ (SFFLÛ BOEÛ"SNFOJBOÛ IBEÛ OPÛ QIBTFÛ PGÛ DPNNPOÛ EFWFMPQNFOU Û XFÛ NVTUÛ MPPLÛ GPSÛJOEJDBUJPOTÛUIBUÛUIFZÛMJWFEÛJOÛFBDIÛPUIFS�TÛOFJHICPVSIPPE ÛUISPVHIÛFBSMZÛMPBOTÛ FUD�Û *Û BNÛ VOBCMFÛ UPÛ QSPEVDFÛ TVDIÛ JOTUBODFT Û CVUÛ UIJTÛ JTÛ NPTUÛEJGGJDVMU�Û #PUIÛ MBOHVBHFTÛ TFFNÛ UPÛ IBWFÛ MPBOTÛ GPSNÛ "OBUPMJBO�"FHFBOÛMBOHVBHFT ÛCVUÛUIJTÛEPFTÛOPUÛNFBOÛUIBUÛUIFZÛXFSFÛJOÛDMPTFÛDPOUBDU�Û*OÛUIJTÛQFSTQFDUJWFÛ *ÛXPVMEÛDPOTJEFSÛLBNVS Û �ÛGE�F RAÛ �CSJEHF��Û5IFÛ(SFFLÛXPSEÛPSJHJOBMMZÛ NFBOUÛ �CFBN� Û BOEÛ UIFÛ "SNFOJBOÛ XPSEÛ IBTÛ CFFOÛ DPNQBSFEÛXJUIÛ )BUUJDÛ IBNVSVXBÛ �CFBN��Û *GÛ UIFSFÛ BSFÛ BÛ HSFBUÛ OVNCFSÛ PGÛBHSFFNFOUT Û UIJTÛ NBZÛ EFQFOEÛ POÛ UIFÛ MBSHF Û PMEÛ BOEÛ XFMM�TUVEJFEÛWPDBCVMBSZÛ PGÛ (SFFL�Û *Û XPVMEÛ DPODMVEFÛ UIBUÛ BÛ MPOHFSÛ QFSJPEÛ PGÛ DMPTFÛDPOUBDUÛDBOÛOFJUIFSÛCFÛEFNPOTUSBUFEÛOPSÛFYDMVEFEÛBUÛUIFÛQSFTFOUÛTUBHF�ÛÛÛÛÛ "TÛUPÛSFMBUJPOTÛXJUIÛPUIFSÛMBOHVBHFT Û,PSUMBOEUÛ��BÛBTTVNFEÛUIBUÛ"SNFOJBOÛTIBSFEÛXJUIÛ5ISBDJBOÛUIFÛMPTTÛPGÛBTQJSBUJPOÛJOÛUIFÛWPJDFEÛBTQJSBUFTÛ UPHFUIFSÛXJUIÛ1ISZHJBO Û%BDP�"MCBOJBOÛBOEÛ#BMUP�4MBWJDÛBOEÛQFSIBQTÛ UIFÛ EFWPJDJOHÛ PGÛ UIFÛ WPJDFEÛ TUPQT�Û 8JUIÛ "MCBOJBOÛ ,PSUMBOEUÛOPUFTÛ UIFÛ GPMMPXJOHÛ BHSFFNFOUTÛ ��BÛ BOEÛ ��B�Û EFQBMBUBMJ[BUJPOÛ PGÛTUPQTÛ CFGPSFÛ SFTPOBOUT Û NPOPQIPOFNJDJ[BUJPOÛ PGÛ UIFÛ DMVTUFSTÛ �TXÛ BOEÛ� X ÛBOEÛUIFÛGBDUÛUIBUÛ, � ÛBOEÛ, � ÛCFGPSFÛFÛBSFÛSFGMFDUFEÛBTÛI��ÛÛÛ 5IJTÛ SFQSFTFOUBUJPOÛ PGÛ UIFÛ MBSZOHFBMTÛ JTÛ BDDPSEJOHÛ UPÛ ,PSUMBOEUÛBMTPÛ GPVOEÛ JOÛ )JUUJUF�Û 0GÛ DPVSTF Û WPDBMJ[BUJPOÛ PGÛ JOJUJBM Û QSFDPOTPOBOUBMÛMBSZOHFBMTÛJTÛBMTPÛGPVOEÛPOÛBÛMBSHFSÛTDBMF�Û(SFFL Û.BDFEPOJBO Û1ISZHJBOÛBOEÛ)JUUJUF�Û 5IFÛEFWFMPQNFOUÛ JTÛ BOÛ BSFBMÛ GFBUVSF ÛXIJDIÛNBZÛQBSUMZÛ CFÛEVFÛUPÛUIFÛTVCTUSBUVNÛMBOHVBHFT�ÛÛ *OÛ TVN Û UIFÛ SFMBUJPOÛXJUIÛ (SFFLÛ IBTÛ CFDPNFÛ EPVCUGVM Û GPSÛ UIBUÛXJUIÛ5ISBDJBOÛBOEÛ"MCBOJBOÛTPNFÛBSHVNFOUTÛDBOÛCFÛBEEVDFE�ÛÛÛ ��%'')28�ÛÛ "SNFOJBOÛTIPXTÛOPÛUSBDFÛPGÛUIFÛ*OEP�&VSPQFBOÛBDDFOU�ÛÛ *OÛUIFÛPMEFTUÛQIBTFÛXFÛDBOÛSFDPOTUSVDU ÛUIFÛQFOVMUJNBUFÛTZMMBCMFÛUIFÛMBTUÛCVUÛPOFÛXBTÛTUSFTTFE�Û"UÛBÛDFSUBJOÛTUBHF ÛBTÛBÛSFTVMUÛPGÛUIJTÛTUSFTTÛUIFÛ GPMMPXJOHÛ �Û MBTUÛ TZMMBCMFÛ XBTÛ MPTU�Û OPUÛ POMZÛ UIFÛ WPXFM Û CVUÛ UIFÛDPOTPOBOUTÛGPMMPXJOHÛUIFÛWPXFMÛEJTBQQFBSFEÛBTÛXFMM ÛFYDFQUÛS � M �OÛOPUFÛUIBUÛBUÛUIJTÛUJNFÛ��NÛIBEÛCFDPNFÛ�O ÛBOEÛBMTPÛ�L Û�Û��FT ÛUIFÛOPNJOBMÛQMVSBMÛFOEJOH�Û&�H�ÛÛÛIJOHÛ�GJWF�Û�Û�QFOL FÛÛÛÛFCFSÛ�IFÛCSPVHIU�ÛGSPNÛ�FCÂSFU��1*&Û�, � ÂCIFSFUÛÛÛÛIBZSÛ�GBUIFS�Û�Û�IBZJSÛ�Û�Q, � U S�Û Û

154� 5��6��3��%HHNHV��

"ÛTZMMBCJDÛSFTPOBOUÛPGÛ1*&ÛGJSTUÛEFWFMPQFEÛJOUPÛBS �BM �BO �BNÛUIFÛMBTUÛGVSUIFSÛUPÛ�BO�ÛÛÛÛFXU O��TFWFO�Û�Û�TÂQUBOÛ�Û�TFQU ÛÛ /PUFÛ UIBUÛ UIFÛ"SNFOJBOÛQSPUIFUJDÛ WPXFMÛXBTÛ BEEFEÛBGUFSÛ UIFTFÛEFWFMPQNFOUT�ÛÛÛBDD�ÛFSJTÛ�UISFF�Û�Û�USJOT��PUIFSXJTFÛUIFÛF�ÛXPVMEÛIBWFÛSFDFJWFEÛUIFÛTUSFTT ÛHJWJOHÛ�FS<O>T�ÛÛÛÛ ��ÛÛ3FUBJOFEÛGJOBMÛWPXFMTÛÛ 5IFÛWPXFMÛPGÛXIBUÛVTVBMMZÛJTÛUIFÛGJOBMÛTZMMBCMFÛDBOÛTUJMMÛCFÛTFFOÛJGÛ TPNFUIJOHÛ BQQFBSFEÛ BGUFSÛ UIBUÛ WPXFM Û CFÛ JUÛ BOÛ FOEJOHÛ JOGMFDUJPO Û BÛTFDPOEÛFMFNFOUÛPGÛBÛDPNQPVOE ÛPSÛBÛDMJUJD�ÛÛ *OGMFDUJPO�Û'PSÛUIFÛJOGMFDUJPOÛPGÛOPVOTÛDG��ÛÛÛHB O ÛQM�ÛHBSJO�L Û�MBNC�ÛÛÛ5IFÛ GPSNTÛ SFQSFTFOUÛ �VS, � � OÛ CVUÛ UIFÛ OPN�Û �OÛ DPOUJOVFTÛ UIFÛ PMEÛBDDVTBUJWF Û �VS, � �FO�FT�Û *OÛ UIFÛ TFDPOEÛ GPSNÛ UIFÛ TVGGJYÛ JTÛ QSFTFSWFEÛCFDBVTFÛ JUÛIBEÛ UIFÛ TUSFTT Û JOÛ UIFÛ GJSTUÛ GPSNÛ UIFÛ TVGGJYÛEJTBQQFBSFE�Û/PUFÛUIBUÛ OPÛ UXPÛ JEFOUJDBMÛ GPSNTÛ BSFÛ DPODFSOFEÛ IFSF Û CVUÛ POMZÛ UXPÛ DMPTFMZÛSFMBUFEÛGPSNT�ÛÛÛÛUBTOÛ�UFO�ÛCVUÛÛNFUBTBOÛ�FMFWFO�ÛÛÛ5BTOÛ�Û�EF ÛIBEÛ��BOÛ�Û�� Û5IFÛWPXFMÛBÛJTÛSFUBJOFEÛJOÛ�FMFWFO�ÛCFDBVTFÛUIJTÛJTÛBÛEFSJWBUJWFÛXJUIÛ�J��ÛUIJTÛ�J�ÛDBOÛBHBJOÛCFÛTFFOÛJOÛJOGMFDUFEÛGPSNT ÛF�H�ÛHFO�ÛNFUBTBO�J�D �Û5IFÛGJOBMÛDPOTPOBOUÛ�OÛJTÛBMXBZTÛQSFTFSWFE�ÛÛ $PNQPVOET�ÛÛÛIJOHÛ�GJWF�ÛCVUÛIOHF�UBTBOÛ�GJGUFFO�ÛÛÛÛ)FSFÛUIFÛGJOBMÛ�FÛPGÛ�QFOL FÛJTÛQSFTFSWFEÛJOÛUIFÛDPNQPVOE ÛXIJDIÛJTÛMPTUÛJOÛUIFÛJTPMBUFEÛGPSN�Û5IFÛ�J�ÛJTÛTZODPQBUFEÛCFGPSFÛUIFÛTUSFTT ÛBTÛJTÛVTVBM�ÛTFFÛ��ÛÛÛÛ PS�L ÛÛ�GPVS�ÛCVUÛ PSFL �UBTBOÛ�GPVSUFFO�ÛÛÛ)FSFÛXFÛ TFFÛ UIFÛ WPXFMÛ PGÛ UIFÛ FOEJOHÛ ��FTÛ SFBQQFBSÛ JOÛ UIFÛ DPNQPVOEÛ�GPVS�Û JTÛ BÛ QMVSBMÛ GPSN�Û /PUFÛ UIBUÛ JOÛ UIJTÛ DBTFÛ UIFÛ GJOBMÛ DPOTPOBOUÛ JTÛBMXBZTÛQSFTFSWFE�ÛÛÛ%JGGJDVMUÛ JTÛ FXU OÛ �TFWFO�Û CFTJEFÛ FXU BOBTVO�Û 1SPCBCMZÛ �BOB�Û IFSFÛSFQSFTFOUTÛBÛTQFDJBMÛEFWFMPQNFOUÛPGÛUIFÛ1*&ÛFOEJOHÛ�� ÛGPMMPXFEÛCZÛUIFÛMBSZOHFBMÛGFBUVSFÛPGÛUIFÛ�E�ÛPGÛ�E PNU, � ÛÛ�Û�TVO�Û4FFÛ��ÛÛÛVNÛEBU�ÛPGÛPWÛ �XIP �ÛCFTJEFÛVNF�L ÛEBU�ÛPGÛP�L Û �TPNFPOF��Û5IFÛWPXFMÛPGÛUIFÛEBUJWFÛ FOEJOHÛ JTÛQSFTFSWFEÛ CFGPSFÛ UIFÛQBSUJDMFÛ �L Û Û XIJDIÛ DPOUJOVFTÛ��L F�Û TFFÛ ��Û PGÛ UIFÛ JOEFGJOJUFÛ QSPOPVO�Û 5IFÛ �F�ÛNVTUÛ CFÛ UIFÛ EBUJWFÛFOEJOHÛ�� JÛ ÛDG�Û4LU�ÛUºTNBJ�Û,PSUMBOEUÛ��B Û��Û<UIJTÛWPM� Û��>�Û8FÛGJOEÛUIFÛTBNFÛJO�ÛÛÛVSFL Û �TPNFXIFSF� Û XIJDIÛ JTÛ UIFÛ MPDBUJWFÛ PGÛ UIFÛ TUFNÛ TFFOÛ JOÛ PL Û


�TPNFCPEZ� ÛHFO�ÛVSVL �ÛÛÛ*UÛTUBOETÛCFTJEFÛVSÛ�XIFSF ��ÛÛÛJ�XJ�L Û JOTUS�Û PGÛ �JL Û �TPNFUIJOH� Û QSFTFSWFTÛ UIFÛ FOEJOHÛ ��CIJÛ CFGPSFÛ UIFÛQBSUJDMFÛ�L �ÛÛÛÛÛÛ/PUFÛ UIBUÛ JOÛ JOEJWJEVBMÛ DBTFTÛ UIFÛ WPXFMÛ PGÛ UIFÛ FOEJOHÛ NBZÛ MJWFÛ POÛCFDBVTFÛ PGÛ BÛ TQFDJBMÛ EFWFMPQNFOU�Û �TVFT SÛ CFDBNFÛ �L FVS Û BGUFSÛXIJDIÛ�FV�Û CFIBWFEÛ BTÛ BÛ EJQIUPOHÛ XIJDIÛ CFDBNFÛ "SN�Û �PZ� Û BTÛ EJEÛ UIFÛ 1*&ÛEJQIUIPOHÛ�FV�ÛÛÛÛ ��Û 1SFUPOJDÛ WPXFMÛ SFEVDUJPOÛ BOEÛ UIFÛ "SNFOJBOÛ WPXFMÛBMUFSOBUJPOTÛÛ "UÛ BÛ MBUFSÛ TUBHF Û UIFÛ WPXFMÛ PGÛ BÛ QSFUPOJDÛ TZMMBCMFÛXBTÛ SFEVDFE�Û5IFÛFGGFDUTÛQFSWBEFÛUIFÛXIPMFÛTZTUFNÛPGÛUIFÛMBOHVBHF�Û5IFÛEFWFMPQNFOUTÛBSFÛBTÛGPMMPXT�ÛÛÛÛTUSFTTFEÛ J� V� � PZ� FB� FBZ� ��JXÛÛÛÛQSFUPOJDÛ � � J� V� F� F�7 �J�$� ��VÛÛÛÛÛÛ/PUÛDIBOHFEÛXFSF�ÛB �F �P��BZ �BX �FXÛBOEÛ�JX�ÛJOÛJOMBVU�ÛÛ*OJUJBMÛJ�ÛBOEÛV�ÛXFSFÛ SFTUPSFE�Û JOL OÛ �TFMG� Û HFO�Û JOL FBO Û VMÛ �LJE� Û HFO�Û VM�PZ�Û 8JUIÛNPOPTZMMBCMFTÛJOÛ�J ��VÛUIFÛWPXFMÛXBTÛSFTUPSFEÛJOÛIJBUVT�ÛKJÛ�IPSTF� ÛHFO�ÛKJ�PZ Û KJ�B�XPSÛ �SJEFS��Û 7PXFMTÛ CFGPSFÛ UIFÛ GJSTUÛ QSFUPOJDÛ TZMMBCMFÛ XFSFÛ BMTPÛSFTUPSFE�Û TJSFBMÛ �CFMPWFE�Û �Û HFO�Û TJSFMPZÛ Û OPUÛ �TSFMPZ�Û "TÛ Û BOEÛ PZÛ EFSJWFÛGSPNÛ�FJÛBOEÛ�FVÛPSÛ�PV ÛSFTQFDUJWFMZ ÛJUÛTFFNTÛUIBUÛUIFÛFÛEJTBQQFBSFE�ÛUIFÛSFEVDUJPOÛJTÛRVJUFÛEJGGFSFOUÛGSPNÛUIBUÛPGÛ�J ��V��8IFOÛ�JXÛ<JV>ÛXBTÛSFEVDFEÛUPÛV ÛUIFÛJÛCFDBNFÛ ÛKVTUÛMJLFÛUIFÛTJOHMFÛJ ÛBOEÛ XÛCFDBNFÛ<V>�Û4FFÛBMTPÛ��Û��Û/PUFÛUIBUÛJXÛCFGPSFÛ ÛXIJDIÛPSJHJOBUFEÛMBUFS ÛSFNBJOFE�ÛTFFÛ��Û5IFÛSFEVDUJPOÛPGÛFBÛBOEÛFBZÛJTÛBHBJOÛRVJUFÛEJGGFSFOU�Û&YBNQMFT�ÛÛÛÛTJSUÛ�IFBSU�Û�ÛHFO�ÛTSU�J��ÛÛÛÛÛFSLJOL Û�IFBWFO�Û�ÛFSLOJD ��ÛÛÛÛÛMVDÛ�ZPLF�Û�ÛHFO�ÛMD�PZ��ÛÛÛÛÛBOBTVO��BOJNBM�Û�ÛHFO�ÛBOBTO�PZ��ÛÛÛÛÛFNVUÛÛBPS�Û�IFÛFOUFSFE�Û�Û�ÛTH�ÛNU�J��ÛÛÛÛÛN HÛ�NJTU ÛGPH�Û�ÛHFO�ÛNJHJ �NJHBNBDÛ�NJTUZ�ÛÛÛÛ uÛ�EPOLFZ�Û�ÛHFO�ÛJu�PZ�ÛÛÛMPZTÛ�MJHIU�Û�ÛMVT�PZ �MVTBXPSÛ�MVNJOPVT�ÛÛÛÛJ PZD ÛBPS��IFÛMFUÛEJTNPVOU�Û�Û�ÛTH�ÛJ VD �J��ÛÛÛÛÛMFBSEÛ�MJWFS�Û�ÛMFSE�J��ÛÛÛÛÛNBUFBOÛ�CPPL�Û�ÛHFO�ÛNBUFO�J��ÛÛÛÛÛBUFBNÛ�*ÛIBUF�Û�ÛJNQG��ÛTH�ÛBUF�J �Û�ÛQM�ÛBUF�BL ÛÛQSFT�ÛTVCK�Û�ÛTH�ÛBUJ�D FNÛÛÛÛÛU JXÛ�OVNCFS�Û�ÛHFO�ÛU V�PZÛÛÛÛ

156� 5��6��3��%HHNHV��

Û ��:3;)07�ÛÛ ��ÛÛ1*&Û�F�ÛÛÛÛÛÛBÛ�FÛ�ÛFÛÛÛÛ�CIFS��CFSFNÛÛ�*ÛCFBS�Û�Û(S�ÛFE�RW Û-BU�ÛGFS ÛÛÛ�NF �, � ��ÛNFD��CJH�Û�Û(S�ÛME�GAJ��ÛÛÛÛÛÛÛÛ8PSETÛ UIBUÛ TFFNÛ UPÛCFHJOÛXJUIÛBÛWPXFMÛIBEÛBOÛ JOJUJBMÛ MBSZOHFBMÛBOEÛXJMMÛCFÛEJTDVTTFEÛXJUIÛUIFÛMBSZOHFBMTÛ��ÛÛÛÛÛÛÛCÛ�FÛ�ÛJÛÛCFGPSFÛOBTBMÛÛÛÛ�QFOL F�Û �IJOHÛ �GJWF�Û �Û(S�ÛPzNTE �4LU�ÛQºÊDB��5IJTÛ JÛ JTÛTVCKFDUÛ UPÛSFEVDUJPOÛXIFOÛUIFÛTUSFTTÛXBTÛSFQMBDFE�ÛDG�ÛIOHF�UBTBOÛ�GJGUFFO��Û5IFÛEFWFMPQNFOUÛFO��JO �UIFO ÛXBTÛFBSMJFSÛUIBOÛUIFÛWPXFMÛSFEVDUJPO��ÛÛ�TFOP��ÛIJOÛ�PME��Û4LU�ÛTºOB� Û(S�Û{NOJ Û-JUI�ÛT OBT�ÛÛÛ/PUFÛUIBUÛ UIFÛFÛXIJDIÛJTÛ UIFÛSFEVDUJPOÛPGÛFBÛ JTÛOPÛMPOHFSÛTVCKFDUÛ UPÛUIJTÛDIBOHF�ÛNBUFBOÛ�CPPL� ÛHFO�ÛNBUFO�J�ÛÛÛÛÛÛDÛ�FÛ�ÛBÛÛ*MÛJTÛPOMZÛGPVOEÛJOÛUIFÛGPMMPXJOHÛXPSET�ÛÛÛÛ�EF N�ÛUBTO��UFO�Û�Û(S�ÛDzKA �-BU�ÛEFDFN �4LU�ÛEº B�ÛÛÛÛ�TVF T��ÛWBU TVOÛ�TJYUZ��ÛDG�ÛWFD Û�TJY�Û�Û�TVF TÛ�ÛÛÛ ,PSUMBOEUÛ DPOTJEFSTÛ UIFÛ BÛ UIFÛ SFEVDFEÛHSBEFÛ WPXFMÛ GSPNÛDBTFTÛXIFSFÛJUÛBQQFBSFEÛSFHVMBSMZÛBTÛTIXB�TFDVOEVNÛUIBUÛSFQMBDFEÛ[FSPÛHSBEFÛWPDBMJTNÛ��C�Û'PSÛDB SÛ�MBVHIUFS�ÛTFFÛ��ÛÛÛ ��ÛÛ1*&Û�PÛÛÛÛÛÛBÛÛ�PÛ�ÛPÛÛPOÛUIFÛDPOEJUJPOTÛTFFÛCFMPXÛÛÛ�VPS PN�ÛÛHPSDÛ�XPSL�Û�Û(S�Û|RGON �ðE�RGONÛÛÛÛ�H PVJP��LPHJÛ�CVUUFS��Û4LU�ÛHºWZB� Û(S�Û�yNNEA� BOIOJÛ�PGÛOJOFÛDPXT��ÛÛÛ�QPE�N�ÛPU�OÛ�GPPU� ÛGSPNÛUIFÛBDD�ÛGPSN�Û(S�ÛPO�D AÛÛÛ��CIPSPT�Û�XPSÛ�DBSSZJOH��Û4LU�ÛCIBSº� Û(S�Û FO�ROJ�ÛÛÛÛÛÛ8PSETÛUIBUÛTFFNÛUPÛCFHJOÛXJUIÛBÛWPXFMÛIBEÛBOÛJOJUJBMÛMBSZOHFBM�ÛUIFTFÛXJMMÛUIFSFGPSFÛCFÛEJTDVTTFEÛXJUIÛUIFÛMBSZOHFBMTÛ��ÛÛÛÛÛÛCÛ�PÛ�ÛVÛCFGPSFÛOBTBMÛÛÛÛ�QPOU�F, � �Û�ÛIVOÛÛ�DIBOOFM ÛGPSE��Û-BU�ÛQPOT �QPOU�JT �0$4ÛQPU �ÛÛÛ� PO�V��DVOS��LOFF��Û(S�ÛGO�NU �4LU�ÛK OV��ÛÛÛÛÛÛÛÛÛÛDÛ �P� �Û BÛ Û JOÛ PQFOÛ TZMMBCMFÛ JOÛ OPO�GJOBMÛ TZMMBCMF�Û PÛ XBTÛ QSFTFSWFE�ÛCFGPSFÛUXPÛDPOTPOBOUT ÛCFGPSFÛXÛBOEÛXIFOÛUIFÛGPMMPXJOHÛTZMMBCMFÛIBEÛBOÛPÛ ,PSUMBOEUÛ ��C Û ��Û <UIJTÛ WPM� Û ��>�Û 5IFÛ DIBOHFÛ P� �� BÛ NBZÛ IBWFÛQSFDFEFEÛEFWFMPQNFOUÛCÛ�PÛ�ÛV�ÛTFFÛ��ÛPOÛBOVOÛFUD�ÛÛÛÛ�QPMJP��BMJ�L ��XBWFT ÛXIJUFÛIBJS��Û(S�ÛPOLIO�J��ÛÛÛÛÛ�QP V��Û BTSÛ �GMFFDF� ÛXJUIÛ �B�Û GSPNÛHFO�Û BTV�Û UIFÛ �SÛXBTÛ BEEFEÛ MBUFS�Û 4LU�ÛQº VÛ�Û�QF V ÛÛ(S�ÛP²KOJ�ÛÛÛÛÛ��P�ÛJOÛDPNQPVOETÛCFDBNFÛ�B� ÛDG�ÛU BH�B�XPSÛ�LJOH�Û�Û��DSPXO�CFBSFS��Û*UÛJTÛQPTTJCMFÛUIBUÛ� �ÛXBTÛVTFEÛJOÛUIJTÛQPTJUJPO ÛCVUÛUIJTÛDBOOPUÛCFÛQSPWFO�ÛÛ


ÛÛÛÛ, � Û CFGPSFÛ DPOTPOBOUÛXBTÛWPDBMJ[FEÛ UPÛ �P ÛXIJDIÛPGUFOÛ GBMMTÛVOEFSÛ UIFÛSVMFTÛHJWFOÛBOEÛTPÛPGUFOÛCFDBNFÛB��Û&�H��ÛÛÛÛ�, � O SJP��Û�PO SZP��ÛBOVS ÛÛ�ESFBN��ÛÛÛÛ�, � E��BUFBN��UPÛIBUF� Û�PE��Û-BU�ÛPEJVN��IBUSFE� Û E ÛQG�Û�*ÛIBUF�ÛÛÛÛÛÛÛÛ'PSÛ UIFÛ FYDFQUJPOTÛ DG�Û �PÛ CFGPSFÛX�Û LPHJÛ BCPWF Û IPWJXÛ �TIFQIFSE�Û �Û�) � FVJ� � MPHBOBNÛ �CBUIF�Û (S�ÛLOU�WÛ�PÛCFGPSFÛBOPUIFSÛP�ÛP PSNÛ �QJUZ� ÛP PL Û�QSBZFS� ÛP PSLÛ�TNPPUI� ÛPSP ÛÛ�MBNC� ÛPSPSÛ�TFB�HVMM� ÛIPMPWFN��SPMM��ÛÛPÛCFGPSFÛUXPÛDPOTPOBOUT�ÛPSC �PSEJ �PSKJL �PSU �PTLS �PTU �PUO ÛHPSD ÛNP[JÛ�DBMG�Û�Û ��[[� Û NP BOBNÛ �GPSHFU�Û ��Û �SOÛ PSÛ �ST�Û BMTPÛ P[OJÛ �IFEHFIPH� Û PSDBNÛ�WPNJU�ÛXIJDIÛHPUÛUIJTÛTUSVDUVSFÛPOMZÛMBUFSÛ�Û�P[JOJ Û�PSVDBNÛÛÛÛÛÛ ��Û1*&ÛÛ� ��Û� ��J�ÛÛÛÛ� SEÛ�ÛTJSUÛ�IFBSU��Û(S�ÛK¡R �Û4LU�ÛI SEJ��Û5IFÛGPSNTÛXJUIÛMPOHÛ ÛEVFÛUPÛBÛMBSZOHFBMÛXJMMÛCFÛEJTDVTTFEÛUIFSF Û��ÛÛÛÛÛÛÛ ��Û1*&Û� ÛÛÛÛ� ÛÛ�ÛV��ÛÛÛÛÛ�E N��ÛUVOÛ�IPVTF��Û(S�ÛDÎ ÛDG�ÛDÎMA �4LU�ÛEBN�ÛÛÛÛ�TO CISJ��ÛOVSCÛ�TNBMM��Û0*D�ÛTOwGS�ÛÛÛ�, � O S�JP��ÛBOVS ��ESFBN��ÛDG�Û(S�ÛŃEIROJ�ÛÛ�, � F, � N S��ÛBXSÛ�EBZ�ÛWJBÛ�BXVS�Û(S�Û£MAR �H�ME�R �ÛÛ5IFÛGPSNTÛXJUIÛMPOHÛÛEVFÛUPÛBÛMBSZOHFBMÛXJMMÛCFÛEJTDVTTFEÛXJUIÛUIFÛMBSZOHFBMTÛ��ÛÛÛ ��(-4,8,32+7ÛÛ 5IFÛ 1*&Û F�Û BOEÛ P�EJQIUIPOHTÛ GFMMÛ UPHFUIFSÛ JOÛ "SNFOJBO Û BTÛGPMMPXT�ÛÛÛÛÛÛ�FJÛÛÛBOEÛ�PJ��Û�Û�FJ��Û ��FVÛ ÛBOEÛ�PV� Û�Û �PVÛ�ÛPZ�Û Û5IFÛ GJSTUÛEFWFMPQNFOUÛ JTÛWFSZÛ GSFRVFOU Û UIFÛMBUUFSÛJTÛOPUÛDPNNPO�Û$G�Û(FSNBOÛFVÛ�ÛÛ<PJ>�ÛÛÛ ��ÛÛ1*&Û�FJ��FJ�� QSFUPOJDÛJ��ÛÛÛÛÛ�EIFJ I�Û�ÛFE [Û�IFÛIFBQFEÛVQ��Û4LU�ÛEÂI�NJ Û-BU�ÛGJOH ��ÛÛ�MFJ I��ÛMJ[FN��UPÛMJDL��Û(S�ÛLEI�CW �4LU�ÛMÂI�NJ�ÛÛ/PUFÛUIBUÛVOTUSFTTFEÛ"SN�ÛJÛNVTUÛEFSJWFÛGSPN��FJÛPSÛ�PJ ÛBTÛ1*&Û�JÛJOÛVOTUSFTTFEÛQPTJUJPOÛXPVMEÛIBWFÛCFDPNFÛVOXSJUUFOÛ �ÛÛÛ�TUFJC��ÛTU QÛ�PGUFO� ÛTUJQFNÛ�VSHF��Û(S�ÛSTEI�BWÛÛ�USFBE ÛTUBNQÛPO��ÛÛÛÛÛÛ/PUFÛUIBUÛUIFÛEFWFMPQNFOUÛTUJMMÛUPPLÛQMBDFÛXIFOÛ�ZÛ�Û�UÛXBTÛJOWPMWFE�ÛÛÛ��F�UJ� �Û� ÛÛ�ÛTH�ÛQSFT�ÛJOE�ÛFOEJOH ÛCFS Û�Û�CIFS�F�UJ�GSPNÛ��FZJ��FZ� ÛXJUIÛBQPDPQF�Û

158� 5��6��3��%HHNHV��

Û ��ÛÛ1*&Û�PJÛÛÛÛÛÛBÛ�PJ�� QSFUPOJDÛJÛÛÛÛ�EIPJ I�Û�ÛE [Û�IFBQ��Û(S�ÛTOkCOJ Û(PUI�ÛEBJHT��ÛÛÛÛÛ�VPJE�Û�ÛHJUFNÛ�UPÛLOPX��Û(S�ÛOmD A Û4LU�ÛWÂE�B �Û(PUI�ÛXBJU��ÛÛÛÛÛ�VPJO�Û�ÛHJOJÛ�XJOF��Û(S�ÛðOkNOJ��CÛ�PJ��BZÛÛ�5IFSFÛBSFÛUXPÛXPSETÛXIJDIÛIBWFÛUIFÛSFGMFYÛBZ�ÛÛÛ�)PJE�Û�ÛBZU��DIFFL� ÛBZUOVNÛ�UPÛTXFMM��Û(S�ÛOI�DE�WÛÛ�UPÛTXFMM�ÛÛÛÛÛ�PJXJ Û�ÛBZHJÛ�WJOF��Û(S�ÛOgH��´ Û�TFSWJDF�USFF� Û-BU�Û WBÛ�HSBQF�Û,PSUMBOEUÛUIJOLTÛUIBUÛUIJTÛEFWFMPQNFOUÛJTÛWFSZÛPMEÛ��C Û��Û<UIJTÛWPM� Û��>ÛBOEÛ UIBUÛ UIJTÛ JTÛ UIFÛSFHVMBSÛEFWFMPQNFOUÛPGÛ �PJ�ÛBUÛ UIFÛCFHJOOJOHÛPGÛUIFÛXPSE�ÛÛÛÛ ��ÛÛ1*&Û�FV��FV��PZ �QSFUPOJDÛVÛÛÛ�MFVL�PTÛ�ÛMPZTÛ�MJHIU�ÛTVCTU��Û(S�ÛLEUKO�J�BEK�ÛÛÛ ��ÛÛ1*&Û�PVÛÛÛÛÛÛ�PV��PZ ÛQSFUPOJDÛV��ÛÛÛÛÛ�CIPVHPTÛ�ÛCPZDÛ�OPVSJTINFOU��Û4LU�ÛCIÌHB�Û�FOKPZNFOU� Û-BU�ÛGVOHPSÛÛÛÛÛÛÛÛ0OFÛ DJUFTÛ Q PZU Û �[FBM��Û (S�Û SPOUDH�Û �IBTUF� Û CVUÛ UIJTÛ DPNQBSJTPOÛ JTÛQSPCMFNBUJD�Û�TQ�ÛSFNBJOTÛTQ�ÛJOÛ"SNFOJBOÛTFFÛ��E�Û5IJTÛXPSEÛNBZÛCFÛOPO�*OEP�&VSPQFBO�ÛÛÛÛÛÛ/PUFÛUIBUÛUIFÛFWJEFODFÛJTÛWFSZÛNFBHSFÛBMTPÛGPSÛ�FV�ÛÛÛÛ ��ÛÛ5IFÛ1*&ÛMPOHÛEJQIUIPOHTÛÛ *Û LOPXÛ PGÛ POMZÛ POFÛ DBTFÛ PGÛ JOEJSFDUÛ FWJEFODFÛ PGÛ BÛ 1*&Û MPOHÛEJQIUIPOH�Û *UÛ JTÛ UIFÛ FOEJOHÛ PGÛ UIFÛ EBUJWFÛ PGÛ UIFÛ JOEFGJOJUFÛ QSPOPVOÛ P�L Û�TPNCPEZ��ÛÛÛ�JP�TN J�L <F>�Û VNF�L �Û UIFÛ FOEJOHÛ XBTÛ IFSFÛ SFUBJOFEÛ CFGPSFÛ UIFÛHFOFSBMJ[JOHÛQBSUJDMFÛ��L <F>ÛPOÛXIJDIÛTFFÛ��ÛDG�Û��ÛPOÛUIFÛSFUBJOFEÛWPXFMT�Û,PSUMBOEUÛ��B Û��Û<UIJTÛWPM� Û��>�ÛÛ,PSUMBOEUÛBTTVNFTÛ JCJEFNÛ UIBUÛVSF�L Û TPNFXIFSF�Û JTÛ BÛ MPDBUJWFÛJOÛ �PJ Û BOEÛ UIBUÛ BMMÛ GJOBMÛ J�EJQIUIPOHTÛNFSHFEÛ JOUPÛ ��FÛ CFGPSFÛ UIFZÛXFSFÛBQPDPQBUFE�Û 5IBUÛ UIFÛ MPOHÛ EJQIUIPOHÛ XBTÛ TIPSUFOFEÛ UPÛ ��PJÛ JTÛ RVJUFÛQSPCBCMF�ÛÛÛ ��8,)�4638,)8-'�:3;)0�ÛÛ "SNFOJBOÛEJEÛOPUÛUPMFSBUFÛXPSETÛCFHJOOJOHÛXJUIÛS��Û1SPUP�*OEP�&VSPQFBOÛEJEÛOPUÛIBWFÛTVDIÛXPSET�ÛXIFSFÛJUÛTFFNTÛUPÛCFÛTP ÛUIFSFÛXBTÛBÛQSFDFEJOHÛMBSZOHFBM�Û5IJTÛMBSZOHFBMÛXBTÛOPUÛSFDPHOJ[FEÛCFDBVTFÛJOÛNPTUÛMBOHVBHFTÛJUÛXBTÛMPTUÛXJUIPVUÛUSBDF�Û*OÛ"SNFOJBOÛJUÛXBTÛWPDBMJ[FE ÛUPÛF� �


B� � P�Û SFTQ��Û TFFÛ ��Û 5IJTÛNFBOTÛ UIBUÛ JOÛ "SNFOJBOÛ BÛ QSPUIFUJDÛ WPXFMÛXBTÛPOMZÛBEEFEÛCFGPSFÛBOÛSÛUIBUÛCFDBNFÛXPSE�JOJUJBMÛTFDPOEBSJMZ�Û0MEFSÛTUVEJFTÛPGUFOÛTUBUFÛ UIBUÛ UIFÛQSPUIFUJDÛWPXFMÛBSPTFÛFWFOÛCFGPSFÛTFDPOEBSZÛJOJUJBMÛS�Û5IJTÛEBUFTÛGSPNÛUIFÛUJNFÛXIFOÛUIFÛMBSZOHFBMÛEFWFMPQNFOUTÛIBEÛOPUÛ ZFUÛ CFFOÛ SFDPHOJ[FE�Û 5IJTÛNFBOTÛ UIBUÛ JGÛ UIFÛ FUZNPMPHZÛ PGÛ BÛXPSEÛTIPXTÛ UIBUÛ UIFÛ S�Û XBTÛ PSÛ TFFNFEÛ JOJUJBMÛ JOÛ 1*&Û BMSFBEZ Û BÛ MBSZOHFBMÛQSFDFEFE�Û"TÛUIJTÛMBSZOHFBMÛXBTÛWPDBMJ[FEÛJOÛ"SNFOJBO ÛBÛSFBMÛQSPUIFUJDÛWPXFMÛ JTÛ JNQPTTJCMF Û BÛ GBDUÛ UIBUÛPGUFOÛ JTÛOPUÛ SFBMJ[FE�Û5IJTÛ JTÛ FWFOÛNPSFÛDMFBSÛXIFOÛ BÛ(SFFLÛ DPHOBUFÛIBTÛ BÛ WPXFMÛ GSPNÛBÛ MBSZOHFBMÛ CFGPSFÛ S��Û *OÛUIBUÛ DBTFÛ UIFÛ "SNFOJBOÛ WPXFMÛ NVTUÛ SFQSFTFOUÛ UIFÛ TBNFÛ MBSZOHFBMÛ BOEÛDBOOPUÛCFÛUIFÛMBUFSÛ"SNFOJBOÛQSPUIFTJT�ÛDG�ÛGVSUIFSÛ��ÛÛ 8IFOÛUIFÛQSPUIFUJDÛWPXFMÛXBTÛBEEFEÛUIFÛBDDFOUÛDPVMEÛOPÛMPOHFSÛCFÛ XJUIESBXOÛ UPÛ JU�Û FSJTÛ �Û �USJOTÛ LFQUÛ UIFÛ TUSFTTÛ POÛ UIFÛ J�Û JUÛ XBTÛ OPUÛXJUIESBXOÛUPÛUIFÛF�CFDBVTFÛXFÛXPVMEÛUIFOÛIBWFÛHPUÛ�FST�ÛÛ 5IFSFÛBSFÛUISFFÛXBZTÛJOÛXIJDIÛBOÛSÛDPVMEÛCFDPNFÛUIFÛJOJUJBMÛPGÛBÛXPSEÛTFDPOEBSJMZ�ÛÛÛ��ÛUISPVHIÛMPTTÛPGÛBÛQSFDFEJOHÛ1*&ÛWPJDFMFTTÛTUPQ Û�Q Û�U Û�L�ÛÛÛ��ÛUISPVHIÛNFUBUIFTJT�ÛÛÛ��ÛUISPVHIÛUIFÛEFWFMPQNFOUÛ�TS�Û�Û ��ÛÛ 5IFÛNBUUFSÛJTÛNVDIÛNPSFÛDPNQMJDBUFEÛUIBOÛJUÛBQQFBST�Û,PSUMBOEUÛBTTVNFTÛ UIBUÛ UIFÛ EFWFMPQNFOUÛ PGÛ BÛ QSPUIFTJTÛ XBTÛ UIFÛ GJSTUÛ TUFQÛ JOÛ UIFÛEFWFMPQNFOUÛ PGÛ UIFTFÛ GPSNT Û GPMMPXFEÛ CZÛ UIFÛ BCPWFÛ QSPDFTTFTÛ MPTT ÛNFUBUIFTJTÛ BOEÛ �TS�� Û &WFSZÛ TUFQÛ XBTÛ JOUFOEFEÛ UPÛ GBDJMJUBUFÛ UIFÛQSPOVODJBUJPOÛBOEÛFWFOUVBMÛFMJNJOBUJPOÛPGÛJOJUJBMÛDMVTUFST�Û5IVTÛ US�ÛXBTÛFBTJFSÛ CFDBVTFÛ PGÛ UIFÛ TZMMBCJGJDBUJPO Û UIBOÛ US��Û "OEÛ SU�Û JTÛ TUJMMÛ FBTJFS�Û8IFOÛPSJHJOBMÛWPJDFMFTTÛTUPQTÛEJTBQQFBSFE Û UIFÛQSPUIFUJDÛWPXFMÛEJEÛOPUÛBSJTF ÛF�H�Û� MVUPT��MV��0OFÛBMTPÛFYQFDUTÛNFUBUIFTJTÛJOÛUIFÛDBTFÛPGÛHSPVQTÛMJLFÛ�EO� Û�EIO� Û� O� Û� M� ÛXIFSFÛUIFÛTUPQÛXPVMEÛIBWFÛCFFOÛQSFTFSWFEÛJOÛQSJODJQMFÛVOMFTTÛUIFSFÛXFSFÛBEEJUJPOBMÛSVMFTÛGPSÛ UIFTFÛDBTFT�Û *ÛIBWFÛOPUÛGPVOEÛFWJEFODFÛGPSÛTVDIÛHSPVQT�ÛÛÛÛÛÛÛ"EÛ��ÛÛÛ�USFJFTÛ�ÛFSFL Û�UISFF��Û4LU�ÛUSºZBT Û(S�ÛTREkJ Û-BU�ÛUS T�ÛÛÛ�L SFQ��Û FSFXJNÛ �BQQFBS��Û (S�ÛPRzPW�Û 5IFÛ JOJUJBMÛNBZÛ IBWFÛ CFFOÛ �L Û OPUÛ�Q� ÛCFDBVTFÛ1*&ÛEJEÛOPUÛIBWFÛSPPUTÛXJUIÛ$�Û�Û$��Û�QSFQ�ÛJTÛBMTPÛQPTTJCMFÛÛÛÛÛÛ"EÛ��ÛÛÛ�ESBL �Û �Û BSUBTV�L Û �UFBST��Û0)(Û USBIBO�ÛXJUIÛNFUBUIFTJTÛ(S�ÛD�KRU Û(PUI�ÛUBHS�Û4FFÛPOÛUIJTÛEJGGJDVMUÛXPSEÛ,PSUMBOEUÛ��B�ÛÛÛ�CISF, � U SÛ �Û F CBZSÛ �CSPUIFS��Û 4LU�Û CIS UBS� Û (S�� FR THR Û -BU�Û GS UFS�Û 5IFÛ ÛBSPTFÛUISPVHIÛEJTTJNJMBUJPOÛCFGPSFÛUIFÛGPMMPXJOHÛS�ÛÛÛ.PSFÛFYBNQMFTÛDBOÛCFÛGPVOEÛJOÛ�� ÛXIFSFÛUIFÛNFUBUIFTJTÛJTÛUSFBUFE�Û

160� 5��6��3��%HHNHV��

ÛÛÛÛÛ"EÛ��ÛÛÛ�TSVUJ��ÛB VÛ�CSPPL��Û4LU�ÛTSVUÆ� Û(S�Û�U�SIJ Û0*S�ÛTSVUI�ÛÛÛ�TSPV��Û�F PHFN��P PHFNÛ�JSSJHBUF��Û4LU�ÛTSºW NJ Û(S�ÛR@E�W�ÛÛÛÛÛÛÛ,PSUMBOEUÛGPMMPXTÛ1FEFSTFOÛJOÛBTTVNJOHÛUIBUÛUIFÛQSPUIFUJDÛWPXFMÛXBTÛF�Û XIJDIÛ DPVMEÛ CFDPNFÛ P� Û TFFÛ UIFÛ OFYUÛ TFDUJPO Û XIJMFÛ B� Û XIJDIÛ JTÛMJNJUFEÛ UPÛ OPVOT ÛXBTÛ PSJHJOBMMZÛ BÛ QSFQPTJUJPO Û JEFOUJDBMÛXJUIÛ 4MBWJDÛ QPÛ��ÛÛ ��:3;)0��'3286%'8-32��%2(��:3;)0�%77-1-0%8-32�ÛÛÛÛÛÛÛ ��ÛÛ7PXFMÛDPOUSBDUJPOÛÛÛÛÛÛÛÛ 8IFOÛBOÛJOUFSNFEJBUFÛDPOTPOBOUÛEJTBQQFBSFE Û UIFÛWPXFMTÛXFSFÛDPOUSBDUFE�Û5IFÛDPOTPOBOUTÛUIBUÛEJTBQQFBSFEÛXFSFÛ�ZÛBOEÛ�T�ÛÛÛÛÛÛÛ5XPÛ JEFOUJDBMÛ WPXFMTÛ SFTVMUFEÛ JOÛ BÛ MPOHÛ POF Û XIJDIÛ XBTÛ TIPSUFOFE�ÛOPUFÛUIBUÛTVDIÛBÛOFXÛ�FÛPSÛ�PÛSFTVMUFEÛJOÛF �P ÛOPUÛJOÛJ �V�ÛÛÛ�USFJFTÛ�ÛFSFL Û�UISFF��ÛÛÛ�CIPTP�H � P��ÛCPLÛ�CBSFGPPU��ÛÛÛ�OJ�TJTE��Û �OJTU�JNÛ �Û OTUJNÛ �TJUÛ EPXO��Û /PUFÛ UIBUÛ UIJTÛ JÛ XBTÛ QSFUPOJDBMMZÛSFEVDFEÛUPÛ �ÛÛÛÛÛÛ8IFOÛUIFÛWPXFMTÛXFSFÛOPUÛ JEFOUJDBM Û UIFÛQIPOFUJDÛ SFTVMUÛXBTÛOFWFSÛFÛJUÛTPÛIBQQFOTÛUIBUÛXFÛIBWFÛPOMZÛFYBNQMFTÛJOÛXIJDIÛPOFÛWPXFMÛXBTÛF�ÛÛÛÛÛÛF��PÛ�ÛP�ÛÛÛ�TVFTPS�FT�ÛL PS�L Û�TJTUFST��ÛÛÛ�L FUVPSFT�Û� FZPS�FL Û�Û PS�L ÛÛ�GPVS��ÛÛÛÛÛÛF��ÛB��B�ÛÛÛÛ�, � FT, � S�Û�FIBSÛ�ÛBS�JXO� �CMPPE�Û,PSUMBOEUÛBTTVNFTÛUIBUÛUIFÛMBSZOHFBMÛXBTÛWPDBMJ[FEÛJOÛUIJTÛGPSNÛ��ÛÛÛÛÛÛB��F��B�ÛÛÛÛÛÛÛ�Q) � UFS�FT�Û �IBZFS�FL Û �Û IBS�L Û �GBUIFST��Û UIFÛ �B�Û DPVMEÛ CFÛ BOBMPHJDBM ÛIPXFWFS�ÛÛÛ/PUFÛ UIBUÛ JOÛ �TVFTS�CIJ�Û �L FISCJÛ UIFÛ IÛ EJTBQQFBSFEÛ CFGPSFÛ UIFÛ SÛ XBTÛWPDBMJ[FE ÛXIJDIÛHBWFÛL FSC�SFHVMBSMZ�ÛDG�Û KFSC�BLBM� �QSJTPOFS�� GSPNÛ� IFTS�CIJ�ÛÛÛÛÛÛÛ/PUFÛUIFÛGPMMPXJOHÛDPOUSBDUJPOÛXIJDIÛSFTVMUFEÛJOÛBÛEJQIUIPOH�ÛÛÛ�TVFT S�Û�L FVSÛ�ÛL PZS�ÛTPÛ�TÛ�Û�IÛPDDVSSFEÛBGUFSÛ� Û�ÛVÛCVUÛCFGPSFÛ�FVÛ��PZÛ,PSUMBOEUÛ��C ÛTUBHFÛ��Û<UIJTÛWPM� Û��>�ÛÛÛ,PSUMBOEUÛ ��Û FYQMBJOTÛ O SÛ GSPNÛ FBSMJFSÛ BOBMPHJDBMÛ �OFZJS Û XIFSFÛ FZ�ÛCFDBNFÛ ÛMJLFÛUIFÛ1*&ÛEJQIUIPOHÛ�FJ�ÛCVUÛDG�Û��ÛPOÛUIJTÛGPSN�ÛÛÛ ��ÛÛ7PXFMÛBTTJNJMBUJPOÛÛ 5IFSFÛXBTÛWPXFMÛBTTJNJMBUJPO ÛF��P��P��PÛBOEÛF��V��P��V�Û


ÛÛ�TSPV��Û�F PHFNÛÛ�ÛP PHFNÛ�JSSJHBUF��ÛÛÛ�, � SV ��Û �FSVD�Û �Û �PSVD�Û �Û PSD�BNÛ �WPNJU��Û (S�Û E�REU�GOMAI�Û 0OÛ UIFÛEFWFMPQNFOUÛPGÛUIFÛMBSZOHFBMÛTFFÛ��ÛÛÛÛÛ ��7)1-:3;)07�Û ÛÛ 5IFÛ 1*&Û TFNJWPXFMTÛ BSFÛ JÛ BOEÛ V�Û 5IFZÛ DBOÛ CFÛ TZMMBCJDÛ PSÛDPOTPOBOUBM�Û /PUFÛ UIBUÛ UIFÛBQQBSFOUÛ MPOHÛ Û ÛXFSFÛ JOÛ GBDUÛ J) �V)�Û UIFZÛBSFÛUSFBUFEÛJOÛ��ÛÛÛÛ ��ÛÛ5IFÛTFNJWPXFMTÛBTÛWPXFMTÛÛ ��ÛÛ4FNJWPXFMTÛBTÛWPXFMT�Û1*&Û�JÛÛÛÛÛÛ�J��J ÛQSFUPOJDÛ ÛOPUÛXSJUUFO��ÛÛÛÛÛÛÛÛ0OÛJ�BÛ�ÛFBÛTFFÛ��Û�JP�ÛBOEÛJXÛ��ÛÛÛ�, � F�VJE�FUÛ�ÛFHJUÛ�IFÛGPVOE� Û�ÛTH�ÛBPS�ÛHU�J�ÛQSFT��HUBOFN�ÛÛÛÛÛÛ�USJ�OTÛ�ÛFSJTÛ�UISFF�ÛBDD�ÛQM��Û(PUI�Û×SJOT��ÛÛÛÛÛ�EJV�Û�ÛUJXÛ�EBZ��ÛDG�Û4LU�ÛEÆW Û�CZÛEBZ�ÛÛÛ ��ÛÛ4FNJWPXFMTÛBTÛWPXFMT�ÛÛ1*&Û�V��V��V ÛQSFUPOJDÛ ÛOPUÛXSJUUFOÛÛÛÛ�EIVH, � U SÛ�ÛEVTUS �HFO�ÛETUFSÛ�EBVHIUFS��Û(S�Û UGA�THR �4LU�ÛEVIJUºS� Û(PUI�ÛEBVIUBS��TOVTPTÛ�ÛOVÛ�EBVHIUFS�JO�MBX��Û(S�ÛNUO�J Û4LU�ÛTOV Û0)(ÛTOVS��ÛÛÛÛÛ�, � O IV�Û �Û BOKVLÛ �OBSSPX��Û 4LU�Û BNIÓ� � �0$4Û P[ L Û (PUI�Û BHHXVT �(S�Û�GC WÛÛÛÛ�VFT�OV�NJÛ�Û[�HFOVNÛ�UPÛESFTT�Û[�ÛJTÛBÛQSFGJY�Û(S�Û{NN MI �4LU�ÛWºT�UF�ÛÛÛ ��ÛÛ5IFÛTFNJWPXFMTÛBTÛDPOTPOBOUTÛÛÛÛÛÛÛÛÛ ��ÛÛ1*&Û�ZÛÛÛÛÛÛÛBÛ�Z�� ÛÛBGUFSÛSFTPOBOUÛS �M �O �N�JGÛUIFÛQSFDFEJOHÛWPXFMÛJTÛOPUÛB��ÛÛÛÛÛ�TUFS�JP��Û TUFS � �TUFSJMF��Û (S�Û STEkRA � 4LU�Û TUBS �Û �CBSSFOÛ DPX� Û -BU�Û TUFSJMJT Û(PUI�ÛTUBÆSP��ÛÛ�, � O S�JP�Û�ÛBOVS Û�ESFBN��Û(S�ÛŃEIRON��POFS�JP��ÛÛÛ�NVOJ�Û �Û NVO Û �EVNC��Û (S�Û MU�NDOJ��MUNARO�J � 4LU�Û NÓOJ�� XJTFÛ NBO ÛCFHFJTUFSU� Û-BU�ÛN UVT��ÛÛÛÛÛ�H IFO�JP�Û�Û O FNÛ�XJQFÛDMFBO�Û�Û� JO ��ÛFUZN�ÛVOLOPXOÛOPUÛDPHOBUFÛXJUIÛ(S�ÛKEeNW �-JUI�ÛHFOJÒÛ��ÛÛÛÛÛÛÛÛÛCÛ��B3Z� Û�P3Z��BZ3� Û�PZ3�ÛÛÛÛÛÛÛ*GÛ UIFÛ QSFDFEJOHÛ WPXFMÛ JTÛ B Û XFÛ HFUÛ ZÛ CFGPSFÛ SFTPOBOUÛ JOGFDUJPO�Û 5IFÛEFWFMPQNFOUÛ JTÛ QBSBMMFMÛ XJUIÛ UIBUÛ JOÛ (SFFL Û XIFSFÛ SFTPOBOUÛ �Û ZÛ HJWFTÛJOGFDUJPOÛBGUFSÛBÛBOEÛP�Û*OÛBMMÛDBTFTÛXFÛQSPCBCMZÛIBWFÛBOUJDJQBUJPOÛPGÛUIFÛZ�

162� 5��6��3��%HHNHV��

B � Z�ÛJGÛUIFÛZÛSFNBJOT ÛUIFÛJOGFDUJPOÛPGÛUIFÛWPXFMÛEPFTÛOPUÛEFWFMPQÛJOUPÛBÛTFQBSBUFÛQIPOFNF�Û4FFÛBMTPÛ��ÛÛÛÛÛ�, � FMJP��BZMÛ�PUIFS��Û(S�Û�LLOJ Û-BU�ÛBMJVT Û(PUI�ÛBMKJT Û0*S�ÛBJMF�ÛÛÛÛÛÛÛ� IXPO�J��KBZOÛ�WPJDF��Û0$4Û[WPO Û�TPVOE��ÛÛÛÛÛÛ-BZOÛ�XJEF�ÛNBZÛSFQSFTFOUÛ�Q U, � �O�J��Û�Û�IM OZ�Û�Û�MBOZ�� ÛÛ 5IFTFÛ BSFÛ UIFÛ CFTUÛ EPDVNFOUFEÛ FYBNQMFT�Û *UÛ JTÛ QSPCBCMZÛ BMTPÛGPVOEÛ JOÛQ BZMFNÛ �TIJOF� Û DG�ÛQ PMQ PMJNÛ �TIJOF Û HMJUUFS� ÛXIJDIÛIBTÛ OPÛHPPEÛFUZNPMPHZ�ÛÛÛÛ 5IFÛ EFWFMPQNFOUÛ PDDVSSFEÛ BGUFSÛ B Û CVUÛ UIFSFÛ JTÛ BMTPÛ TPNFÛFWJEFODFÛ GPSÛ JUÛ BGUFSÛ P�Û 5IJTÛ DPOTJTUTÛ POMZÛ PGÛ CPZMÛ �DPNQBOZ Û BTTFNCMZ�ÛCFTJEFÛCPMPSÛ�BMM ÛFOUJSF� ÛXIJDIÛIBTÛOPÛFUZNPMPHZ�Û5IFÛTBNFÛIPMETÛGPSÛU PZMÛUBMÛ �HJWFÛQFSNJTTJPO� ÛBOEÛO�uPZMÛ �MJHIU�ÛCFTJEFÛuP Û �SBZ Û GMBTIÛPGÛ MJHIU��Û5IJTÛFWJEFODFÛJTÛSBUIFSÛVODFSUBJO�Û5IFÛFUZNPMPHZÛPGÛP Û �XIPMF ÛTPVOE�ÛJTÛVODFSUBJOÛ�PMZP� �ÛÛÛÛÛÛ4FFÛGVSUIFSÛ��Û��ÛÛÛÛÛÛÛÛÛÛDÛ�ZÛÛ�Û[FSP�ÛÛXPSE�JOJUJBMMZÛBOEÛCFUXFFOÛWPXFMTÛÛÛ8PSEÛJOJUJBMÛZ�ÛÛÛÛ�JP�TÛ�ÛPSÛÛ�XIP��ÛÛÛ5IFÛ EFWFMPQNFOUÛ JOÛ BOMBVUÛ JTÛ EFCBUFEÛ ,PSUMBOEUÛ ��B�Û 5XPÛXPSETÛTFFNÛUPÛQPJOUÛ UPÛZ��Û M��Û MVD� �ZPLF�Û�Û�ZVH�ÛBOEÛ MFBSEÛ �MJWFS�Û JGÛ GSPNÛ�ZFL �Û)PXFWFS Û BÛ EFWFMPQNFOUÛ UPÛ MÛ JTÛ QIPOFUJDBMMZÛ JNQSPCBCMF�Û 5IFÛ XPSETÛNVTUÛIBWFÛCFFOÛJOGMVFODFEÛCZÛPUIFSÛXPSETÛUIFÛGJSTUÛCZÛMVDBOFNÛ �UPÛCJOEÛMPPTF� �ÛUIFÛTFDPOEÛXPSEÛQSPWJEFTÛTFWFSBMÛEJGGJDVMUJFT�Û(FSNBOJDÛBMTPÛIBTÛBOÛM� ÛDG�ÛMJWFS�ÛJUÛXBTÛTVQQPTFEMZÛUBLFOÛGSPNÛUIFÛXPSEÛGPSÛ�GBU� Û�MJQ� �DG�Û(S�ÛLIPARO�J�Û5IFÛMÛJTÛOPUÛUPÛCFÛSFDPHOJ[FEÛBTÛBÛSFHVMBSÛEFWFMPQNFOU�ÛÛÛÛÛÛÛ"SN�Û VSÛ�XBUFS�ÛIBTÛCFFOÛDPOOFDUFEÛXJUIÛ-JUI�ÛK SBÛ�TFB��Û)PXFWFS ÛUIFÛ-JUIVBOJBOÛXPSEÛJTÛVTVBMMZÛDPOOFDUFEÛXJUIÛ4LU�ÛW SJ ÛTPÛJUTÛ �ÛDBOOPUÛCFÛPSJHJOBM�Û "MTP Û "SN�Û KFSÛ �ZPVÛ QM��Û CFTJEFÛ 4LU�Û Z ZºN� IBTÛ CFFOÛ BEEVDFE�Û,PSUMBOEUÛ BTTVNFTÛ UIBUÛ UIFÛ QSPUP�GPSNÛ NVTUÛ IBWFÛ IBEÛ �ZV�Û �Û F Û JOÛ BOÛBDDVTBUJWFÛ�ZXFÛQBSBMMFMÛUPÛ�UXFÛ�ÛL F�ÛBOEÛUIBUÛ�ZXÛGFMMÛUPHFUIFSÛXJUIÛUIFÛSFGMFYÛPGÛ� IX��ÛÛ 5IFÛFUZNPMPHZÛPGÛUIFÛQSPOPNJOBMÛTUFNÛP�ÛQPJOUTÛUPÛ�Z�ÛUIFÛJEFBÛUIBUÛ JUÛ SFQSFTFOUFEÛ �L P�Û DBOOPUÛ CFÛ SFUBJOFE Û BTÛ �L Û EJEÛ OPUÛ EJTBQQFBSÛCFDPNJOHÛ[FSPÛBTÛUIFÛSFHVMBSÛSFGMFYÛBTÛF�H�ÛJOÛ(SFFL�ÛÛÛÛÛÛÛÛÛÛÛÛ�Z��[FSPÛCFUXFFOÛWPXFMTÛÛÛÛ�USFJ�FTÛ�ÛFSFL ��UISFF��Û4LU�ÛUSºZBT �(S�ÛTREkJ �Û-BU�ÛUS T��F F� � F�Û�ÛEFOPNJOBUJWFÛTVGGJYFTÛ�FN ��BN��ÛÛÛÛUIFÛOPN�ÛQM�ÛPGÛUIFÛJ�TUFNTÛ��FJ�FT��FL ��L �ÛÛ


ÛÛÛÛÛÛEÛ�ZJ�Û�ÛTUSFTTFEÛIJ� ÛBOEÛ1"SN�Û�IJ�Û�ÛVOTUSFTTFEÛ �ÛÛÛÛÛÛ5IJTÛJTÛBÛWFSZÛDPNQMJDBUFEÛNBUUFS ÛBMTPÛCFDBVTFÛPGÛQSPCMFNTÛXJUIÛUIFÛOPUBUJPO�Û,PSUMBOEUÛ��B Û��G�Û<UIJTÛWPM� Û��G�>�ÛÛÛÛÛÛ*UÛIBTÛCFFOÛPCTFSWFEÛUIBUÛUIFÛEJBMFDUTÛIBWFÛBÛTFFNJOHMZÛTFDPOEBSZÛI�ÛJUTÛIJTUPSZÛEPFTÛOPUÛDPODFSOÛVTÛIFSF ÛCVUÛGVSUIFSÛUIBUÛTPNFÛEJBMFDUTÛIBWFÛBÛWPJDFMFTTÛI�ÛCFTJEFÛBÛWPJDFEÛ)� ÛOPUFEÛI�ÛBOEÛ)�ÛSFTQFDUJWFMZ�Û5IJTÛ)�ÛJOÛTPNFÛ DBTFTÛ DPOUJOVFTÛ $MBTT�Û Z� Û UIFÛ QSFQPTJUJPO�Û *UÛ JTÛ BMTPÛ GPVOEÛ JOÛ UIFÛJOUFSSPHBUJWFÛ QSPOPVO�� )VWÛ �Û PWÛ �XIP � Û)VNÛ �Û VNÛ EBU�Û �GPSÛ XIPN � Û)VSÛ�ÛVSÛ�XIFSF ��ÛÛÛÛÛÛ/PXÛ ,PSUMBOEUÛ BTTVNFTÛ UIFÛ GPMMPXJOHÛ TFRVFODFÛ PGÛ FWFOUT�Û 1*&Û �Z�ÛCFDBNFÛ)� Û XIJDIÛ XBTÛ MPTUÛ JOÛ UIFÛ $MBTTJDBMÛ MBOHVBHFÛ CVUÛ SFUBJOFEÛ JOÛTPNFÛ EJBMFDUTÛ FYDFQUÛ CFGPSFÛ J�Û TPÛ �ZJ�Û CFDBNFÛ �)J��Û -BUFSÛ UIJTÛ )�ÛEJTBQQFBSFEÛUPP�Û� FO, � U SÛ�Û�)JO�Û�Û O�Û�ÛO� ÛJOÛO SÛ�CSPUIFS�TÛXJGF��ÛÛÛÛÛÛÛ'VSUIFSÛ UIJTÛ �)J�Û CFDBNFÛ IJ�Û XJUIÛ UIFÛ �OPSNBM� Û WPJDFMFTTÛ IÛ XIFOÛTUSFTTFE�Û 5IJTÛ XBTÛ UIFÛ DBTFÛ JOÛ BÛ NPOPTZMMBCMFÛ PUIFSÛ JOTUBODFTÛ BSFÛVOLOPXO�Û *OÛ UIJTÛ XBZÛ EBU�Û IJNÛ �GPSÛ XIPN �Û JTÛ FYQMBJOFE�Û 4PÛ JUÛ JTÛTVQQPTFEÛ UIBUÛ UIJTÛ GPSNÛ HPFTÛ CBDLÛ UPÛ �ZJN�Û <*Û UIJOLÛ UIBUÛ JUÛ DPOUJOVFTÛ�FTN JÛ Û SFDUFÛ �, � FTN JÛ XJUIÛ UIFÛ Z�Û PGÛ ZPTÛ �Û PWÛ �XIP �Û UIFÛ �WÛ BTÛ ZFUÛVOFYQMBJOFEÛ BEEFE Û CVUÛ DG�Û [�HFOVNÛ �ESFTT�Û �Û �VFTOVNJ�Û 5IFÛ WBSJBOUÛ JNÛNBZÛIBWFÛMPTUÛI�ÛBGUFSÛQSFQPTJUJPO�>ÛÛÛÛÛÛ5IFÛ GPMMPXJOHÛ IBTÛ OPUIJOHÛ UPÛ EPÛ XJUIÛ �Z� Û CVUÛ NBZÛ CFÛ USFBUFEÛ IFSFÛCFDBVTFÛ UIFÛ EJGGFSFODFÛ CFUXFFOÛ I�Û BOEÛ )�Û JTÛ DPODFSOFE�Û ,PSUMBOEUÛBTTVNFTÛUIBUÛIJ�Û XJUIÛWPJDFMFTTÛI� ÛF�H�ÛGSPNÛ1*&Û�QJ�Û BTÛ JOÛIJOHÛ �GJWF�Û�Û�QFOL F�ÛTFFÛCFMPXÛPOÛZJTVOÛ�GJGUZ� ÛCFDBNFÛ) �ÛXJUIÛUIFÛWPJDFEÛI�ÛCFGPSFÛUIFÛTUSFTT�ÛUIFÛ ÛJTÛUIFÛXFMM�LOPXOÛSFEVDUJPOÛPGÛQSFUPOJDÛJ�Û5IJTÛJTÛQFSIBQTÛGPVOEÛJOÛ NQFNÛ �UPÛESJOL� Û JGÛ UIFÛQSPUPGPSNÛJTÛDPSSFDUMZÛSFDPOTUSVDUFEÛBTÛ�QJNC��ÛIFSFÛUIFÛ)ÛBHBJOÛEJTBQQFBST ÛCVUÛUIFÛ ÛJTÛSFUBJOFEÛCFGPSFÛBÛDMVTUFS�ÛÛÛ 5IFÛUXPÛTFSJFTÛPGÛEFWFMPQNFOUTÛNBZÛCFÛTVNNBSJ[FEÛBTÛGPMMPXT�ÛÛÛÛÛ1*&Û�ZJ�ÛÛ�ÛÛ�)J�ÛÛ�ÛÛTUSFTTFEÛÛÛÛIJ�ÛÛÛBTÛJOÛIJNÛÛ�Û�ZJNÛÛÛÛÛ1*&Û�QJÛÛÛ�ÛÛÛÛÛÛÛIJ�ÛÛ�ÛÛQSFUPOJDÛ�) �Û�Û �ÛÛBTÛJOÛ NQFNÛÛÛÛÛÛ*UÛJTÛBTTVNFEÛUIBUÛJOÛUIFÛXPSEÛZJTVOÛUIJTÛ�) �ÛCFDBNFÛ)J ÛBOEÛUIBUÛUIJTÛXBTÛXSJUUFOÛZJ�ÛCFDBVTFÛ$MBTTJDBMÛ"SNFOJBOÛIBEÛOPÛTJHOÛGPSÛ) ÛBOEÛBTÛZJÛJTÛQIPOFUJDBMMZÛJEFOUJDBMÛUPÛ)J�Û5IJTÛQPJOUÛJTÛSBUIFSÛEJGGJDVMU ÛUIFÛNPSFÛTPÛBTÛUIFÛQSFGPSNÛPGÛZJTVOÛJTÛVODFSUBJO�Û4PÛUIFÛQSPCMFNÛJTÛUIBUÛXFÛIBWFÛIFSFÛOPUÛ POMZÛ UIFÛ EFWFMPQNFOUÛ PGÛ QSFUPOJDÛI�Û CFGPSFÛ J Û CVUÛ BMTPÛ UIBUÛ PGÛ UIFÛEFWFMPQNFOUÛ PGÛ QSFUPOJDÛ JÛ JOÛ UIJTÛ QPTJUJPO�Û 5XPÛ TPMVUJPOTÛ NBZÛ CFÛDPOTJEFSFE�Û�Û8FÛTUBSUÛGSPNÛ�QFOL E PNU, � Û�Û�IJOHTVOÛ�Û�IJ TVOÛXIJDIÛCFDBNFÛ �IJZTVO� �� ) ZTVO� �� )JTVO�Û #FUUFSÛ JTÛ QFSIBQTÛ �Û UPÛ TUBSUÛ GSPNÛ�QFOL FE PNU) � Û�Û�IJOH TVO��Û�IJOHJTVOÛ�Û�IJ JTVO��IJZJTVO��)JTVO�ÛÛÛÛÛÛÛ&WFOÛNPSFÛ DPNQMJDBUFEÛ JTÛ UIFÛ IJTUPSZÛ PGÛ UIFÛ QSFQPTJUJPOÛ JO Û CVUÛ UIJTÛ

164� 5��6��3��%HHNHV��

IJTUPSZÛMJFTÛPVUTJEFÛ$MBTTJDBMÛ"SNFOJBO ÛTPÛ*ÛSFGFSÛUPÛ,PSUMBOEU�TÛBSUJDMF�ÛÛÛÛÛÛÛFÛ$MVTUFSTÛXJUIÛ�Z�Û$POTPOBOUÛ�ÛZÛPGUFOÛHJWFTÛTQFDJBMÛEFWFMPQNFOUT ÛXIJDIÛBSFÛEJTDVTTFEÛXJUIÛUIFÛDMVTUFSTÛJOÛ��Û5IFZÛBSF�Û�TZ Û �UZ ��EZ ��EIZ�� Z ��L Z�ÛÛÛÛÛÛÛ ��ÛÛÛ1*&Û�X�ÛÛÛÛÛÛB��X��ÛHÛFYDFQUÛXIFOÛXPSE�GJOBMÛBGUFSÛUIFÛBQPDPQFÛÛÛÛ�XPS �Û�ÛHPSDÛ�XPSL��Û(S�Û|RGON Û.P)(Û8FSL��ÛÛÛÛÛ�XPJE�Û�ÛHJUFNÛ�LOPX��Û4LU�ÛWÂEB �(S�ÛOmDA�ÛÛÛ�XFT�Û �Û [�HFOVNÛ �UPÛ DMPUIÛPOFTFMG� Û GSPNÛ � FT�OFV�NJ� Û [�Û JTÛ BÛQSFGJY�Û(S�Û{NNUMAI��{NN MI�ÛÛÛ�H PX�J)P�Û�ÛLPHJÛ�CVUUFS��ÛTFFÛC�ÛPOÛ�DPX��ÛÛÛÛ�EF, � J�XFSÛ�ÛUBZHSÛ�IVTCBOE�TÛCSPUIFS��Û4LU�ÛEFWºS� Û(S�ÛD H�R �-JUI�ÛEJFWFSÅT �BDD�ÛEÆFWFS Û0$4ÛE WFS ��ÛÛÛÛÛ� POX��ÛDVOH�L Û�LOFF�T�ÛQM�ÛÛÛÛÛÛÛ0OÛ UIFÛ QSPCMFNÛ TFFÛ ,PSUMBOEUÛ ��Û /PUFÛ UIBUÛ �FV�Û QSPCBCMZÛ IBEÛCFDPNFÛ�PV��FBSMJFS ÛTPÛUIBUÛ�FH�ÛJTÛVOLOPXO�Û*ÛIBWFÛOPUÛGPVOEÛJOTUBODFTÛPGÛ�JH��ÛÛÛÛÛÛÛ*UÛ JTÛ OPUÛ DFSUBJOÛXIBUÛ IBQQFOFEÛXJUIÛXÛ CFGPSFÛ T�Û CPUIÛ HÛ BOEÛ LÛ IBWFÛCFFOÛQSPQPTFE�ÛÛÛ�, � FVT�T�JÛMPD��ÛBZHÛ�EBXO��ÛUIFÛXPSEÛXPVMEÛEFSJWFÛGSPNÛBOÛFYQSFTTJPOÛMJLFÛ�BUÛEBXO� Û OE�BZHÛXIJDIÛXPVMEÛBUÛUIFÛTBNFÛUJNFÛFYQMBJOÛUIFÛBCTFODFÛPGÛI�ÛBGUFSÛUIFÛQSFQPTJUJPO�ÛCVUÛBOPUIFSÛEFSJWBUJPOÛGSPNÛUIJTÛTUFNÛXPVMEÛBMTPÛCFÛQPTTJCMF�ÛÛÛ�, � FVT��ÛBHBOJNÛ�QBTTÛUIFÛOJHIU� ÛJGÛUIJTÛJTÛUIFÛDPSSFDUÛSFDPOTUSVDUJPO�ÛÛÛ�, � FVT�O��Û BLBO� �L Û �FBST� Û QM�Û PGÛVO�LOÛ �Û �, � VT�O��Û *OÛ UIJTÛ DBTFÛ UIFÛIÛ �Û TÛXPVMEÛIBWFÛEFWPJDFEÛUIFÛHÛ�Û �Û5IJTÛJTÛJNQPTTJCMFÛCFDBVTFÛJUÛXPVMEÛIBWFÛSFTVMUFEÛJOÛL �ÛÛÛÛÛÛÛÛÛÛÛÛCÛ�X��Û�X�XSJUUFOÛ�W�BGUFSÛPÛXPSE�GJOBMMZÛBGUFSÛWPXFM�ÛÛÛ�EJV�Û�ÛUJXÛ�EBZ��Û4LU�ÛEJW Û�CZÛEBZ�ÛÛÛ�OF, � X �Û�ÛOBXÛ�TIJQ��Û4LU�ÛOºV� �(S�ÛNAÀJ �-BU�ÛO WJT�ÛÛÛÛ�H PX��ÛLPWÛ�DPX��Û4LU�ÛHºW� �(S�ÛBOÀJ�ÛTFFÛBCPWFÛPOÛLPHJ�ÛÛÛÛ�, � SFX�Û�ÛBSFXÛ�TVO��Û4LU�ÛSBWÆ��ÛUIFÛGPSNÛXJUIÛHÛJOÛBSFH�BLOÛ�TVO� ÛMJUU�Û�TVO�TQSJOH�ÛXJUIÛBSFH�Û�ÛHFO�Û�BSFXPT ÛXIFSFÛUIFÛHÛXBTÛQSPUFDUFEÛCZÛUIFÛSFGMFYÛPGÛUIFÛT�ÛÛÛÛÛÛÛ5IFÛXÛPGUFOÛTQSFBEÛUISPVHIÛBOBMPHZ�ÛOBX�J�FUD� ÛEFSJW�ÛOBX�BL��CPBU��ÛÛÛÛÛÛÛÛDÛ�XÛ�ÛX�W�ÛJOÛPUIFSÛQPTJUJPOT ÛÛÛÛÛÛ5IJTÛEFWFMPQNFOUÛJTÛBTTVNFEÛGPSÛWB FNÛ�UPÛLJOEMF� Û-JUI�ÛWÅSUJ��UPÛDPPL� �


0$4ÛWBSJUJ�JE�Û*OJUJBMÛH�ÛJTÛTPÛGSFRVFOU ÛUIBUÛUIFÛFUZNPMPHZÛNVTUÛQSPCBCMZÛCFÛHJWFOÛVQ�ÛÛÛÛÛÛ7FD Û �TJY�Û CFMPOHTÛIFSFÛ JGÛ GSPNÛ UIFÛ T�MFTTÛ GPSNÛ�VF TÛ �TX��XPVMEÛIBWFÛHJWFOÛ �L ��Û *UÛ JTÛ FYQMBJOFEÛ GSPNÛ BÛ -JOEFNBOÛ GPSNÛ �TV FLT � PSÛ XJUIÛ V�JOUSPEVDFEÛ GSPNÛ UIFÛ PSEJOBMÛ �VLT�P�� ,PSUMBOEUÛ ��C�Û 5IFOÛ �TÛ CFDBNFÛ[FSP Û BOEÛ WPDBMJDÛ VÛ CFGPSFÛ WPXFMÛ XBTÛ SFUBJOFEÛ BTÛ �X�Û XSJUUFOÛ W�Û$PNQBSFÛ GPSÛ UIJTÛEFWFMPQNFOU�Û J�WFSPZÛ Û �VQPO�Û GSPNÛ�TVQFSJ�Û *OÛB CJXSÛ�TQSJOH�ÛGSPNÛ�CISF, � VS��CS XS�BOEÛBMJXSÛ�GMPVS�ÛGSPNÛ�, � MF, � VS��BM XSÛUIFÛXÛJTÛQSFTFSWFEÛEJSFDUMZÛCFGPSFÛUIFÛSÛBGUFSÛUIFÛMPTTÛPGÛUIFÛGJOBMÛTZMMBCMF�Û0OÛUIFÛHFO�ÛTH�ÛPGÛUIFTFÛGPSNTÛTFFÛUIFÛOFYUÛTFDUJPO ÛE�ÛÛÛÛÛÛ4JXOÛ�QJMMBS�Û(S�ÛK WN Û.ZD�ÛLJXP��L X O��ÛUIFÛXPSEÛJTÛOPXÛTIPXOÛUPÛCFÛ*OEP�&VSPQFBO� IBTÛ OPUÛ QSFTFSWFEÛ UIFÛ X�Û *UÛ EFWFMPQFEÛ GSPNÛ UIFÛBDDVTBUJWFÛ�L XPO��TJXVO��TJVOÛ�ÛTJXOÛV�XSJUUFOÛX�Û5IJTÛJTÛDPOGJSNFEÛCZÛKJXOÛ�TOPX� ÛXIJDIÛEFSJWFTÛGSPNÛ� IJPO��KJVOÛXSJUUFOÛKJXO�Û�Û��ÛÛÛÛÛÛEÛ�XÛ�Û[FSPÛCFGPSFÛS ÛCFGPSFÛBOEÛBGUFSÛVÛÛÛÛÛ5IJTÛEFWFMPQNFOUÛDBOÛCFÛDPOTJEFSFEÛSFHVMBSÛCFGPSFÛVÛBOEÛS ÛBTÛ UIJTÛJTÛQIPOFUJDBMMZÛQSPCBCMF�ÛÛÛÛÛÛ�OFV�SP��XJUIÛ�SP�ÛSFQMBDJOHÛ�P��ÛOPSÛ�OFX�ÛWJBÛ�OPV�SP��ÛÛÛÛÛ�CSFXS�PT Û UIFÛ SFTIBQFEÛ HFO�Û TH�Û PGÛ B CJXSÛ �TQSJOH�Û TFFÛ UIFÛ QSFDFEJOHÛTFDUJPOÛ�ÛB CFS�ÛJOÛUIFÛTBNFÛXBZÛBMFSÛGSPNÛBMJXS�ÛÛÛÛÛ5IJTÛ SVMFÛ XBTÛ CMPDLFEÛ CZÛ BÛ QSFDFEJOHÛ DPOTPOBOUÛ XIFSFÛ XFÛ GJOEÛOPSNBMÛH�ÛÛÛÛ�EF, � JV S� �� UBZHSÛ �IVTCBOE�TÛCSPUIFS��Û 0OFÛNJHIUÛBMTPÛ GPSNVMBUFÛ �ZX��ZH�ÛÛÛÛÛÛ5IFÛMPTTÛPGÛUIFÛXÛJOÛOFBSEÛ�GJCSF ÛTJOFX�ÛGSPNÛ�TOF, � �UJ��JTÛBOBMPHJDBMÛBGUFSÛUIFÛPMEÛPCMJRVFÛDBTFT�Û�TOF, � �VS��Û�TO XS� ÛXIJDIÛMPTUÛJUTÛXÛKVTUÛMJLFÛB CFSÛBCPWF�ÛÛÛ� J) PO��Û TJXOÛ �QJMMBS��Û �TJXVO�Û �Û �TJVOÛ �Û TJXO�Û 5IFÛ EFWFMPQNFOUÛ JTÛQBSBMMFMÛUPÛ� IJPN�Û�Û�KJVOÛ�ÛKJXO ÛXIJDIÛTIPXTÛUIBUÛUIFÛXÛJOÛTJXOÛEPFTÛOPUÛDPOUJOVFÛUIFÛPSJHJOBMÛ�X�ÛÛÛÛ'PSÛ UIFÛ QPTJUJPOÛ BGUFSÛ VÛ DPNQBSFÛ BMTPÛ [JOVPSÛ �TPMEJFS�Û �Û �[JOV�XPSÛÛ�CFBSJOHÛBSNT�ÛXÛ�Û�CI�Û�'VSUIFSÛDQ�ÛB V TÛ�GPY� Û(S�Û�LW�PHX�ÛÛÛÛÛ*OÛJOOÛ �OJOF�ÛUIFSFÛXBTÛOPÛX�Û5IFÛ1*&ÛXPSEÛXBTÛ�, � OFVO�Û5IJTÛCFDBNFÛ�FOFVO�� JOO ÛXJUIÛ MPTTÛPGÛ UIFÛEJQIUPOHÛ FVÛ JOÛ MBTUÛ TZMMBCMFÛ OPUFÛ UIBUÛ UIFÛWPDBMJ[BUJPOÛ JTÛ BÛ QPTU�1*&Û QIFOPNFOPO�Û ,PSUMBOEUÛ Q�D�Û 0SÛ �JOFVOÛ XBTÛSFQMBDFEÛCZÛ �JOVO�Û GSPNÛ UIFÛPSEJOBMÛ �JOVO�P��5IFÛ GPSNÛ FSLBO� �NJMMTUPOF�ÛEJEÛOPUÛIBWFÛX ÛDG�Û-JUI�ÛHÅSOB��H SF, � �O��HSBO��ÛÛÛÛÛÛÛFÛÛ$MVTUFSTÛXJUIÛ�X�ÛÛ'PSÛUIFÛDMVTUFSTÛ�TX Û�UX Û�EX �� X�TFFÛ��ÛÛÛ�

166� 5��6��3��%HHNHV��

� ��6)732%287�ÛÛ ��ÛÛ4ZMMBCJDÛSFTPOBOUTÛÛ ��ÛÛÛ1*&Û� Û�ÛBS�ÛÛÛÛÛ�QS �TL��ÛIBSD �BOFNÛ�BTL��Û4LU�ÛQ DDI� Û-BU�ÛQPTD Û�Û�QPSD�TD Û0)(�GPSTL O�ÛÛÛ�CIS IV��ÛCBSK�SÛ�IJHI��Û4LU�ÛC I�ºOU� Û)JUU�ÛQBSLVT�ÛÛÛ�TSCI��BSC�FOBNÛ�ESJOL��Û-BU�ÛTPSCF Û-JUI�ÛTVSCJÒ�ÛÛÛ�) � SOV��ÛB OVNÛ�UBLF��Û(S�Û��RNUMAIÛÛ�XJO ÛHBJO� Û"W�Û S OV�Û�QSPWJEF��ÛÛÛ�NSUP��ÛNBSEÛ�NBO��Û4LU�ÛN Uº�Û�NPSUBM� Û(S�ÛBROTO�J�ÛÛÛ ��ÛÛ1*&Û� Û�ÛBMÛÛÛÛÛÛÛ�HMLU�T Û�NÛOPN� ÛBDD��Û�LB D Û�ÛEJBM�ÛLBYD ÛSFTQ�Û�LB U OÛ�Û$MBTT�ÛLBU OÛ�NJML��Û(S�ÛGA�LA ÛHFO�ÛGA�LAKTOJÛÛ�Û�HMBLU� Û-BU�ÛMBD �MBDU�JT�Û5IFÛEJBMFDUBMÛGPSNÛJT ÛPGÛDPVSTF ÛQPTU�$MBTTJDBMÛ,PSUMBOEUÛ��D Û��Û<UIJTÛWPM� Û��>�ÛÛÛ ��ÛÛ1*&Û� Û�ÛBO�ÛÛÛÛÛ� ��ÛBO�Û�VO��ÛF�H�ÛBOLJOÛ�XJUIPVUÛXJGF��Û4LU�ÛBO� Û(S�ÛA��N Û-BU�ÛJO� Û(PUI�ÛVO��ÛÛÛ ��Û Û 1*&Û � Û �Û BN� Û Û $FSUBJOÛ FYBNQMFTÛ IBWFÛ GJOBMÛ � Û XIJDIÛCFDBNFÛ�O�ÛÛÛ�TFQUN�ÛFXU OÛ�TFWFO��Û4LU�ÛTBQUº Û(S�ÛxPTA� Û-BU�ÛTFQUFN�ÛÛÛ�EF N�ÛUBTOÛ�UFO��Û4LU�ÛEº B Û(S�ÛDzKA Û-BU�ÛEFDFN�ÛÛÛ�QPE�NÛBDD��ÛPUOÛ�GPPU��Û4LU�ÛQ E� Û(S�ÛPO�D Û-BU�ÛQFE� Û0)(ÛGVP[�ÛÛÛ�EVJ�E NUJ�ÛL TBOÛ�UXFOUZ��Û"W�ÛW TBJUJ Û(S�Û%PS�ÛðI�KATI Û-BU��W HJOU Û�ÛÛÛ ��ÛÛ$POTPOBOUBMÛSFTPOBOUTÛÛÛ ��ÛÛ1*&Û�S��S��S�� CFGPSFÛOÛÛÛÛÛÛ1*&ÛEJEÛOPUÛIBWFÛXPSETÛCFHJOOJOHÛXJUIÛ S��Û8PSETÛ UIBUÛIBEÛ �QS� � �US�ÛMPTUÛUIFÛTUPQ�ÛUIFÛSFNBJOJOHÛ�S�ÛHPUÛBÛQSPUIFUJDÛWPXFMÛF��PSÛB��ÛTFFÛDI�Û��ÛrÛ0OÛ�)S��TFFÛ��ÛrÛ0OÛ�ST�ÛTFFÛUIFÛOFYUÛTFDUJPO Û��ÛÛÛ�, � FS, � USPN� �ÛBSBXS� �QMPVHI��Û(S�Û�ROTRON �0*S�ÛBSBUIBS �0*D�ÛBSÉS�Û-BU�ÛBS USVN��ÛÛÛÛÛ� FS, � �P�Û�ÛDFS��PMEÛNBO��Û(S�ÛGE�RWN �4LU�ÛKºSBOU��ÛÛÛ�EP, � SPNÛ�ÛUVSÛ�HJGU��Û(S�ÛDÎRON �0$4ÛEBS ��ÛÛÛÛÛ�EIVS�N� �Û EV O� �EPPS�Û GSPNÛ UIFÛ BDD�Û GPSN�Û (S�Û KU�R Û -BU�Û GPS T Û (PUI�ÛEBVS��ÛÛÛ�QFSVUJÛ�ÛIFSVÛ�MBTUÛZFBS��Û(S�ÛPE�RUSI Û4LU�ÛQBSÓU��ÛÛÛÛÛ�QSF, � UPTÛ�ÛFSBTUBOL Û�CBDL ÛBOVT��Û(S�ÛPRWKTO�J�ÛÛÛ�USFZFTÛ�ÛFSFL Û�UISFF��Û4LU�ÛUSºZBT �(S�ÛTREkJ Û-BU�ÛUS T�ÛÛ


Û ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛS�ÛTFFÛ��ÛÛÛÛÛÛ ��ÛÛÛ1*&Û�M��M��M�� CFGPSFÛDPOTPOBOUÛFYDFQUÛ�Z��ÛÛÛÛÛ�MFVLPTÛ�ÛMPZTÛ�MJHIU�ÛTVCTU��Û(S�ÛLEUKO�J Û4LU��SPLº��ÛÛÛ�MJL �Û �Û F�MJL Û �IFÛ MFGU�Û �Û TH�Û BPS��Û ML BOFNÛ �MFBWF�Û QSFT��Û (S�Û | LIP E � 4LU�Ûº�SJD�BU Û(PUI�ÛMFJ BO��ÛÛÛÛÛ�, � FM�FO�Û�ÛF OÛ�EFFSÛDPX��Û(S�Û|NELOJÛXJUIÛNFUBUIFTJT �yLLO�J��Û�FM�OP� Û0$4ÛKFMFO ��ÛÛÛÛÛ�NFMJ��ÛNF S��IPOFZ��Û(S�ÛME�LI Û-BU�ÛNFM Û(PUI�ÛNJMJ×��ÛÛ�QMF, � � ��UP�Û�ÛMOVNÛ�GJMM� ÛMJÛ�GVMM��Û4LU�ÛQS Uº� �-BU�ÛDPNQM UVT�Û��ÛÛÛÛÛ� MVUPTÛ�ÛMVÛ�GBNPVT��Û(S�ÛKLUTO�J �4LU�Û SVUº� �-BU�ÛJODMVUVT��ÛÛÛ ��ÛÛ/BTBMT�Û1*&Û�OÛÛÛÛÛ�B��O��O��ÛÛÛÛÛ�OJTEPTÛ�ÛOJTUÛ�QPTJUJPO ÛTFBU��Û4LU�ÛO Eº� �-BU�ÛO EVT �Û&OH�ÛOFTU�ÛÛÛ�TOVTÌTÛ �Û OV� �EBVHIUFS�JO�MBX��Û (S�Û NUO�J � 4LU�Û TOV Û -BU�Û OVSVT Û 0)(ÛTOVS��ÛÛÛÛÛ� FO, � PTÛ�ÛDJOÛ�CJSUI��Û(S�ÛGE�NOJ �4LU�ÛKºOBT Û-BU�ÛHFOVT��ÛÛÛÛÛ�, � O IV�Û�ÛBOKVLÛ�OBSSPX��Û4LU�ÛBNIÓ� �0$4ÛP[ L �(PUI�ÛBHHXVT��ÛÛÛ�QFOL FÛ�ÛIJOHÛ�GJWF��Û(S��PE�NTE �4LU�ÛQºÊDB��ÛÛÛÛCÛ�O��[FSPÛÛCFGPSFÛT�ÛBUÛUIFÛFOEÛPGÛBÛXPSEÛBGUFSÛWPXFMÛBUÛBOÛFBSMZÛTUBHFÛFYDFQUÛJOÛNPOPTZMMBCMFT�ÛTFFÛPOÛUVOÛCFMPX Û��ÛÛÛÛÛ��NF, � OT��Û BNJTÛ �NPOUI�Û GSPNÛ �BN�NJT� �NPOUIÛ PGÛ BÛ ZFBS��Û (S�Û MEI�JÛ �Û�N OT �-BU�ÛN OTJT�ÛÛÛ��ÛÛÛÛÛ�USJOTÛ�ÛFSJTÛBDD�ÛQM�Û�UISFF��Û(PUI�Û×SJOT��ÛÛÛÛÛÛÛ*UÛJTÛBTTVNFEÛUIBUÛ�O �MJLFÛ�N �XBTÛMPTUÛBGUFSÛBÛWPXFM�Û5IJTÛFYQMBJOTÛUIBUÛUIFÛ OPNJOBUJWFÛ TJOHVMBSÛ PGÛ UIFÛO�TUFNTÛ PSJHJOBMMZÛ IBEÛ OPÛ �O� BTÛ JOÛ B JL��HJSM� Û NBOVL� �DIJME� Û XIFSFÛ UIFÛ BCTFODFÛ PGÛ BOÛ �OÛ JOÛ UIFÛ OPNJOBUJWFÛ JTÛPUIFSXJTFÛIBSEÛUPÛFYQMBJO�ÛFS D Û�QSJFTU� ÛXIJDIÛJTÛBOÛV�TUFNÛJOÛUIFÛTJOHVMBSÛBOEÛ BOÛO�TUFNÛ JOÛ UIFÛ QMVSBM Û DBOÛ UIVTÛ CFÛ FYQMBJOFEÛ GSPNÛ BÛ OPN�Û TH�Û � ÛXJUIPVUÛ�O ÛOPN�ÛQM�Û�PO�FT��5IFÛ�OÛXIJDIÛXFÛOPSNBMMZÛGJOEÛJOÛUIFÛOPN�ÛTH�Û IBTÛ CFFOÛ SFTUPSFE�Û5IFÛ TPVSDFÛXBTÛ UIFÛ BDDVTBUJWFÛ FOEJOHÛ � �XIJDIÛCFDBNF� �O�� 5IJTÛ JTÛ OPÛ EPVCUÛ UIFÛ SFBTPOÛ XIZÛ SPPUÛ OPVOTÛ MJLFÛ PUOÛ �GPPU�ÛKPJOFEÛUIFÛO�TUFNT�ÛrÛ"ÛXPSEÛGJOBMÛOBTBMÛBSPTFÛJOÛUIFÛMBOHVBHFÛGSPNÛUIFÛTZMMBCJDÛ OBTBMT Û �� XIJDIÛ GFMMÛ UPHFUIFSÛ JOÛ �O�Û 4FFÛ BMTPÛ POÛ �N�Û,PSUMBOEUÛ��D�ÛÛÛ ��ÛÛ/BTBMT�Û1*&Û�N�ÛÛÛÛÛBÛ�N��N�ÛFYDFQUÛJOÛBVTMBVU ÛCFGPSFÛT �BOEÛCFGPSFÛOÛÛÛÛ�NF, � U SÛ�ÛNBZSÛ�NPUIFS��Û(S�ÛM THR �MH�THR �4LU�ÛN UºS� �-BU�ÛN UFS��Û

168� 5��6��3��%HHNHV��

ÛÛ�, � NPJHIPTÛ�ÛN HÛ�NJTU ÛGPH��Û4LU�ÛNFHIº� �(S�ÛO�MI�CLH �-JUI�ÛNJHM¹��ÛÛÛÛÛ�, � NFJ I�Û �ÛNJ[FNÛ �VSJOBUF��Û 4LU�ÛNÂIBUJ Û (S�Û O�MEI�CW � -BU�ÛNFJ ��0SÛ BSFÛUIFTFÛ*SBOJBOÛMPBOXPSET �ÛÛÛÛÛ�, � NPTÛ�ÛJNÛ�NJOF��Û(S�ÛyMO�J��ÛÛÛÛÛÛÛÛÛÛÛÛCÛ�NÛ�Û[FSPÛCFGPSFÛTÛÛÛÛ�, � PNTPTÛ�ÛVTÛ�TIPVMEFS��Û(S�ÛÐMOJ �4LU�ÛºNTB� �(PUI�ÛBNT��ÛÛÛÛÛ�N NTP�ÛPSÛ�NFNTP��ÛNJTÛ�NFBU��Û4LU�ÛN NTºN �0$4ÛN TP Û(PUI�ÛNJN[��ÛÛÛÛÛÛD��N��O��XPSE�GJOBMMZ ÛSFUBJOFEÛJOÛNPOPTZMMBCMFT ÛBOEÛCFGPSFÛ�TÛBOEÛKÛÛÛÛ�L F, � NÛ�ÛL BO��BT�Û��ÛÛÛÛÛ�E NÛ�ÛUVOÛ�IPVTF��Û4LU�ÛEºN� Û(S�ÛDÎ��DW>MA��ÛÛÛÛ� IJPN�N�BDD�Û�ÛKJXO��TOPX�ÛUISPVHIÛ�KJVNO�Û(S�ÛCIW�N �"W�Û[ZBN��ÛÛÛÛÛ�, � NF�Û�ÛEBU�Û JOKÛ �UPÛNF��Û GSPNÛIFSFÛ UIFÛOÛXBTÛFYUFOEFEÛ UPÛPUIFSÛ GPSNT�ÛJO O � JOFXÛ UIFÛNÛ JTÛQSFTFSWFEÛJOÛUIFÛHFO�Û JN�Û UIFÛNÛNBZÛBMTPÛIBWFÛCFFOÛBTTJNJMBUFEÛUPÛOÛCFGPSFÛT�ÛBDD�ÛMPD�ÛPGÛ�* ÛNF�ÛJTÛ�Û�JOT��JN�T�Û5IJTÛQSPDFTTÛGVSUIFSÛQFSIBQTÛ JOÛ JO Û �TPNFUIJOH� Û �ZJN�L JE Û UIPVHIÛIFSFÛ UIFÛ �OÛNBZÛIBWFÛ BSJTFOÛ JOÛ XPSEÛ GJOBMÛ QPTJUJPO�Û UIJTÛ JTÛ QSPCBCMFÛ BTÛ PUIFSXJTFÛ UIFÛDPOTPOBOUÛBGUFSÛOÛXPVMEÛIBWFÛCFFOÛWPJDFE ÛHJWJOHÛ�JO �ÛÛÛÛÛÛÛÛEÛ�NÛ��[FSPÛXPSE�GJOBMMZÛFYDFQUÛJOÛNPOPTZMMBCMFTÛTFFÛD��BCPWFÛÛÛÛÛÛ8PSE�GJOBMÛ �NÛ GFMMÛ UPHFUIFSÛXJUIÛ �O Û CVUÛXFÛ EPÛ OPUÛ LOPXÛ FYBDUMZÛ BUÛXIBUÛ TUBHF�Û-BUFSÛ JUÛEJTBQQFBSFEÛ QSPCBCMZÛ UISPVHIÛBÛOBTBMJ[FEÛWPXFM�ÛÛ5IFÛBDDVTBUJWFTÛPGÛUIFÛWPXFMÛTUFNTÛ��PN ��JN ��VNÛEJTBQQFBSFEÛFOUJSFMZ ÛBOEÛ TPÛ GFMMÛ UPHFUIFSÛ XJUIÛ UIFÛ OPNJOBUJWFTÛ BGUFSÛ UIFÛ �T Û IBEÛ CFFOÛSFNPWFE�ÛTFFÛ��ÛÛÛÛÛÛÛÛFÛ�NO��XO�BGUFSÛWPXFM�ÛJGÛBOÛPÛQSFDFEFE ÛUIJTÛCFDBNFÛV�GJSTU�ÛOPUÛBGUFSÛPMEÛV�BOEÛJOÛNPOPTZMMBCMFT�ÛÛÛÛÛÛÛÛ5IFÛ QSPDFTTÛ XBTÛ QSPCBCMZÛ SFDFOUÛ BTÛ JUÛ JTÛ TUJMMÛ QSPEVDUJWF�Û QBuUBXO��XPSTIJQ�ÛDPNFTÛGSPNÛ�NO� �BTÛBQQFBSTÛGSPNÛUIFÛ(%-�ÛQBuUBNBO�ÛBOEÛUIFÛXPSEÛ JTÛ BOÛ *SBOJBOÛ MPBO�Û "MTPÛ HP JXOÛ �TIPVUJOH Û DSZ�Û GSPNÛ �JNO Û CFTJEFÛ(%-ÛHP NBO�ÛÛÛÛ��P�N, � OPT�Û�VOÛGSPNÛ��PNOPTÛ�Û�VNOPTÛ ÛWFSCBMÛBEKFDUJWF��ÛÛÛÛÛÛÛÛ"GUFSÛ V � �NO�Û XBTÛ SFTUPSFEÛ SBUIFSÛ UIBOÛ QSFTFSWFE�Û TBS VNO��NPWFNFOU��Û *OÛNPOPTZMMBCMFTÛ�NOÛXBTÛBMTPÛQSFTFSWFEÛPSÛSFTUPSFE�ÛIJNO��GPVOEBUJPO� ÛLBNOÛ�GMBJM�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛGÛ�N��XÛCFGPSFÛ1"SN�Û�V�ÛÛÛÛÛ�, � OF, � NO�ÛBOVOÛ�OBNF��ÛPO NOÛ�ÛBOVNOÛ�Û�BOVXOÛ�ÛBOVO�Û4LU�ÛO NB Û(S�ÛŃOMA Û-BU�ÛO NFO�Û�ÛÛÛ


ÛÛ�, � F, � N S�Û BXS� �EBZ�Û �Û �BNVS�Û (S�Û £MAR��H@ME�RA�Û 5IJTÛ GPSNÛ JTÛ BMTPÛFYQMBJOFEÛXJUIÛX�FQFOUIFTJT Û�BNVSÛ�Û�B NSÛ�ÛBXS ÛQBSBMMFMÛUPÛ�BOJSÛ�ÛB � OSÛ�ÛBZSÛ�NBO��Û)PXFWFS ÛUIFÛSVMFÛ�NV�Û�Û�V�ÛXBTÛQSPCBCMZÛPMEFSÛUIBOÛUIFÛMPTTÛPGÛGJOBMÛTZMMBCMFT��ÛÛÛ��PNPN�Û JOÛ �VL Û �ÛQM�Û FOEJOHÛPGÛ UIFÛ BPS�Û TVCK�Û,PSUMBOEUÛ ��D Û ��Û <UIJTÛWPM� Û��G�>ÛBTTVNFTÛUIBUÛUIJTÛGPSNÛCFDBNFÛ�VNVÛ�Û�VVÛ�ÛV ÛBGUFSÛXIJDIÛUIFÛQMVSBMÛJOEJDBUJPOÛ�L Û�Û��TÛXBTÛBEEFE��Û ÛÛ ��4-)��W�ÛÛ ��ÛÛ1*&Û�TÛÛÛÛ�TÛÛ�ÛTÛÛCFGPSFÛQ �U �Y�ÛPOÛTLÛTFFÛPOÛDMVTUFSTÛ��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛBOEÛBGUFSÛO �NÛXIJDIÛEJTBQQFBSÛÛÛÛÛÛÛÛÛÛ�ÛÛI�ÛJOÛBOMBVUÛQSFTFSWFEÛCFGPSFÛJÛÛÛÛÛÛÛÛÛÛ�ÛÛ�L ÛÛXPSE�GJOBMMZÛÛÛÛÛÛÛÛÛÛ�ÛÛ[FSPÛFMTFXIFSFÛÛÛÛÛÛÛÛÛÛÛÛ�TÛJTÛQSFTFSWFEÛCFGPSFÛQ �U �Y�ÛBOEÛBGUFSÛO �N�ÛJUÛCFDPNFTÛ[FSPÛJOÛBOMBVUÛCVUÛ I�Û CFGPSFÛ J Û CFUXFFOÛ WPXFMT Û BOEÛ CFGPSFÛ M � O � N�Û XPSE�GJOBMMZÛ JUÛCFDBNFÛ�Û��IÛ�Û�� Û�Û�L ��ÛÛÛÛÛÛBÛÛ�T��T�ÛCFGPSFÛQ �U �YÛBOEÛBGUFSÛO �N �XIJDIÛEJTBQQFBS��ÛÛÛÛ�TUFS�JP�Û �ÛTUFS ��TUFSJMF��Û(S�ÛSTEkRA �4LU�ÛTUBS � �-BU�ÛTUFSJMJT �(PUI�Û�TUBÆSP��ÛÛÛÛÛ�TUJC�Û�ÛTUJQFN��ESBOHFO��Û(S�ÛSTEI�BWÛ�USFBE ÛTUBNQÛPO��ÛÛÛÛ�, � TU SÛ�ÛBTU ��TUBS��Û(S�ÛA�STH�R �4LU�ÛTUºS� �-BU�ÛTU MMB�ÛÛÛÛ�VFT�UV��Û[�HFTU��DMPUIJOH��Û-BU�ÛWFTUJT �(PUI�ÛXBTUJ��ÛÛÛ�TQSOF, � NJÛ�ÛTQB OBNÛ�NFOBDF��Û-BU�ÛTQFSO �0*D�ÛTQFSOB�CPUI��QVTIÛBXBZ��ÛÛÛÛ�TL, � FM� Û�ÛTYBMFNÛ�TUVNCMF ÛFSS��Û4LU�ÛTLIBMBUF��ÛÛÛÛ�USJOTÛ�ÛFSJTÛBDD�Û�UISFF��Û(PUI�Û×SJOT��ÛÛÛÛÛ�NF, � OTPTÛ�ÛBNJTÛ�NPOUI�ÛWJBÛ�N OTPTÛGSPNÛ�BN�NJT��NPOUIÛPGÛUIFÛZFBS��Û(S�ÛMEI�J �4LU�ÛN T��NBBT��NF, � T� �-BU�ÛN OTJT�ÛÛÛÛ�, � PNTPTÛ�ÛVTÛ�TIPVMEFS��Û(S�ÛÐMOJ �4LU�ÛºNTB� �(PUI�ÛBNT��ÛÛÛÛÛ�N NTPNÛ�ÛNJTÛ�NFBU��Û4LU�ÛN NTºN �0$4ÛN TP Û(PUI�ÛNJN[��ÛÛÛÛÛÛÛÛCÛÛ�T��ÛI�ÛJOÛBOMBVUÛCFGPSFÛJ�ÛÛÛÛÛ�TFO��ÛIJOÛÛ�PME��Û4LU�ÛTºOB� Û(S�Û{NOJ Û-JUI�ÛT OBT�Û/PUFÛUIBUÛUIFÛSVMFÛPGÛMPTTÛPGÛ�IÛ�Û�TÛPQFSBUFEÛBGUFSÛF��JÛCFGPSFÛOBTBM�ÛÛÛÛÛÛ/PUFÛUIBUÛI�Û�HPPE�ÛI�[BXSÛ�TUSPOH�Û�Û�IBWJOHÛHPPEÛGPSDF�ÛJTÛQSPCBCMZÛBÛMPBOÛGSPNÛ*SBOJBOÛIV��ÛÛÛÛÛÛÛDÛÛ�TÛ�Û�L ÛÛXPSE�GJOBMMZÛÛÛ��FTÛQMVS�ÛFOEJOH�Û�FL �ÛUIFÛWPXFMÛJTÛQSFTFSWFEÛJOÛ PSFL UBTBOÛ�GPVSUFFO��ÛÛÛ��NFT Û�UFT�Û�NL ��ZL ÛÛ� Û�ÛQM�ÛFOEJOHT�ÛÛÛ5IJTÛEFWFMPQNFOUÛIBTÛTVSQSJTFEÛTDIPMBST ÛCVUÛUIFSFÛBSFÛPUIFSÛJOTUBODFTÛXIFSFÛ�I Û�YÛCFDBNFÛ�LI ÛBTÛJOÛ4PVUIÛ1PMJTIÛEJBMFDUT Û4FSCP�$SPBUJBOÛBOEÛ

170� 5��6��3��%HHNHV��

(FSNBOÛEJBMFDUT�ÛÛÛÛÛÛ8IFOÛ UIFÛ �L Û DBNFÛ UPÛ CFÛ VTFEÛ GPSÛ UIFÛ OPNJOBMÛ QMVSBMÛ FOEJOH Û UIFÛBDDVTBUJWFÛTH�ÛXBTÛTVCTUJUVUFEÛGPSÛUIFÛOPN�ÛTH�Û*OÛPUIFSÛQPTJUJPOTÛUPPÛUIFÛSFGMFYÛPGÛ��TÛXBTÛSFNPWFE ÛBTÛJOÛUIFÛHFO�ÛTH�ÛJOÛ��PT�Û5IFÛPCKFDUJPOÛUIBUÛUIFÛBUUSJCVUJWFÛBEKFDUJWFÛIBTÛBÛ[FSPÛFOEJOHÛ JOÛ UIFÛOPNJOBUJWFÛQMVSBM Û DBOÛCFÛSFNPWFEÛCZÛBTTVNJOHÛUIBUÛUIFÛBEKFDUJWFÛUPPLÛQSPOPNJOBMÛFOEJOHT ÛBTÛF�H�ÛJOÛ(FSNBOJD�ÛUIJTÛFOEJOHÛEJTBQQFBSFEÛQIPOFUJDBMMZ�ÛÛÛÛÛÛ5IJTÛ FYQMBOBUJPOÛ BMTPÛ BDDPVOUTÛ GPSÛ UIFÛ MBSHFÛ OVNCFSÛ PGÛ QMVSBMJBÛUBOUVNÛJOÛ"SNFOJBOÛ,PSUMBOEUÛ��D�ÛÛÛÛÛÛEÛÛ�TÛ�Û[FSPÛFMTFXIFSF ÛJ�F�ÛJOÛBOMBVUÛFYDFQUÛCFGPSFÛJ ÛCFUXFFOÛWPXFMT ÛCFGPSFÛM �O �NÛPOÛTS �STÛTFFÛ��ÛCFMPXÛPOÛDMVTUFSTÛÛÛ�TFQUN��FXU OÛ�TFWFO��Û4LU�ÛTBQUº Û(S�ÛxPTA� Û-BU�ÛTFQUFN�ÛÛÛ�TN)��Û BNÛ �ZFBS��Û 4LU�Û TºN� Û 0*S�Û TBN�Û DG�Û �TN, � FS��Û BNB �OÛ �TVNNFS��Û0)(�TVNBS�ÛÛÛ�TSCI�Û�ÛBSCJÛ�*ÛESVOL��Û-BU�ÛTPSCF �Û-JUI�ÛTVSCJÒ�ÛÛÛÛÛ�CIPTP��ÛCPLÛ�CBSFGPPU�Û�Û�CIPTP�H P��Û-JUI�ÛC¼TBT Û0$4ÛCPT Û0)(ÛCBS�ÛÛÛ�TVFT S�ÛL PZSÛ�TJTUFS�Û�Û��FIVS�Û4LU�ÛTWºTBS� ÛÛ-BU�ÛTPSPS Û(PUI�ÛTXJTUBS�ÛÛÛ�TOVTPT�ÛOVÛ�EBVHIUFS�JO�MBX��Û4LU�ÛTOV Û(S��NUO�J Û-BU�ÛOVSVT Û0)(ÛTOVS�ÛÛÛ�TNJ, � ��ÛNJÛÛ�POF��Û(S�ÛMI�A�ÛDG�Û-BU�ÛTFN�FMÛ�POF��ÛÛÛ�, � FTNJ�ÛFNÛ�*ÛBN��Û4LU�ÛºTNJ Û(S�ÛEI�MI� Û0$4ÛKFTN �ÛÛÛÛÛ�JP�TN��Û�VNÛQSPO�ÛEBU��MPD��Û4LU�Û�TNBJ Û(PUI�Û× B�NNB�ÛÛÛ�VFT�OPN�ÛHJOÛ�QSJDF��Û4LU�ÛWBTOºN Û-BU�ÛW OVN�Û5IFÛSFDPOTUSVDUJPOÛ�V TOPNÛJTÛJNQSPCBCMF�ÛTFFÛUIFÛGPMMPXJOHÛÛÛ�VFT�OV��Û[�HFOVNÛ �DMPUIFÛPOFTFMG��Û4LU�ÛWBT� Û(S�Û{NN MI�Û"TÛBQQFBSTÛGSPNÛUIFÛQSFDFEJOHÛUXPÛFUZNB ÛUIFÛEFWFMPQNFOU�T/��/ÛQSFDFEFEÛF/��J/ �P/��V/ ÛTPÛJOÛ[�HFOVNÛUIFÛWPXFMÛNVTUÛIBWFÛCFFOÛSFTUPSFE�ÛÛÛÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛ�TÛTFFÛ��ÛÛÛÛ ��78347�� Û5IFÛ"SNFOJBOÛDPOTPOBOUÛTIJGU�ÛÛÛÛÛÛ'PSÛ1*&ÛXFÛSFDPOTUSVDUÛXJUIÛUIFÛEFOUBMTÛBTÛFYBNQMFÛGPSÛUIFÛTUPQT�ÛÛ �U� �E� �EIÛ Û XIJDIÛ BSFÛ SFQSFTFOUFEÛ JOÛ"SNFOJBOÛBT�ÛÛ ÛU � �U� �EÛ Û 5IFÛ TFDPOEÛ DPMVNO Û UIFÛ 1*&Û�WPJDFEÛ TUPQT�ÛXFSFÛ JOÛ GBDUÛHMPUUBMJ[FE�Û5IFJSÛWPJDFÛXBTÛOPUÛ UIFÛFTTFOUJBMÛGBDUPS ÛCVUÛUIFZÛNVTUÛIBWFÛCFFOÛrÛQIPOFUJDBMMZÛrÛMFOFT�ÛUIFÛHMPUUBMJ[BUJPOÛXBTÛUIFÛQIPOFNJDBMMZÛEJTUJOHVJTIJOHÛGFBUVSF�Û/PXÛUIJTÛHMPUUBMJ[BUJPOÛIBTÛCFFOÛ QSFTFSWFEÛ JOÛ "SNFOJBOÛ VQÛ UPÛ UIFÛ QSFTFOUÛ EBZ Û BTÛ JOÛ -BUWJBOÛ,PSUMBOEUÛ��B�Û��D�ÛBMTPÛ��B�ÛÛÛÛÛÛÛ5IFÛ DPOTPOBOUÛ TIJGUÛ DPOTJTUFEÛ JOÛ UIFÛ GPMMPXJOHÛ EFWFMPQNFOUT�Û �Û


XFBLFOJOHÛ PGÛ UIFÛ QMBJO Û WPJDFMFTTÛ TUPQT�Û �Û EFWPJDJOHÛ PGÛ UIFÛ �WPJDFE� ÛHMPUUBMJ[FEÛTUPQT�Û�ÛMPTTÛPGÛBTQJSBUJPOÛJOÛUIFÛBTQJSBUFEÛTUPQT�Û5IFÛSFMBUJWFÛDISPOPMPHZÛPGÛUIFTFÛDIBOHFTÛJOÛ"SNFOJBOÛBDDPSEJOHÛUPÛ,PSUMBOEUÛXBTÛBTÛGPMMPXT�ÛÛÛBÛMPTTÛPGÛBTQJSBUJPOÛPGÛUIFÛBTQJSBUFEÛTUPQT�ÛTUBHFÛ��ÛÛÛCÛEFWPJDJOHÛPGÛUIFÛHMPUUBMJ[FEÛTUPQT�ÛÛÛDÛWPJDJOHÛPGÛQMBJO ÛWPJDFMFTTÛTUPQTÛBGUFSÛSFTPOBOU�ÛÛÛÛEÛMFOJUJPOÛPGÛUIFÛQMBJO ÛWPJDFMFTTÛTUPQT�Û�Q��U��LÛ�Û�FÛ� Û� ÛTUBHFÛ��ÛÛÛFÛ�TIPSUFOJOH�ÛPGÛUIFTFÛGSJDBUJWFTÛUPÛQ �U �L ÛTUBHFÛ��ÛÛÛ ��ÛÛ5IFÛMBCJBMTÛÛ ��Û1*&Û�Q�ÛÛÛÛÛÛÛÛBÛÛ�QÛÛ�ÛI�ÛJOÛBOMBVUÛCFGPSFÛWPXFMÛ�Û[FSPÛCFGPSFÛPÛÛÛÛÛÛCÛÛÛÛÛÛÛÛ�Û[FSPÛÛÛJOÛBOMBVUÛCFGPSFÛPÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛJOÛBOMBVUÛCFGPSFÛDPOTPOBOUÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛBGUFSÛOBTBMÛÛÛÛÛÛDÛÛÛÛÛÛ�ÛXÛÛÛÛFMTFXIFSF ÛJ�F�ÛBGUFSÛWPXFMÛPÛ�ÛXÛCFGPSFÛDPOTPOBOUÛ�ÛV�ÛJÛ�ÛXÛCFGPSFÛWPXFMÛHJWFTÛVÛÛÛÛÛÛBÛÛ�Q��I�ÛÛJOÛBOMBVUÛCFGPSFÛWPXFMÛFYDFQUÛPÛÛÛ�Q, � U S�ÛIBZSÛ�GBUIFS��Û4LU�ÛQJUºS� Û(S�ÛPATH�R Û-BU��QBUFS�ÛÛÛ�QFOL F�ÛIJOHÛ�GJWF��Û4LU�ÛQºÊDB Û(S�ÛPzNTE Û-BU�ÛRVJORVF�ÛÛÛ�QV)S��ÛIVSÛ�GJSF��Û(S�ÛPU>R Û)JUU��QBIIVS Û6NCS�ÛQJS Û0)(ÛGJVS�ÛÛÛÛÛ�QFE�PN�ÛIFUÛ�USBDF ÛGPPU��Û4LU�ÛQBEºNÛ�TUFQ� Û(S�ÛPE�DONÛ�CPUUPN��ÛÛÛ�QFSVUJ�ÛIFSVÛ�MBTUÛZFBS�Û�Û�IFSVZ�Û4LU�ÛQBSÓU Û(S�ÛPE�RUSI�ÛÛÛ�QPOU)��ÛIVOÛ�SPBE ÛGPSE��Û4LU�ÛQºOUI Û(S�ÛPO�NTOJÛ�TFB� Û-BU��QPOT �QPOU�JT Û0$4Û QPU �Û 5IFÛ XPSEÛ NVTUÛ IBWFÛ QSFTFSWFEÛ JUTÛ I�Û CFDBVTFÛ UIFÛ �P�Û IBEÛCFDPNFÛ�V�ÛPSÛJUÛDPNFTÛGSPNÛUIFÛPCMJRVFÛDBTFTÛXJUIÛ�QOU)��ÛÛÛÛÛÛ0OÛ�QJÛ�ÛIJÛ�ÛZJÛTFFÛ��EÛÛÛÛÛÛCÛ Û �QÛ �Û [FSPÛ Û Û Û JOÛ BOMBVUÛ CFGPSFÛ P�Û JOÛ BOMBVUÛ CFGPSFÛ DPOTPOBOU�Û BGUFSÛOBTBMÛÛÛ�QPE�NÛÛBDD��ÛPUOÛ�GPPU��Û4LU�ÛQ E� Û(S�ÛPO�D AÛBDD� Û-BU�ÛQFE�FN Û0)(ÛGVP[�ÛÛÛ�Q M)��ÛVMÛ�LJE��Û(S�ÛPW>LOJ�Û(PUI�ÛGVMBOÛ�Û�QMI��ÛÛÛ�QPMJ��Û BMJ�L Û QM�Û �XBWFT�Û XIJUFÛ IBJS��Û (S�Û POLIO�JÛ �HSFZ� Û 4LU�Û QBMJ�Uº�Û�Û�QFMJ� Û0)(Û GBMPÛ �Û �QPM�VP��Û/PUFÛ UIBUÛ UIFÛ �Q�Û�Û �I�ÛIFSFÛEJTBQQFBSFEÛCFGPSFÛPÛCFDBNFÛB�ÛÛÛ�QMF, � �UPT�ÛMJÛ�GVMM��Û4LU��QS Uº� Û-BU�ÛDPN�QM UVT�ÛÛÛ�QSF, � U�P��ÛFSBTU�BO�L Û CVUUPDLT��Û(S�ÛPRWKTO�JÛ�Û�QSP, � UP��Û4FFÛ��Û/PUFÛUIFÛQSPUIFUJDÛWPXFM�ÛÛÛ�QUFS�PN�ÛU FSÛ�TJEF��Û(S�ÛPTE�RONÛ�XJOH��ÛÛÛ�QUFM��ÛU F JÛ�FMN��Û(S�ÛPTELE�A Û-BU�ÛUJMJB�Û5IFTFÛXPSETÛNBZÛCFÛMPBOTÛGSPNÛBÛ

172� 5��6��3��%HHNHV��

.FEJUFSSBOFBOÛMBOHVBHF�ÛÛÛ�QSFJTL �V��ÛFS D Û �FMEFS ÛQSJFTU��Û-BU�ÛQS TDVTÛ �BODJFOU��Û(S�ÛPRE�SBUJ Û,SFUBOÛPREI>GUJÛXIJDIÛIBWFÛ�H �Û/PUFÛUIFÛQSPUIFUJDÛWPXFM�ÛÛÛÛ�O�Q M)�P��Û BN�VMÛ �TUFSJMF��Û TFFÛ POÛ VMÛ BCPWF�Û 5IFÛ NÛ TIPXTÛ UIBUÛ UIFÛDPNQPVOEÛXBTÛGPSNFEÛXIFOÛUIFÛQ�ÛXBTÛTUJMMÛUIFSF�ÛÛÛ�O�QVUS�JP��Û BNVSJÛ Û �VONBSSJFE�Û �Û �DIJME�MFTT�Û 4LU�ÛQVUSº�Û �DIJME� Û DG�Û -BU�ÛQVFS�ÛÛ5IFÛ�QÛ�Û�IÛXBTÛMPTUÛBGUFSÛUIFÛQSFGJY�ÛÛÛÛÛÛÛÛÛÛÛÛÛDÛÛ�Q��XÛÛBGUFSÛWPXFMÛÛÛ�TFQUN�ÛFXU OÛ�TFWFO��Û4LU�ÛTBQUº Û(S��xPTA� Û-BU�ÛTFQUFN�ÛÛÛ�, � FQJ�ÛFXÛ�BOE ÛBMTP��Û4LU�ÛºQJ Û(S�ÛyPI��ÛÛÛ�TVPQOP��ÛL VOÛ�TMFFQ��Û4LU�ÛTWºQOB� Û-BU��TPNOVT�Û)FSFÛXFÛIBWFÛVÛGSPNÛP�X�ÛÛÛ�, � S JQJP��ÛBSDVJÛ�FBHMF� ÛGSPNÛ�BSDJXJ�Û4LU�Û KJQZº��ÛÛÛÛ�QSFQ��Û FSFXJNÛ �TFFN��Û (S�Û PRzPW � 0*S�Û SJDIUÛ �GPSN Û TIBQF��Û 0OFÛ BMTPÛDPOTJEFSTÛBÛ SPPUÛ �L SFQ� Û BTÛ JOÛ1*&Û SPPUTÛ UIFÛ GJSTUÛ BOEÛ UIFÛ MBTUÛ DPOTPOBOUÛXFSFÛTFMEPNÛJEFOUJDBM�ÛÛÛ ��ÛÛ1*&Û�CÛÛÛÛÛÛÛ�CÛ �Û Q� Û Û /PUFÛ UIBUÛ UIFÛ QIPOFNFÛ �CÛ XBTÛ WFSZÛ SBSFÛ JOÛ 1SPUP�*OEP�&VSPQFBO�ÛÛÛ�TUJC��Û TUJQFNÛ �VSHF Û DPNQFM��Û (S�Û STEI�BWÛ �USFBE Û TUBNQÛ PO� Û STIBARO�JÛ�TUSPOH��ÛÛÛ�CJ�C, � ��Û NQFNÛ�ESJOL�ÛXJUIÛBÛOBTBMÛJOGJY�ÛPUIFSÛSFDPOTUSVDUJPOTÛBSFÛBMTPÛQSPQPTFE ÛCVUÛUIFZÛBSFÛOPUÛSFMFWBOUÛGPSÛUIFÛ�C�ÛÛÛ ��ÛÛ1*&Û�CI�ÛÛÛBÛÛ�CIÛ�ÛCÛÛXIFOÛOPUÛJOUFSWPDBMJDÛÛÛCÛÛÛÛÛÛÛÛ�ÛXÛCFUXFFOÛWPXFMTÛXIJDIÛEJTBQQFBSTÛBGUFSÛV�ÛÛÛÛÛÛÛBÛ�CIÛ�ÛCÛÛFYDFQUÛCFUXFFOÛWPXFMTÛÛÛ�CIFS��ÛCFSFNÛ�DBSSZ��Û4LU�ÛCIBS� Û(S�ÛFzRW Û-BU�ÛGFS Û(PUI�ÛCBÆSB�ÛÛÛ�CIF, � �OJ��ÛCBOÛ �XPSE��Û0*D�ÛCÌO�Û(S�ÛFWNH�Û�Û�CIP, � �OF, � ��UIFÛXPSEÛNBZÛIBWFÛIBEÛ[FSPÛHSBEF ÛIPXFWFS��ÛÛÛÛÛ�CISF, � U S�ÛF CBZSÛ�CSPUIFS��Û4LU�ÛCIS UBS� Û(S�ÛFR THR Û-BU�ÛGS UFS�ÛÛÛÛÛ�, � PSCI�PT�ÛPSCÛ�PSQIBO��Û(S�ÛO�RFANO�J Û-BU��PSCVT�ÛÛÛ�TSCI��ÛBSCJÛ�ESJOL�ÛBPS� ÛQSFT�ÛBSCFOBN�Û-BU�ÛTPSCF Û-JUI�ÛTVSCJÒ�ÛÛÛ��CIJÛJOTUS�ÛQM�ÛFOEJOH�Û�CÛBGUFSÛDPOTPOBOU�ÛHB BN�CÛHB OÛ�MBNC��ÛÛÛCÛÛÛ�CIÛ�ÛXÛÛCFUXFFOÛWPXFMTÛÛÛ�, � CIFM��ÛZ�BXFMVNÛ�JODSFBTF��ÛÛÛ��CIJÛJOTUS�ÛQM�ÛFOEJOH�Û�XÛBGUFSÛWPXFM�Û�BX ��JX ��PWÛ�ÛÛÛ��CIPS��ÛU BHB�XPSÛ�LJOH�Û�Û�DSPXO�CFBSFS��ÛÛ


Û ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛMBCJBMÛTFFÛ��ÛÛÛÛÛÛ ��ÛÛ%FOUBMT�ÛÛ ��ÛÛ1*&Û�U�ÛÛÛÛÛÛÛÛBÛ�UÛ�Û[FSPÛJOÛBOMBVUÛCFGPSFÛDPOTPOBOUÛÛÛÛÛÛCÛÛÛÛÛ�ÛUÛÛBGUFSÛ"SNFOJBOÛTJCJMBOUÛÛÛÛÛÛÛDÛÛÛÛÛÛ�ÛEÛÛBGUFSÛS �M �O �NÛÛÛÛÛÛÛEÛÛÛÛÛ�Û[FSPÛXPSE�GJOBMMZÛBGUFSÛOÛÛBGUFSÛUIFÛMPTTÛPGÛGJOBMÛWPXFMTÛÛÛÛÛÛFÛÛÛÛÛÛ�ÛU ÛÛFMTFXIFSF ÛFYDFQUÛJOÛUIFÛGPMMPXJOHÛJOTUBODFT��ÛÛÛÛÛGÛÛÛÛÛÛÛ�ÛXÛCFGPSFÛS �M �O �N�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛHÛÛÛÛÛÛ�ÛZÛCFUXFFOÛWPXFMTÛBUÛBÛTUBHFÛCFGPSFÛUIFÛMPTTÛPGÛGJOBMÛWPXFMTÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛCVUÛXÛCFUXFFOÛCBDLÛWPXFMT�ÛÛÛÛÛÛÛÛÛÛÛBÛ�UÛ�Û[FSPÛJOÛBOMBVUÛCFGPSFÛDPOTPOBOUÛÛÛÛ�USFJFT�ÛFSFL Û�UISFF��Û4LU�ÛUSºZBT Û(S�ÛTREI>J Û-BU�ÛUS T�ÛÛÛÛÛ8FÛIBWFÛOPÛFWJEFODFÛGPSÛPUIFSÛDPOTPOBOUT ÛCVUÛ�UÛIBSEMZÛFWFSÛPDDVSSFEÛCFGPSFÛPUIFSÛDPOTPOBOUT�ÛÛÛÛÛÛCÛ �UÛ �Û UÛ Û BGUFSÛ TJCJMBOUÛ BMTPÛ BGUFSÛ BÛ TJCJMBOUÛ UIBUÛ GJSTUÛ BSPTFÛ JOÛ"SNFOJBO�ÛÛÛÛ0OÛ�TU�ÛTFFÛ��ÛÛÛ�QSF, � UP��ÛFSBTUBOL Û�IJOE�ÛCPEZQBSU�Û(S��PRWKTO�JÛ�Û�QSP, � UPT�ÛÛÛÛ�EIV , � U S�ÛEVTUSÛ�EBVHIUFS��Û4LU��EVIJUºS� Û(S�Û UGA�THR Û-JUI�ÛEVLU Û(PUI�ÛEBVIUBS�ÛÛÛÛÛÛDÛÛ�UÛ�ÛEÛÛBGUFSÛS �M �O �N�ÛÛÛÛÛ�NSUPT�ÛNBSEÛ�NBO ÛIVNBOÛCFJOH��Û4LU�ÛN Uº� Û(S�ÛBROTO�J�ÛÛÛ�CISUJ��ÛCBSEÛ�IFBQ��Û4LU�ÛCI UÆ� Û-BU�ÛGPST �GPSUJT Û(PUI��HB�CBÓS× TÛ�CJSUI��ÛÛÛ�, � SUV��Û BSEÛ �PSEFSJOH ��Û 4LU�Û UÓ� Û (S�Û A�RTU�WÛ �BSSBOHF Û QSFQBSF� Û -BU�ÛBSUVTÛ�KPJOU ÛMJNC��ÛÛÛ�, � FOUFSP�� OEFS�L Û�JOUFTUJOFT��Û(S�Û|NTERA�Û0*D�ÛJÉS Û0$4ÛPUSPCB�ÛÛÛÛ�, � FO, � �UF, � �Û ES�BOEÛ �EPPS�QPTU Û �GSBNF�Û ES�Û GSPNÛ �EVS�Û �EPPS��Û 4LU��U � Û-BU�ÛBOUBFÛQM� Û0*D�ÛPOEÛ�GSPOU�SPPN��ÛÛÛÛÛÛEÛ�UÛ�Û[FSPÛÛÛÛXPSE�GJOBMMZÛBGUFSÛOÛBGUFSÛUIFÛMPTTÛPGÛÛGJOBMÛWPXFMTÛÛÛ�EXJ�E NUJ�ÛL TBOÛ�UXFOUZ��Û"W�ÛW TBJUJ Û(S�Û%PS�ÛðeKATI Û-BU��W HJOU �ÛÛÛ��E PNU, � �Û�TVOÛ�EFDBEF� ÛF�H�ÛVU �TVOÛ�FJHIUZ��Û(S�Û KONTA Û-BU�Û�HJOU �ÛÛÛ*OÛ DBTFTÛ MJLFÛ ES�BOEÛ TFFÛ BCPWFÛ UIFÛ �EÛ XBTÛ SFTUPSFEÛ BGUFSÛ UIFÛ PCMJRVFÛDBTFT ÛMJLFÛHFO�ÛESBOEJÛXIFSFÛUIFÛ�OE�ÛXBTÛQSFTFSWFEÛJOUFSOBMMZ�ÛÛ�, � FOUJ�Û OEÛ �UPXBSET Û JO Û BUÛ FUD��Û QSPDMJUJDÛ GPSÛ �IBOE�Û 4LU�Û ºOUJ Û(S�Û�NTI� Û)JUU�ÛIBOUJ Û-BU�ÛBOUF�Û5IFÛQSPDMJTJTÛXJMMÛCFÛUIFÛSFBTPOÛGPSÛUIFÛNBJOUFOBODFÛPGÛ UIFÛ �EÛ UIFÛXPSEÛ GPSNJOHÛBÛVOJUZÛXJUIÛ UIFÛ GPMMPXJOHÛXPSE ÛBOEÛBMTPÛGPSÛUIFÛEJTBQQFBSBODFÛPGÛUIFÛI��Û5IFSFÛXBTÛOPÛSFTUPSBUJPOÛJO�ÛÛÛ�QPOU�F, � ��Û IVOÛ �GPSE��Û IFSFÛ UIFÛ �EÛ XBTÛ OPUÛ SFTUPSFEÛ CFDBVTFÛ JOÛ JUTÛ

174� 5��6��3��%HHNHV��

BSDIBJDÛ JOGMFDUJPOÛ CPUIÛ F, � Û BOEÛ, � Û EJTBQQFBSFEÛ F�I�Û HFO�Û �QPOU�, � �PTÛ �ÛIVO�ÛÛÛÛÛÛÛFÛ�UÛ�ÛU ÛÛJOÛPUIFSÛQPTJUJPOTÛOPU�ÛJOÛBOMBVUÛCFGPSFÛDPOTPOBOU ÛCFGPSFÛPSÛBGUFSÛSFTPOBOU ÛCFUXFFOÛWPXFMTÛÛÛ�UST��ÛU B BNJNÛ�XJUIFS��Û(S�ÛTE�RSOMAI Û-BU�ÛUPSSF Û0)(ÛEFSSFOÛ�ESZÛVQ��ÛÛÛÛ�TFQUN�ÛFXU OÛ�TFWFO��Û4LU�ÛTBQUº Û(S��xPTA� Û-BU��TFQUFN�ÛÛÛ�, � PVUJ��ÛBXU Û �BÛQBTTJOHÛUIFÛOJHIU��Û(S�ÛAU�LH�Û �DPVSUZBSE� ÛI�AU�WÛ �QBTTÛUIFÛOJHIU��Û/PUFÛUIBUÛBGUFSÛ ÛUIFÛEFOUBMÛCFDBNFÛU �ÛÛÛÛÛÛÛGÛÛ�UÛ�ÛXÛÛCFGPSFÛS �M �O �NÛÛ�ÛÛÛÛÛ�Q, � US�PTÛHFO�ÛTH��ÛIBXSÛ�PGÛUIFÛGBUIFS��Û(S�ÛPATRO�J Û-BU�ÛQBUSJT�ÛÛÛ�, � S, � USPN�ÛBSBXSÛ�QMPVHI��Û(S�Û�ROTRON Û0*S�ÛBSBUIBS Û0*D�ÛBSÉS�ÛÛÛ�, � M, � US��ÛB BXS�JÛÛ�GFNBMFÛXIPÛHSJOETÛDPSO��Û(S�ÛA�LETRI�J�ÛÛ�ÛÛÛÛÛÛ� O, � �UM��ÛDOBX ��QBSFOU��Û(S�ÛGENE�TWR�ÛÛÛÛÛÛÛHÛ�UÛ�ÛZÛ ÛCFUXFFOÛWPXFMTÛ Û BUÛBÛTUBHFÛCFGPSFÛUIFÛ MPTTÛPGÛ GJOBMÛWPXFMTÛCVUÛXÛCFUXFFOÛCBDLÛWPXFMTÛÛÛ�Q, � U S�ÛIBZSÛ�GBUIFS��Û4LU�ÛQJUºS� Û(S�ÛPATH�R Û-BU�ÛQBUFS�Û5IFÛ�UÛ�Û� ÛCFDBNFÛZÛBUÛUIFÛTUBHFÛ�IB JS�ÛÛÛ�CI, � UJ��ÛCBZÛ�XPSE��Û(S�ÛFA�SIJ��FA�TIJ��ÛÛÛÛÛ�QFSVUJ�Û IFSVÛ �MBTUÛ ZFBS�Û �ZÛ BGUFSÛ VÛ EJTBQQFBSFE�Û 4LU�Û QBSÓU Û (S�ÛPE�RUTIÛ�ÛPE�RUSI�ÛÛÛ� MVUPT�ÛMVÛ�GBNPVT��Û4LU�Û SVUº� Û(S��KLUTO�J Û-BU�ÛJO�DMVUVT�ÛÛÛ��, � �UP�Û �BXÛ �ÛTH�Û NJEEMFÛ FOEJOH�Û ,PSUMBOEUÛ ��Û BTTVNFTÛ UIFÛ TBNFÛEFWFMPQNFOUÛJO�DBOBXU Û�LOPXO� ÛGSPNÛ� O, � �UPTÛXJUIÛBÛTFDPOEBSZÛTVGGJYÛ�UP��ÛTFFÛ��ÛÛÛÛÛÛÛ ��ÛÛ1*&Û�E ÛÛÛÛÛ�EÛ�ÛU��ÛÛÛÛÛ�EP, � SPNÛ�ÛUVSÛ�HJGU��Û(S�ÛDW>RON Û0$4ÛEBS �ÛÛÛ�E N�ÛUVOÛ�IPVTF��Û4LU�ÛEºN� Û(S�ÛDW>�MA��DO�MOJ Û-BU��EPNVT�ÛÛÛ�EF, � JV S�ÛUBZHSÛ�IVTCBOE�TÛCSPUIFS��Û4LU�ÛEFWºS� Û(S�ÛD H�R Û-JUI�ÛEJFWFSÅT�ÛÛÛ�QPE�N�ÛPUOÛ�GPPU��Û4LU�ÛQºE� Û(S�ÛPO�D Û-BU�ÛQFE��ÛÛÛ�QFE�PN�ÛIFUÛ�USBDF ÛUSBJM ��Û4LU�ÛQBEºN Û(S�ÛPE�DON�ÛÛÛ� SE�ÛTJSUÛ�IFBSU��Û(S��KH>R��KRADI�H Û-BU�ÛDPS �DPSEJT Û(PUI�ÛIBÆSUP�ÛÛÛ�, � PTE�PT�ÛPTUÛ�CSBODI��Û(S�Û´ZOJ Û(PUI�ÛBTUT�ÛÛÛ ��ÛÛ1*&Û�EI ÛÛÛÛÛ�EIÛ�ÛE�ÛÛÛÛÛ�EIVS��ÛEVS�L Û�HBUF��Û(S�ÛKU�R Û-JUI�ÛEVS� Û(PUI�ÛEBÓS�ÛÛÛ�EIV , � �U S�ÛEVTUSÛ�EBVHIUFS��Û(S�Û UGA�THR Û-JUI�ÛEVLU Û(PUI�ÛEBVIUBS��ÛÛÛÛÛÛ�EIPJ I��ÛE [Û�IFBQ��Û(S�ÛTOI>COJ Û(PUI�ÛEBJHT�ÛÛÛ�EIM, � SPT�ÛEBMBSÛ�HSFFO�ÛGSFTI��Û(S�Û ALERO�J�Û


ÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛEFOUBM�ÛTFFÛ��ÛÛÛÛÛÛ��ÛÛÛ ��ÛÛ1BMBUBMTÛÛ ��ÛÛ1*&Û� ÛÛÛÛÛÛÛÛBÛ� Û�Û[FSPÛÛJOÛBOMBVUÛCFGPSFÛSFTPOBOUÛPOMZÛBUUFTUFEÛCFGPSFÛMÛÛÛÛÛÛCÛÛÛÛÛ�ÛXÛJOÛJOMBVUÛCFGPSFÛSÛÛÛÛÛÛÛÛÛÛÛÛÛÛDÛÛÛÛÛÛ�ÛTÛÛÛFMTFXIFSFÛÛÛÛÛÛÛBÛÛ� Û�ÛÛ[FSPÛÛJOÛBOMBVUÛCFGPSFÛSFTPOBOU�ÛPOMZÛBUUFTUFEÛCFGPSFÛMÛÛÛ� MVUPT�ÛMVÛ�GBNPVT��Û4LU�Û SVUº� Û(S��KLUTO�J Û-BU�ÛJO�DMVUVT�ÛÛÛÛÛÛ"ÛQSPCMFNÛHJWFTÛTSVO�L Û�TIJO�CPOF��Û,PSUMBOEUÛ��CÛGPMMPXTÛ1FEFSTFOÛJOÛQPTJUJOHÛ� SVO��Û*ÛXPOEFSÛXIFUIFSÛJUÛJTÛOPUÛBÛMPBOÛGSPNÛ*SBOJBO ÛDG�Û"W�ÛTSBPOJ�Û�CVUUPDLT�ÛJOÛTQJUFÛPGÛUIFÛEJGGFSFODFÛJOÛNFBOJOH�ÛTVDIÛTIJGUTÛBSFÛOPUÛJNQPTTJCMFÛJOÛUIFÛDBTFÛPGÛCPEZ�QBSUT�ÛDG�Û&OH�ÛIBN�ÛÛÛÛÛÛÛCÛ� Û�ÛXÛJOÛJOMBVUÛCFGPSFÛSÛÛÛÛÛ�TN7 SV�� NBXSV�L Û �CFBSE��Û 4LU�Û Nº SV� Û -JUI�Û TN¼LSBT�Û 5IFSFÛ JTÛEJTDVTTJPOÛPOÛUIFÛWPXFM�ÛEJBM�ÛNJSVL ÛNBZÛCFÛUIFÛGPSNÛXJUIÛ�FÛ�FX��JX��J� ÛUIFÛBÛCFJOHÛBÛSFEVDFEÛWPXFMÛJOÛUIFÛ[FSPÛHSBEF�ÛÛÛÛÛÛDÛ� Û�ÛTÛÛÛÛ� SE�ÛTJSUÛ�IFBSU��Û(S�ÛKH>R��KRADI�H Û-BU�ÛDPS �DPSEJT Û-JUI�ÛuJSEÅT�ÛÛÛÛ� J)V O�ÛTJXOÛ�QJMMBS��Û(S��K WN Û.ZD�ÛLJXPÛ�L X O��ÛÛÛ�EF N�ÛUBTOÛ�UFO��Û4LU�ÛEº B Û(S�ÛDE�KA Û-JUI�ÛE uJNU Û-BU�ÛEFDFN�ÛÛÛÛ�� PNU, � �ÛFSF�TVOÛ�UIJS�UZ��Û(S�Û KONTA Û-BU�Û�HJOU �ÛDG�Û4LU�Û� BU�ÛÛÛÛ ��ÛÛ1*&Û� ÛÛÛÛÛ� Û�ÛDÛÛÛÛ� FO, � PT�ÛDJOÛ�CJSUI��Û4LU�ÛKºOBT Û(S�ÛGzNOJ Û-BU�ÛHFOVT�ÛÛÛ� POV�ÛDVOSÛ�LOFF��Û4LU�ÛK OV Û(S�ÛGO�NU�Û-BU��HFO �ÛÛÛ� FS, � ��ÛDFSÛ�PME��ÛDG�Û4LU�ÛKºSBOU� Û(S�ÛGzRWN Û0TTFUJDÛ[¿SPOE�ÛÛÛ�, � FT��ÛBDFNÛ�DBSSZ��Û-BU��HFS �ÛÛÛÛ�VPS PN�ÛHPSDÛ�XPSL��Û(S��E��RGON��ðE�RGON Û0)(�XFSD�ÛÛÛÛ�NF , � ��ÛNFDÛ�CJH��Û4LU�ÛNºI� Û(S�ÛME�GAJ Û(PUI��NJLJMT�ÛÛÛÛ�, � S ��ÛBSDBU Û�TJMWFS��Û"W�Û S [BUB� Û-BU�ÛBSHFOUVN�ÛDG�Û(S�Û�RGUROJ�ÛÛÛÛÛÛ ��ÛÛ1*&Û� IÛÛÛ� IÛ�ÛK��[ÛCFUXFFOÛWPXFMTÛÛÛ� IFTS��ÛKF OÛ�IBOE��Û)JUU�ÛLFuuBS Û(S�ÛCEI�R�ÛÛÛÛ� IJ N�ÛKJXOÛ�TOPX��Û"W�Û[ZBN� Û(S�ÛCIW�N�Û-BU��IJFNT�ÛÛÛ� IJN�FS��Û KNF �OÛ �XJOUFS��Û 4LU�Û IÂNBO� Û IJNº� Û (S�� CEIMW�N�� CEIMERINO�J��DU�S CIMOJ ÛÛ)JUU�ÛHJNBOU� Û-JUI�� JFN¹ Û-BU�ÛI CFSOVT�ÛÛ

176� 5��6��3��%HHNHV��

ÛÛ�CIS IV��ÛCBSK�SÛ�IJHI��Û4LU�ÛC I�BOU� Û)JUU�ÛQBSLVT�ÛÛÛ�NF ITSJ�ÛNFSKÛ �OFBSÛCZ�Û XJUIÛNFUBUIFTJT�Û(S�ÛMzCRI�Û,PSUMBOEUÛ��C Û��Û <�Û UIJTÛWPM� Û ��>Û SFDPOTUSVDUTÛ UIFÛ GPSNÛXJUIÛ BOÛ �T� Û CFDBVTFÛPUIFSXJTFÛUIFÛ WFMBSÛ XPVMEÛ IBWFÛ CFFOÛ EFQBMBUBMJ[FEÛ CFGPSFÛ S�Û 5IJTÛ BMMPXTÛ UIFÛFYQMBOBUJPOÛUIBUÛUIFÛGPSNÛIBTÛBSJTFOÛGSPNÛ�NF�� ITSJÛÛ�BUÛIBOE��ÛÛÛ�NFJ I��ÛNJ[�FN��VSJOBUF��Û"W�ÛNB [B� Û(S�ÛO�MEI�CW �-BU�ÛNFJ �NJOH ��EIPJ IP��ÛE [��IFBQ��Û(S�ÛTOI>COJ Û"W�ÛQBJSJ�EB [B��FODMPTVSF� Û(PUI�ÛEBJHT�ÛÛÛ ��ÛÛ%FQBMBUBMJ[BUJPOÛÛ *UÛ JTÛ BTTVNFEÛ UIBUÛ CFGPSFÛ SÛ UIFÛ QBMBUBMTÛ XFSFÛ EFQBMBUBMJ[FEÛ,PSUMBOEUÛ ��C Û ��Û <�Û UIJTÛ WPM� Û ��>�Û 5IJTÛ JTÛ TFFOÛ JO� NBXSVL Û �CFBSE�Û �Û�TN7 SV�ÛTFFÛ��Û'PS�NFSKÛ�OFBS�ÛTFFÛBCPWF�ÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛBÛQBMBUBM�ÛTFFÛ��ÛÛÛÛ ��ÛÛ-BCJPWFMBSTÛBOEÛQVSFÛWFMBSTÛÛ 5IFÛMBCJPWFMBSTÛMPTUÛUIFJSÛMBCJBMJ[BUJPOÛTUBHFÛ��ÛQBSUMZÛBMSFBEZÛBUÛ��ÛBOEÛTPÛDPJODJEFEÛXJUIÛUIFÛQVSFÛWFMBST�Û5IFÛFYJTUFODFÛPGÛQVSFÛWFMBSTÛJTÛBÛIPUMZÛEFCBUFEÛJTTVF ÛCVUÛJOÛTPNFÛDBTFTÛUIFJSÛFYJTUFODFÛDBOOPUÛCFÛEFOJFE�Û*UÛ JTÛ HFOFSBMMZÛ BENJUUFEÛ UIBUÛ BGUFSÛ VÛ POMZÛ QVSFÛ WFMBSTÛ PDDVSSFE Û CVUÛ JOÛ"SNFOJBOÛWFMBSTÛXFSFÛQBMBUBMJ[FEÛBGUFSÛVÛBOEÛUIFOÛGFMMÛUPHFUIFSÛXJUIÛUIFÛ1*&Û QBMBUBMT�Û rÛ *OÛ UIFÛ GPMMPXJOHÛXFÛ TIBMMÛ HJWFÛ FYBNQMFTÛ PGÛ QVSFÛ WFMBSTÛBGUFSÛ UIPTFÛXJUIÛ MBCJPWFMBST�Û5IFÛWFMBSTÛXFSFÛQBMBUBMJ[FEÛCFGPSF� F � JÛ TFFÛ�� ÛCVUÛPGUFOÛUIFÛOPO�QBMBUBMJ[FEÛGPSNTÛXFSFÛSFTUPSFE�ÛÛÛ ��ÛÛ1*&ÛÛ�L ÛBOEÛ�LÛÛÛÛÛÛBÛ�L ÛÛÛ�Û[FSPÛÛÛJOÛBOMBVUÛCFGPSFÛDPOTPOBOU�ÛÛÛÛÛCÛÛÛÛÛÛÛÛ��HÛÛBGUFSÛSFTPOBOUÛS �M �O �N�ÛOPUÛBMMÛBSFÛBUUFTUFEÛÛÛÛÛDÛÛÛÛÛÛÛÛ��L ÛÛÛFMTFXIFSFÛÛÛÛEÛÛÛÛÛÛÛÛÛ�ÛTÛÛÛBGUFSÛVÛÛÛÛ0OÛUIFÛTFRVFODFÛ�OL �ÛTFFÛ��ÛÛÛÛÛBÛ�L Û�Û[FSPÛÛÛJOÛBOMBVUÛCFGPSFÛDPOTPOBOUÛÛÛÛ�L UXS��ÛL �BTVOÛ�GPSUZ��ÛTFFÛ��ÛÛÛÛÛÛÛ�L SFQ��ÛFSFXJNÛ �TFFN��Û(S�ÛPRE�PW Û JGÛPOFÛQSFGFSTÛUIFÛSFDPOTUSVDUJPOÛXJUIÛ�L ÛDG�Û��D�ÛÛÛÛÛCÛ�L ��HÛÛÛBGUFSÛSFTPOBOUÛS �M �O �N�ÛOPUÛBMMÛBSFÛBUUFTUFEÛÛÛ�QFOL F�ÛIJOHÛ�GJWF��Û4LU�ÛQºÊDB Û(S�ÛPzNTE Û"FPMJDÛPzMPE�Û5IFÛWFMBSÛXBTÛOPUÛQBMBUBMJ[FEÛBGUFSÛUIFÛOBTBM�ÛÛÛ�, � FSL PT�ÛFSHÛ�TPOH��Û4LU�ÛBSLº��Û


ÛÛ�, � SL��ÛBSHFMÛ �PCTUBDMF��Û(S�ÛA�RKE�W�Û-BU��BSDF �Û5IFÛOPO�QBMBUBMJ[FEÛGPSNÛXBTÛUBLFOÛGSPNÛBÛGPSNÛXJUIÛBÛGPMMPXJOHÛP ÛDG�Û(S�Û�RKOJ�ÛÛÛÛÛDÛÛÛÛ�L ��L ÛÛÛ�L F, � N�ÛL BOÛ�BT��Û-BU�ÛRVBN�ÛÛÛÛ�MFJL ��Û FMJL Û BPS� Û ML BOFNÛ QSFT�Û �MFBWF��Û 4LU�Û SJD� Û (S�Û LEI�PW�Û |LIPON Û -BU�ÛMJORV Û (PUI�Û MFJ BO�Û 5IFÛ OPO�QBMBUBMJ[FEÛ GPSNÛ XBTÛ JOUSPEVDFEÛ GSPNÛGPSNTÛXJUIÛP�WPDBMJTN�ÛÛÛÛÛÛ��L F�Û �L Û JOEFGJOJUFÛ QBSUJDMF�Û Û 4LU�Û Lº �DB Û -BU�Û RVJT�RVF Û (PUI�Û �I�Û 5IFÛGPSNÛNBZÛOPUÛIBWFÛCFFOÛQBMBUBMJ[FEÛCFDBVTFÛJUÛXBTÛBOÛFODMJUJD�ÛÛÛ�LFS��ÛL FSFNÛ�TDSBUDI��ÛÛÛÛÛEÛÛ�L Û�ÛTÛÛÛBGUFSÛVÛÛÛ�MFVLPT�ÛMPZTÛ�MJHIU�ÛTVCTU��Û4LU�ÛSPLº� Û(S�ÛLEUKO�JÛ�XIJUF� Û-JUI�ÛMB LBT�ÛÛÛÛ"TÛXBTÛNFOUJPOFEÛ BCPWF Û JUÛ JTÛ HFOFSBMMZÛ BTTVNFEÛ UIBUÛ BGUFSÛ VÛ POMZÛ BÛQVSFÛWFMBSÛPDDVSSFE�ÛTPÛFYBNQMFTÛXJUIÛ�L ÛBSFÛOPUÛUPÛCFÛFYQFDUFE�ÛÛÛ ��ÛÛ1*&Û�H �BOEÛ�HÛÛÛ�H Û�ÛLÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ�H PV��ÛLPWÛ�DPX��Û4LU�ÛHºW� Û(S�ÛBOU>J Û0*S�ÛCÌ Û0)(ÛLVP�ÛÛÛÛÛ�, � SFH PT�ÛFSFLÛ�FWFOJOH��Û4LU�ÛSºKBT Û(S�Û|REBOJ Û(PUI�ÛSJRJT�ÛÛÛÛ�H , � FJ�� LFBNÛ �MJWF��Û GPSÛ UIFÛSFDPOTUSVDUJPOÛPGÛ UIFÛSPPUÛ,PSUMBOEUÛ��B Û��Û<UIJTÛWPM� Û��G�>�ÛÛÛ�H FO, � Û �ÛLJOÛ �XPNBO�Û QM�ÛLBOBZL Û�Û�H O, � �Û �Û4LU�Û KºOJ� ÛHO � Û(S�ÛGUNH� Û0$4Û FOB Û0*S�ÛCFO�Û5IFÛL�ÛXBTÛJOUSPEVDFEÛGSPNÛGPSNTÛXJUIÛ[FSPÛHSBEFÛPGÛUIFÛSPPU�ÛÛÛ�H FS, � �P��ÛLFSÛ �GPPE��Û4LU�ÛHJSºUJ Û(S�ÛBI BRW� SKW ÛCPUIÛXJUIÛ[FSPÛHSBEF�Û-JUI�Û HFSJÒ�Û 5IFÛ WFMBSÛ HPFTÛ CBDLÛ UPÛ BÛ GPSNÛ XJUIÛ P�WPDBMJTNÛ XIJDIÛ JTÛFYQFDUFEÛIFSFÛBOEÛUPPLÛUIFÛWPXFMÛGSPNÛUIFÛWFSC�ÛÛ�HÛÛÛÛÛÛ0OÛUIFÛTFRVFODFÛ��BVH�ÛTFFÛ��ÛÛÛ ��ÛÛ1*&Û�H IÛBOEÛ�HIÛÛÛÛÛÛÛ�H IÛ�ÛH�ÛÛÛ�H I ��ÛHBO�FNÛ�TUSJLF��ÛQFSIBQTÛUIFÛSPPUÛXBTÛ�H IPO��Û4LU�ÛIBO� Û(S�Û EI�NW�ÛFO�NOJ�Û-JUI�ÛHFOJÒ�ÛÛÛ�H IFM I��ÛHF K�L Û QMVS�Û �HMBOET��Û0$4Û M [BÛ �Û � FM[B�Û 5IFÛ WFMBSÛ JTÛ OPUÛQBMBUBMJ[FE�ÛXBTÛJUÛUBLFOÛGSPNÛUIFÛ[FSPÛHSBEF ÛÛÛÛ ��ÛÛ1BMBUBMJ[BUJPOÛPGÛMBCJPWFMBST�ÛÛ 1BMBUBMJ[BUJPOÛPGÛ MBCJPWFMBSTÛIBTÛCFFOÛBÛQPJOUÛPGÛEJTDVTTJPO ÛBTÛ

178� 5��6��3��%HHNHV��

SBUIFSÛPGUFOÛ UIFÛOPO�QBMBUBMJ[FEÛ GPSNÛIBTÛ CFFOÛ SFTUPSFE�Û)PXFWFS Û JUÛ JTÛNPTUÛQSPCBCMFÛUIBUÛUIFÛQBMBUBMJ[BUJPOÛCFGPSFÛFÛBOEÛJÛXBTÛSFHVMBSÛBOEÛUIBUÛEFWJBUJPOTÛ NVTUÛ CFÛ FYQMBJOFEÛ CZÛ HFOFSBMJ[BUJPOÛ PGÛ UIFÛ OPO�QBMBUBMJ[FEÛGPSNÛXIJDIÛ PSJHJOBUFEÛ CFGPSFÛ PÛ PSÛ JOÛ UIFÛ [FSPÛ HSBEFÛ CFGPSFÛ DPOTPOBOU�Û1BMBUBMJ[BUJPOÛNBZÛIBWFÛCFFOÛCMPDLFEÛCZÛBÛQSFDFEJOHÛOBTBM ÛBTÛJOÛIJOHÛ�Û�QFOL FÛ,PSUMBOEUÛ��B�ÛÛÛÛÛ5IFÛEFWFMPQNFOUTÛNBZÛCFÛTVNNBSJ[FEÛIFSF�Û*ÛBEEÛUIFÛSFQSFTFOUBUJPOÛPGÛUIFÛ1*&ÛQBMBUBMTÛGPSÛDPNQBSJTPO�ÛÛÛÛQBMBUBMTÛÛÛÛÛÛÛÛÛÛÛÛÛ� ÛÛÛÛÛ�HÛÛÛÛÛÛ� IÛÛÛÛÛÛÛÛ"SN�ÛÛÛÛT�Û��D��ÛK �[ÛÛÛÛÛÛ[ÛJOUFSWPDBMJDÛÛÛÛMBCJPWFMBSTÛÛÛÛÛ�L ��H ��H IÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛL ��L��HÛÛÛÛÛÛÛÛÛÛÛ ÛÛÛÛÛÛÛÛCFGPSFÛF �JÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ�� Û ÛÛÛÛÛÛ ÛJOUFSWPDBMJDÛÛÛÛÛÛ#FGPSFÛZÛXFÛGJOEÛUIFÛMBUUFSÛEFWFMPQNFOU ÛCVUÛUIFÛZÛEJTBQQFBST�ÛXFÛTIBMMÛUIFSFGPSFÛ EJTDVTTÛ UIJTÛ EFWFMPQNFOUÛ VOEFSÛ UIFÛ DMVTUFSTÛ ��Û *UÛ JTÛLOPXOÛ POMZÛ JOÛ UIFÛ DBTFÛ PGÛ UIFÛ 1*&Û WPJDFMFTTÛ TPVOE�Û 8IFOÛ UIFÛEFWFMPQNFOUÛPDDVSTÛJOÛUIFÛMBTUÛTZMMBCMF ÛXIJDIÛEJTBQQFBSFE ÛJUÛJTÛEJGGJDVMUÛPSÛJNQPTTJCMFÛUPÛEFDJEFÛXIFUIFSÛJÛPSÛZÛGPMMPXFE�ÛÛÛÛÛÛÛBÛÛ1*&Û�L ÛÛCFGPSF�F �JÛÛ�Û ÛÛÛ�L FUVPSFT�Û PSL Û�GPVS��Û4LU�ÛDBUW SBT Û(S�ÛTE�SSAREJ Û-BU�ÛRVBUUVPS�ÛÛÛ�, � L �J, � �ÛB �L Û QMVS�Û �FZFT��ÛIFSFÛXFÛBSFÛOPUÛTVSFÛXIFUIFSÛUIFÛFOEJOHÛEFWFMPQFEÛUPÛ� ÛPSÛ� F�ÛGPSÛUIFÛMBUUFSÛQPTTJCJMJUZÛTFFÛCFMPXÛVOEFSÛ$MVTUFST�ÛÛÛ��L JE�Û [�JO� Û �BOZUIJOH��Û 0UIFSÛ QSPOPVOTÛ IBWFÛ �L Û BTÛ HFOFSBMJ[JOHÛQBSUJDMF Û XJUIPVUÛ QBMBUBMJ[BUJPO�Û UIFÛ QBSUJDMFÛ IBTÛ CFFOÛ JEFOUJGJFEÛ XJUIÛ��L F ÛBTÛJOÛ-BU�ÛRVJT�RVF Û4LU�ÛLº �DB�Û5IFÛQBSUJDMFÛBQQBSFOUMZÛCFIBWFEÛMJLFÛ-BU�ÛOFDÛCFTJEFÛOFRVF ÛPSÛBDÛCFTJEFÛBURVF�ÛJUÛXBTÛTIPSUFOFEÛUPÛ��LÛXJUIÛMPTTÛPGÛUIFÛ�F ÛTPÛUIBUÛUIFSFÛXBTÛOPÛQBMBUBMJ[BUJPO�ÛÛÛ�) � PJV�L JE �ÛP Û�OPU� ÛCVUÛUIFÛWPDBMJTNÛQSFTFOUTÛEJGGJDVMUJFT�Û(S�ÛOU�K�ÛÛÛÛ�MOH I�J, � �ÛMBO �L Û�Û�MBOH�Z� ÛBÛEVBMÛGPSN Û�CSFBTU��Û0*D��MVOHBÛ�MVOH��Û5IFÛDPNQBSJTPOÛJTÛBUUSBDUJWF ÛCVUÛOPUFÛUIBUÛUIFÛ"SNFOJBOÛGPSNÛDBOOPUÛHPÛCBDLÛUPÛ UIFÛ SPPUÛ �, � MFOH I�Û �MJHIU� Û GSPNÛXIJDIÛ UIFÛ XPSEÛ GPSÛ �MVOH�Û JTÛNPTUMZÛEFSJWFE�ÛÛÛÛÛÛÛ$PNQBSFÛBMTPÛTVDIÛHSPVQTÛBTÛQ L BNÛ�IJTT�Û�ÛQ FNÛ�CMPX ÛJOGMBUF��ÛÛÛÛÛÛÛCÛÛ1*&Û�H ÛÛCFGPSFÛF �J�Û�Û ÛÛÛ�, � FVH��ÛB FNÛ �HSPX��Û-JUI�ÛºVHV Û-BU�ÛBVHF �Û/PUFÛUIBUÛIFSFÛUIFÛV ÛXIJDIÛQBMBUBMJ[FT ÛXBTÛMPTUÛWFSZÛFBSMZÛTUBHFÛ��ÛÛÛ�HFN��Û NMFNÛ�QSFTT ÛTRVFF[F��Û0$4Û UJ�ÛÛÛÛÛÛÛDÛ1*&Û�H IÛÛCFGPSF�F �J�Û�Û �Û ÛÛCFUXFFOÛWPXFMTÛÛÛ�H IFSN��Û FSNÛ�XBSN��Û(S�Û ERMO�J�ÛXJUIÛ�P�Û4LU�ÛHIBSNº� Û-BU�ÛGPSNVT�Û


ÛÛ�H IJ)TM��Û JMÛ�TJOFX��Û-JUI�ÛHÖTMB�ÛÛÛÛ�, � FHIJT�Û J Û �TOBLF��Û4LU�ÛºIJ� Û(S�Û|CIJ�Û0OÛ�FÛ�Û�FJÛ�Û ÛPGÛXIJDIÛ J�ÛJTÛUIFÛSFEVDFEÛGPSN ÛTFFÛ��ÛÛÛÛÛÛÛEÛ�TLÛ�ÛuÛÛCFGPSFÛF �JÛÛÛÛÛÛÛ5IJTÛ HSPVQÛ JTÛ NFOUJPOFEÛ IFSFÛ CFDBVTFÛ PGÛ JUTÛ TQFDJBMÛ DIBSBDUFS��/PUFÛUIBUÛ�TLÛ�ÛD Û JOÛPUIFSÛDBTFT�ÛTFFÛ��G�Û5IFÛDPOEJUJPOTÛPGÛ UIJTÛDIBOHFÛIBWFÛOPUÛ ZFUÛ CFFOÛ FTUBCMJTIFEÛXJUIÛ DFSUBJOUZ�Û$GÛ BMTPÛ POÛ UIFÛ EFWFMPQNFOUÛ PGÛUIFÛDMVTUFSTÛ� T�BOEÛ�L T�UPÛu�JOÛ��DÛBOEÛ��E�ÛÛÛ�TLFE��Û uFSUÛ �DIJQ��Û (S�Û �SKI�D NHMI Û BPS�Û �SKED A�SAIÛ �TDBUUFS Û EJTQFSTF��Û5IFÛ DPOOFDUJPOTÛ PGÛ UIJTÛ SPPUÛ BSFÛ EJGGJDVMU�Û 5IFÛNFBOJOHÛ EPFTÛ OPUÛ BHSFFÛXFMMÛXJUIÛUIBUÛPGÛUIFÛ"SNFOJBOÛXPSE�Û-JUI�ÛLFEFSW¹Û�TQMJOUFS�ÛBHSFFTÛXFMM ÛCVUÛJUTÛSPPUÛNVTUÛIBWFÛCFFOÛ�TLFEI�ÛCFDBVTFÛPGÛ8JOUFS�TÛMBX�Û,PSUMBOEUÛ��B Û��Û<UIJTÛWPM� Û��>�ÛÛÛ�TLFM��ÛuF Û�PCMJRVF ÛBTMBOU��Û(S�ÛSKOLIO�JÛ�DVSWFE ÛCFOU� Û0)(ÛTDFMBI�ÛÛÛÛ ��0%6=2+)%07�� ;SVH�MRMXMEP�PEV]RKIEP��ÛÛÛÛÛ ��ÛÛ8PSE�JOJUJBMÛMBSZOHFBMÛCFGPSFÛWPXFM Û)7�ÛÛÛÛÛ 5IFÛ EJTDPWFSZÛ UIBUÛ "SN�Û I�Û JOÛ TPNFÛ DBTFTÛ SFGMFDUTÛ BOÛ PSJHJOBMÛMBSZOHFBMÛ JTÛQFSIBQTÛFWFOÛNPSFÛ JNQPSUBOUÛ UIBOÛ UIFÛSFDPHOJUJPOÛ UIBUÛ UIFÛPMEÛ �QSPUIFUJDÛ WPXFM�Û XIJDIÛ BHSFFTÛ XJUIÛ UIFÛ (SFFLÛ POFÛ EFSJWFTÛ GSPNÛ BÛMBSZOHFBMÛPOÛXIJDIÛTFFÛUIFÛOFYUÛTFDUJPO Û��Û*UÛXBTÛOPUBCMZÛ(SFQQJOÛXIPÛGJSTUÛBSHVFEÛGPSÛUIJTÛQIFOPNFOPO�ÛÛÛÛÛÛÛ, � EJTBQQFBST ÛCVUÛ, � BOEÛ, � CFGPSFÛF�BSFÛSFQSFTFOUFEÛBTÛI�Û ,PSUMBOEUÛ��CÛ BOEÛ ��C�Û *UÛ JTÛ SFNBSLBCMFÛ UIBU Û UIPVHIÛ "SNFOJBOÛ IBTÛ CVUÛ BÛMJNJUFEÛOVNCFSÛPGÛSFMJBCMFÛFUZNPMPHJFT ÛUIJTÛDPOUSJCVUJPOÛDBOÛUPÛNZÛNJOEÛCFÛDPOWJODJOHMZÛEFNPOTUSBUFE�ÛÛÛÛÛ"GUFSÛ BÛ GFXÛ JOUSPEVDUPSZÛ SFNBSLTÛ XFÛ TIBMMÛ EJTDVTTÛ UIFÛ FWJEFODFÛ QFSÛMBSZOHFBMÛJOÛUIFÛGPMMPXJOHÛPSEFS�ÛÛÛÛ, � �ÛÛCFGPSFÛF�ÛCFGPSFÛP��ÛCFGPSFÛ$ÛBOEÛ3$�ÛÛÛÛ, � �ÛÛÛÛ ÛÛÛÛÛÛÛÛÛÛÛÛÛÛ ÛÛÛÛÛÛÛÛÛÛÛÛÛÛ ÛÛÛÛ, � �ÛÛÛÛ ÛÛÛÛÛÛÛÛÛÛÛÛÛÛ ÛÛÛÛÛÛÛÛÛÛÛÛÛÛ ÛÛ"UÛUIFÛFOEÛTPNFÛTQFDJBMÛQSPCMFNTÛXJMMÛCFÛEJTDVTTFE�ÛÛÛÛÛÛÛÛ1SFMJNJOBSZÛSFNBSLT�ÛÛÛÛÛÛ4PNFÛQSFMJNJOBSZÛSFNBSLTÛNVTUÛCFÛNBEF�Û*UÛ JTÛTPNFUJNFTÛTUBUFEÛUIBUÛ"SN�Û I�Û JTÛ VOSFMJBCMF Û CFDBVTFÛ UIFSFÛ BSFÛ XPSETÛ XIJDIÛ PDDVSÛ XJUIÛ BOEÛXJUIPVUÛJU�Û)PXFWFS ÛUIJTÛDPODFSOTÛBÛWFSZÛTNBMMÛOVNCFSÛPGÛXPSET ÛXIJMFÛUIFÛMBSHFÛNBKPSJUZÛFJUIFSÛEPÛIBWFÛI�ÛPSÛOPU�ÛJUÛJTÛDMFBSMZÛQIPOFNJD�Û*UÛNVTUÛBMTPÛCFÛSFBMJ[FEÛUIBUÛJUÛJTÛQSFDJTFMZÛUIFÛI�ÛGSPNÛMBSZOHFBMÛXIJDIÛHBWFÛUIFÛ

180� 5��6��3��%HHNHV��

JNQSFTTJPOÛ UIBUÛ UIFÛPDDVSSFODFÛXBTÛ JOFYQMBJOBCMF�Û *OÛ GBDU Û UIJTÛIBTÛCFFOÛDJUFEÛ BTÛ POFÛ PGÛ UIFÛ NPTUÛ IPQFMFTTÛ QIFOPNFOBÛ PGÛ "SNFOJBOÛ IJTUPSJDBMÛMJOHVJTUJDT�ÛÛÛÛÛ*UÛ IBTÛ CFFOÛ QPJOUFEÛ PVUÛ UIBUÛ JOÛ.JEEMFÛ "SNFOJBOÛ BOÛ I�Û JTÛ BEEFEÛ UPÛXPSETÛ XIJDIÛ EJEÛ OPUÛ IBWFÛ JUÛ FBSMJFS�Û )PXFWFS Û UIJTÛ EPFTÛ OPUÛ DPODFSOÛ$MBTTJDBMÛ"SNFOJBO ÛXIJDIÛXFÛEJTDVTTÛ IFSF�Û.PSFPWFS Û UIFÛ TJUVBUJPOÛ JOÛ.JEEMFÛ"SNFOJBOÛJTÛMBSHFMZÛDMBSJGJFE�ÛÛÛÛÛ'VSUIFSÛ JUÛ NVTUÛ CFÛ OPUFEÛ UIBUÛ JOJUJBMÛ I�Û EJTBQQFBSTÛ XIFOÛ TPNFUIJOHÛDPNFTÛ CFGPSFÛ JUÛ DG�Û &OH�Û TIFQIFSE�Û 5IVTÛ JOÛ UIFÛ DBTFÛ PGÛ SFEVQMJDBUJPO�ÛIPUPUJN��TNFMM�Û��IPU�IPU��Û'VSUIFS ÛBGUFSÛUIFÛGJSTUÛNFNCFSÛPGÛBÛDPNQPVOE�ÛKF O�BUÛ �XJUIÛIJTÛIBOEÛDVUÛPGG� ÛGSPNÛIBOU�Û �DVU��Û"GUFSÛ[ÛBOEÛZÛBOÛI�ÛBMTPÛEJTBQQFBST�Û[JO Û �XIBU�Û GSPNÛUIFÛSPPUÛTFFOÛJOÛIJNÛ �XIZ��ÛZ�FUÛ �BGUFS�Û GSPNÛIFUÛ �USBDF��Û *UÛ JTÛ DPODFJWBCMFÛ UIBUÛ TPNFUJNFTÛ TVDIÛ BÛ GPSNÛXJUIPVUÛI�ÛXBTÛHFOFSBMJ[FE�ÛÛÛÛÛÛÛ8IFOÛ BÛXPSEÛ PDDVSTÛXJUIÛ BOEÛXJUIPVUÛ BOÛ I� Û JUÛ JTÛ QSPCBCMFÛ UIBUÛ UIFÛGPSNÛXJUIPVUÛI�ÛBSPTFÛJOÛUIJTÛXBZ ÛPSÛUISPVHIÛJOGMVFODFÛPGÛSFMBUFEÛGPSNTÛXJUIPVUÛI�ÛJOÛUIFÛQBSBEJHNÛPSÛPVUTJEFÛPGÛJU�ÛÛÛÛÛ/PUFÛ UIBUÛ1*&ÛEJEÛOPUÛLOPXÛBÛQVSFMZÛWPDBMJDÛBOMBVU�ÛFWFSZÛXPSEÛ UIBUÛTFFNTÛUPÛCFHJOÛXJUIÛBÛWPXFMÛJOÛGBDUÛIBEÛBÛQSFDFEJOHÛMBSZOHFBM�ÛÛÛÛÛÛ5IFÛSFGMFYFT ÛUIFO ÛBSFÛBÛGPMMPXT�ÛÛÛÛ�, � F��F�� , � P��P��, � F��IB� , � P��P��, � F��IP� , � P��P��ÛÛÛ5IFÛQJDUVSFÛJTÛNVDIÛDPNQMJDBUFEÛCZÛUIFÛGBDUÛUIBUÛQSFUPOJDÛPÛPGUFOÛCFDBNFÛB�Û 5IFÛ DPOEJUJPOTÛ NBZÛ CFÛ TVNNBSJ[FEÛ IFSFÛ BTÛ GPMMPXT�Û PÛ CFDBNFÛ BÛ JOÛQSFUPOJDÛTZMMBCMFT ÛFYDFQU�Û��ÛCFGPSFÛUXPÛDPOTPOBOUT�Û��ÛCFGPSFÛPÛJOÛUIFÛOFYUÛTZMMBCMF�Û ��Û JNNFEJBUFMZÛ CFGPSFÛ HÛ �Û � � �� X�Û *OÛ UIJTÛ XBZ Û IPÛ DPVMEÛCFDPNFÛ IB�Û Û "Û TFDPOE Û NJOPSÛ DPNQMJDBUJPOÛ JTÛ UIBUÛ PÛ CFDBNFÛ VÛ CFGPSFÛOBTBM ÛBTÛJOÛVTÛ�TIPVMEFS�Û�Û�PNT��ÛÛ'VSUIFSÛXFÛIBWFÛJOJUJBMÛ�PJ��ÛBZ�ÛXIJDIÛ,PSUMBOEUÛEBUFTÛWFSZÛFBSMZ�ÛÛ"MTPÛXFÛGJOEÛ�PQÛ�Û�PXÛ�ÛV��8FÛBMTPÛIBWFÛUPÛSFDLPOÛXJUIÛF��JÛCFGPSFÛOBTBM ÛBTÛJTÛGPVOEÛJOÛJOOÛ�OJOF�Û�Û�, � O��ÛÛÛÛ*UÛ JTÛ FTTFOUJBMÛ UIBUÛ UIFSFÛ XBTÛ BOÛ PQQPTJUJPOÛ CFUXFFOÛ , � FÛ BOEÛ , � P�Û5IPVHIÛ UIFÛ SFGMFYFTÛPGÛ UIFÛWPXFMTÛ BSFÛ UIFÛ TBNFÛ JOÛNPTUÛ MBOHVBHFT Û UIFÛUXPÛXFSFÛEJTUJOHVJTIFEÛJOÛ1*&ÛBOEÛUIFZÛSFNBJOFEÛTPÛJOÛUIFÛFBSMZÛTUBHFTÛPGÛ UIFÛ TFQBSBUFÛ MBOHVBHFT�Û 5IJTÛ JTÛ EFNPOTUSBUFEÛ CZÛ UIFÛ MJNJUBUJPOÛ PGÛ#SVHNBOO�TÛ -BXÛ UPÛ �BCMBVUJOH�P� Û BTÛ PQQPTFEÛ UPÛ �OPO�BCMBVUJOH�P� Û J�F�Û, � F�Û5IJTÛMJNJUBUJPOÛXBTÛBMSFBEZÛGPSNVMBUFEÛCZÛ#SVHNBOOÛIJNTFMGÛXIPÛDPVMEÛOPUÛZFUÛTUBUFÛUIFÛQIFOPNFOPOÛJOÛMBSZOHFBMÛUFSNTÛBOEÛSFEJTDPWFSFEÛCZÛ-VCPUTLZ�Û&�H�Û �, � FQ�PTÛ Û�Û4LU�ÛºQBT ÛOPUÛ � QBT�Û"ÛEJTUJODUJPOÛCFUXFFOÛFÛ BOEÛ PÛ JTÛ BUUFTUFEÛ JOÛ MJWJOHÛ MBOHVBHFT Û BTÛ JOÛ /PSUIÛ "NFSJDBOÛ4IVTXBQ�Û


ÛÛÛ5IFÛ SFGMFYFTÛXFSFÛ FYQMBJOFEÛ CZÛ ,PSUMBOEUÛ BTÛ GPMMPXT�Û *OÛ 1*& Û BÛXPSEÛDPVMEÛ OPUÛ CFHJOÛ XJUIÛ F� Û CVUÛ IBEÛ BÛ QSFDFEJOHÛ , � ��Û 5IFÛ MBSZOHFBM ÛUIFSFGPSF ÛJTÛBVUPNBUJD ÛIBTÛOPÛQIPOFNJDÛWBMVF�Û*OÛUIFÛDBTFÛPGÛBÛGPMMPXJOHÛP Û BÛ MBSZOHFBMÛ XBTÛ BVUPNBUJDBMMZÛ SPVOEFE Û XIJDIÛ JTÛ XIZÛ UIFÛ UISFFÛMBSZOHFBMTÛ NFSHFE�Û )FSFÛ BHBJOÛ UIFÛ MBSZOHFBMÛ JTÛ BVUPNBUJD Û BTÛ BÛ XPSEÛDPVMEÛ OPUÛ IBWFÛ KVTUÛ P� Û CVUÛ �BVUPNBUJDBMMZ�Û IBEÛ BÛ QSFDFEJOHÛ SPVOEFEÛMBSZOHFBM�Û *UÛ JTÛ UIFSFGPSFÛ VOEFSTUBOEBCMFÛ UIBUÛ UIFÛ BVUPNBUJDÛ MBSZOHFBMTÛXFSFÛOPUÛSFGMFDUFEÛBTÛI�ÛJOÛ"SNFOJBO�ÛÛÛÛÛ8FÛTIBMMÛOPXÛMPPLÛBUÛUIFÛFWJEFODF�ÛÛÛÛÛÛ, � ÛCFGPSFÛFÛÛÛÛÛ5IFSFÛJTÛOPÛFWJEFODFÛUIBUÛ, � �ÛFWFSÛSFTVMUFEÛJOÛI��Û$PNQBSF�ÛÛÛÛ�, � F� ÛUIFÛBVHNFOU�ÛF�LOÛ�IFÛDBNF��Û(S�Û| �BH Û4LU�Ûº�HBO�ÛÛÛÛ�, � F�Û EFJDUJDÛ QBSUJDMF�Û F�U FÛ �UIBU Û XIFO� Û CFTJEFÛ U F�Û (S�Û y KEI>NOJ Û 3VTT�ÛÁ�UPU Û4LU�ÛB�TºV�ÛÛÛÛ�, � FTNJ�ÛFNÛ�*ÛBN��Û4LU�ÛºTNJ Û(S�ÛEI�MI��ÛÛÛÛ�, � F �Û�ÛFTÛ�*��Û(S�ÛyGW� Û-BU�ÛFH �ÛÛÛÛÛÛÛ�, � FQJ�ÛFXÛ�BOE��Û(S�ÛyPI� Û4LU�ÛºQJ�ÛÛÛÛ�, � FSL PT�ÛFSHÛ�TPOH��Û4LU�ÛBSLº��ÛÛÛÛ�, � F I��ÛF[SÛ�CPSEFS�FEHF ÛTIPSF��Û-JUI�ÛF ÛÛ�CPSEFS ÛGSPOUJFS��ÛÛÛÛ�, � FM, � �FO��ÛFMOÛ�IJOE ÛSPF��Û(S�Û|LAFOJ Û|LLOJ Û-JUI�ÛÂMOJT Û0$4ÛKFMFO �ÛÛÛ�, � FM�FV��ÛF FXJOÛ�DFEBS��Û3VTT�ÛKBMPWFDÛ�KVOJQFS�Û�Û� M� ÛDG�Û(S�ÛyLA�TH�ÛÛÛÛÛÛ0UIFSÛ XPSETÛ CFHJOOJOHÛ XJUIÛ F�Û GPSÛ XIJDIÛ BOÛ FUZNPMPHZÛ IBTÛ CFFOÛTVHHFTUFEÛ IBWFÛ BÛ QSPUIFUJDÛ WPXFMÛ GSPNÛ BÛ MBSZOHFBMÛ PSÛ UIFÛ MBUFSÛ"SNFOJBOÛ POF Û PSÛ HPÛ CBDLÛ UPÛ GPSNTÛ XJUIÛ T��Û 8FÛ DPODMVEFÛ UIBUÛ, � �Û JTÛOFWFSÛSFQSFTFOUFEÛCZÛ"SN�ÛI��Û5IFÛDPOOFDUJPOÛPGÛIF Û�EJTDPSE ÛTUSJGF�ÛXJUIÛ(S�Û|RIJÛNVTUÛUIFSFGPSFÛCFÛHJWFOÛVQ�Û5IJTÛXPSEÛNVTUÛIBWFÛIBEÛ�QFS��ÛÛÛÛÛ, � ÛCFGPSFÛP�ÛÛÛÛÛ�, � P I�J)�O��Û P[OJÛ �IFEHFIPH��Û (S�Û yCI>NOJ Û -JUI�Û F ZT Û 0$4Û KF Û 0)(ÛJHJM�ÛÛÛ�, � PST��ÛP Û �SVNQ��Û(S�Û´RROJ Û *PO�ÛO�RSOPU�GION Û)JUU�ÛBSSBT Û0)(ÛBST�Û UIFÛXPSEÛIBEÛ, � �ÛJGÛ0*S�ÛFSSÛ�UBJM�ÛCFMPOHTÛIFSF�ÛÛÛÛÛÛÛ,PSUMBOEUÛ QSPQPTFTÛ UPÛ EFSJWFÛVUFNÛ �FBU�Û GSPNÛ �TPN� ��, � FE�Û ��B Û ��Û<UIJTÛWPM� Û��>�ÛÛÛÛÛÛÛ, � ÛCFGPSFÛ$ÛBOEÛ3$ÛÛÛÛ4FFÛ��ÛBOEÛ��ÛÛÛÛÛ, � ÛCFGPSFÛFÛÛÛÛÛ, � F�Û TFFNTÛ UPÛ CFÛ SFGMFDUFEÛ CZÛ IB��Û *OÛ UIFÛ GPMMPXJOHÛ DBTFTÛ UIJTÛ

182� 5��6��3��%HHNHV��

SFDPOTUSVDUJPOÛTFFNTÛDMFBS�ÛÛÛÛ�, � FO��ÛIBOÛ�HSBOENPUIFS��Û-BU�ÛBOVT Û)JUU�ÛIBOOBT�ÛÛÛÛ�, � FV)�PT�ÛIBXÛ�HSBOEGBUIFS��Û-BU�ÛBWVT Û)JUU�ÛIVIIBT�ÛÛÛÛ�, � FV�JT�ÛIBXÛ�CJSE��Û-BU�ÛBWJT�ÛÛÛÛ�, � FV��ÛIBHBDJNÛ�CFDPNFÛBEEJDUFEÛUP��Û-BU�ÛBWF �ÛÛÛÛ�, � FJ��ÛIBZD FNÛ�CFH ÛCFTFFDI��Û-BU�ÛBFSVTD SF�Û-JUI�ÛJFuLÌUJ�ÛÛÛ"OEÛQSPCBCMZÛBMTP�ÛÛÛÛ�, � FE��ÛIBUBOFNÛ�DVU��Û)JUU�ÛIBUUBJ�ÛÛÛÛ�, � FS, � �VS Û�VFO��ÛIBSBXVOL Û�GJFME��Û(S�Û�ROURA�Û0*S�ÛBSCPS�ÛÛÛÛ�, � FO ��Û IBT�BOFNÛ �BSSJWFÛ BU Û PCUBJO��Û (S�Û �POD HNEKH�JÛ �SFBDIJOHÛ UPÛ UIFÛGFFU�Û Û �Û Û 0*S�Û QFSG�Û ºOBJDÛ �SFBDIFE��Û 5IJTÛ SPPUÛ FYJTUFEÛ CFTJEFÛ �, � OF �Û�CSJOH� Û (S�Û BPS�Û �NEGK ON�Û yNEGK EI>N�Û ;FSPÛ HSBEFÛ �, � O �� NBZÛ BMTPÛ CFÛQPTTJCMF�ÛÛÛ�, � FVT�T�JÛ MPD��Û BZHÛ �EBXO��Û UIFÛ XPSEÛ JTÛ TVQQPTFEÛ UPÛ EFSJWFÛ GSPNÛFYQSFTTJPOTÛ MJLFÛ �BUÛ EBXO� Û DG�Û (S�Û �I KANO�JÛ �DPDL�Û �Û ��TJOHJOHÛ BUÛ EBXO��Û5IFÛBCTFODFÛPGÛI�Û JTÛFYQMBJOFEÛGSPNÛUIFÛQPTJUJPOÛBGUFSÛQSFQPTJUJPO ÛBTÛ JOÛOE�BZHÛ�BUÛEBXO��ÛÛÛÛÛÛÛÛÛÛÛ, � ÛCFGPSFÛPÛÛÛÛÛ5IFÛNBJOÛQPJOUÛJTÛUPÛFYQMBJOÛXIZÛNBOZÛGPSNTÛXJUIÛ, � ÛIBWFÛB�ÛXJUIPVUÛI��Û /PUFÛ UIBUÛ IB�ÛNBZÛ JOÛ TPNFÛDBTFTÛ DPOUJOVFÛPMEFSÛ �)P��Û5IFSFÛBSFÛUXPÛ QPTTJCJMJUJFT�Û P�HSBEFÛ PSÛ [FSPÛ HSBEF�Û 8FÛ TIBMMÛ TFFÛ CFMPXÛ UIBUÛ UIFÛTBNFÛIPMETÛGPSÛ, � �ÛÛÛÛÛ*GÛ, � ÛCFGPSFÛ1*&Û�PÛHBWFÛOPÛSFGMFY ÛXFÛDBOÛFYQMBJO�ÛÛÛ�, � PJE��Û BZUOVNÛ �TXFMM� Û BZUÛ �DIFFL��Û (S�ÛOI�DE�W�Û UIFÛ, � Û JTÛ TIPXOÛ CZÛ-BU�ÛBFNJEVTÛ�UVNJEVT ÛJOGM UVT�ÛHMPTT ÛJGÛUIJTÛJTÛSFMBUFEÛ�BJE�TN��ÛÛÛ�, � PJ�TL��ÛBZD Û �WJTJU Û JOTQFDUJPO��Û0)(ÛFJTD O�Û0MEÛ �TUFNTÛPGUFOÛIBWFÛP�WPDBMJTN�ÛÛÛ5IFÛXPSEÛDPOUBJOTÛUIFÛTBNFÛSPPUÛBTÛIBZD FNÛHJWFOÛBCPWF�Û*UÛJTÛOPÛTVSQSJTFÛUIBUÛUIFSFÛJTÛBMTPÛBÛGPSNÛIBZD ÛXJUIÛI�ÛGSPNÛUIFÛWFSC�ÛJUÛHPUÛBÛTMJHIUMZÛEJGGFSFOUÛNFBOJOH Û�JORVJSZ��ÛÛÛ�, � PNT�P��Û VTÛ �TIPVMEFS��Û (S�Û ÎMOJÛ �Û �PNTPT Û 4LU�Û ºNTB� Û (PU�Û BNT�Û JGÛ5PDI�Û"Û FT Û#Û OUTFÛQPJOUÛ UPÛB�WPDBMJTN Û UIFÛ SPPUÛNVTUÛIBWFÛ, � ��Û DG�Û BMTPÛ(S�Û�ME�SW �W�MOPL�TAI�HMPTT ÛPSJHJOÛVOLOPXO�ÛÛÛÛÛ, � ÛCFGPSFÛ$ÛBOEÛ3$ÛÛÛÛ*OÛ UIFÛ DBTFÛ PGÛ [FSPÛHSBEFÛXFÛNVTUÛEJTUJOHVJTIÛ BUÛ MFBTUÛ UXPÛ DBTFT�Û)$�ÛBOEÛ)3�ÛXJUIÛ3ÛUSBEJUJPOBMMZÛDPOTJEFSFEÛTZMMBCJD�Û5IFÛGJSTUÛSFTVMUTÛJOÛUIFÛTP�DBMMFEÛ QSPUIFUJDÛ WPXFM��Û $G�Û BZSÛ �NBO�Û �Û �, � O S�Û *OÛ UIJTÛ XBZÛ DBOÛ CFÛFYQMBJOFE�ÛÛÛÛ�, � �FT��Û BDFNÛ �DBSSZ��Û -BU�Û HFS� �Û OPUÛ UPÛ CFÛ DPNQBSFEÛ XJUIÛ -BU�Û BH Û �Û�, � F ��Û


ÛÛ'PSÛ)3�ÛJ�F�Û)3�$ÛTFFÛ��Û5IJTÛTFRVFODFÛSFTVMUTÛJOÛ73�$�Û$G�ÛÛÛ�, � SU �P��ÛBS Û�CFBS��Û4LU�Û L B� Û)JUU�ÛIBSUBHHB� Û(S�Û�RKTOJ�ÛÛÛÛ�, � S ��ÛBSDBU Û�TJMWFS��Û"W�Û S [BUB��Û4LU�ÛSBKBUºN�Û-BU�ÛBSHFOUVN�ÛÛÛÛ�, � SL��ÛBSHFMÛ�IJOESBODF ÛPCTUBDMF��ÛDG�Û(S�Û�RKE�W�ÛÛÛ'PSÛBSEÛ�GPSN�ÛTFFÛ��ÛPOÛ)3$��ÛÛÛÛÛÛ"Û TQFDJBMÛ TVCDBUFHPSZÛ BSFÛ UIFÛ XPSETÛ CFHJOOJOHÛ XJUIÛ )3)�$Û TFFÛ��Û5IFTFÛHJWFÛB3B$��ÛÛÛ�, � S, � USPN�ÛBSBXSÛ�QMPVHI��Û(S�Û��ROTRON Û0*S�ÛBSBUIBS�Û5IBUÛUIJTÛUZQFÛIBEÛ[FSPÛHSBEFÛJTÛTIPXOÛCZÛ-JUI�ÛÅSLMBT��PBS�Û�Û�, � S, � ��ÛÛÛ�, � M, � US��ÛB BXS�JÛ�GFNBMFÛXIPÛHSJOETÛDPSO��Û(S�Û�LETR I�JÛÛJE�ÛÛÛÛÛ/PUFÛUIBUÛJOÛUIJTÛXBZÛUIFÛEJGGFSFODFÛCFUXFFOÛIBSBXVOL Û�GJFME�ÛBOEÛBSBXSÛ�QMPVHI�ÛJTÛFYQMBJOFE�Û5IFSFÛJTÛBÛGPSNÛIBS SÛCFTJEFÛBS S ÛBSBXS ÛXIJDIÛNBZÛIBWFÛJUTÛI�ÛGSPNÛBÛSFMBUFEÛGPSNÛXJUIÛ�, � FS, � ��ÛÛÛÛÛÛÛÛ�, � �CFGPSFÛFÛÛÛÛ, � FÛ JTÛQSPCBCMZÛSFQSFTFOUFEÛCZÛIP��Û5IFÛUXPÛXPSETÛXJUIÛIP�Û UIBUÛIBWFÛBOÛFUZNPMPHZÛQSPCBCMZÛIBEÛ, � F��ÛÛÛ�, � FE��Û IPUÛ �TNFMM��Û (S�Û ´ZW Û -BU�Û PEPSÛ �, � �FE� T Û -JUI�Û ÓPE JVÛ �Û �, � FE�ÛXJUIÛMPOHÛWPXFMÛBOEÛBDVUFÛJOUPOBUJPOÛBDDPSEJOHÛUPÛUIFÛ8JOUFS�,PSUMBOEUÛ-BX�Û 5IFÛ FYDMVTJWFÛ P�WPDBMJTNÛ QPJOUTÛ UPÛ , � ��Û *GÛ UIFÛ "SNFOJBOÛ XPSEÛDPOUJOVFTÛBÛOFVUFSÛT�TUFN ÛXFÛFYQFDUÛF�WPDBMJTN�ÛÛÛÛÛ�, � FVJ��Û IPWJXÛ �TIFQIFSE��Û (S�Û ÍJ Û -BU�Û PWJT Û -VX�Û IBXJ�Û XIJDIÛEFNPOTUSBUFTÛ, � Û PSÛ, � �Û UIFÛ FYDMVTJWFÛ P�WPDBMJTNÛ GBWPVSTÛ, � Û XIJDIÛ JTÛDPOGJSNFEÛ CZÛ 4LU�Û ºWJ�Û BTÛ UIJTÛ EPFTÛ OPUÛ IBWFÛ BÛ MPOHÛ WPXFMÛ GSPNÛ#SVHNBO�TÛ -BX Û XIJDIÛ EPFTÛ OPUÛ PQFSBUFÛ POÛ �P�Û GSPN�, � FÛ BDDPSEJOHÛ UPÛ-VCPUTLZ�TÛ-BX�ÛÛÛ�, � FTL��IBD JÛ�BTIÛUSFF��Û(S�Û±XU�H �Û0*D�ÛBTLS�ÛÛÛÛ8FÛDBOÛOPXÛFYQMBJOÛIVN�BTÛGPMMPXT�ÛÛÛ�, � F)NPT Û�, � F, � NPT�ÛIVNÛ�SBX��Û(S�ÛW�MO�J Û4LU�Û Nº��Û5IFÛXPSEÛEJEÛOPUÛIBWFÛMFOHUIFOFEÛHSBEFÛCVUÛSBUIFSÛ)7)��Û5IFOÛXFÛOFFEÛF�WPDBMJTNÛUPÛHFUÛUIFÛI�ÛXJUIÛ, � �ÛUIFÛJOJUJBMÛMBSZOHFBMÛNVTUÛCFÛ, � ÛPSÛ, � �ÛÛÛÛÛÛ0UIFSÛXPSETÛXJUIÛIV�ÛEFSJWFÛGSPNÛ�QP�Û�ÛOBTBM�ÛÛÛÛÛÛ, � ÛCFGPSFÛPÛÛÛÛÛ, � �ÛGPMMPXFEÛCZÛPÛJTÛWFSZÛEJGGJDVMUÛUPÛEFNPOTUSBUF�ÛÛÛ�, � PSCI��ÛPSCÛ �PSQIBO��Û(S�Û±RFANO�J�Û-BU�ÛPSCVT Û(PUI�ÛBSCJ Û0*S�ÛPSCÛTIPXÛGVMMÛ HSBEF Û BOEÛ UIFÛ HFOFSBMÛ P�WPDBMJTNÛ TVHHFTUTÛ , � ��Û #VUÛ [FSPÛ HSBEFÛ, � SCI�ÛJTÛBMTPÛQPTTJCMFÛBTÛJTÛ�, � PSCI�ÛPSÛ�, � PSCI��ÛÛÛ�, � PTE��ÛPTUÛ�CSBODI��Û(S�Û´ZOJ Û)JUU�ÛIBTEVFJS�Û(PUI�ÛBTUTÛTIPXTÛGVMMÛHSBEF ÛBOEÛUIFÛHFOFSBMÛP�WPDBMJTNÛTVHHFTUTÛ, � �ÛCVUÛ�, � TE�ÛJTÛBMTPÛQPTTJCMF�ÛÛÛ�, � PTU�V��ÛPTLSÛ�CPOF��Û(S�Û±STE�ON�Û)JUU�ÛIBTUBJ�Û4LU�ÛºTUIJÛTIPXTÛGVMMÛHSBEF�Û

184� 5��6��3��%HHNHV��

CVUÛ�, � TU�ÛJTÛBMTPÛQPTTJCMF�ÛÛÛ�, � PJV�F, � �� BZHJÛ �WJOF��Û (S�Û ´ �� OgHÛ �TFSWJDF�USFF� Û -BU�Û WBÛ �HSBQF� Û -JUI�ÛJFW¹ ÛBMMÛGSPNÛ�PJV �Û5IFÛMBSZOHFBMÛDBOOPUÛCFÛBTDFSUBJOFE�ÛÛÛÛ�, � PVH��ÛPZDÛ �DPME��Û0*S�ÛÓBDIUÛ �PVH�UV�Û �DPMEOFTT� Û-JUI�ÛºVuUJ�Û5IFÛSPPUÛNBZÛIBWFÛIBEÛ, � ��ÛÛÛÛÛÛ, � �CFGPSFÛ$ÛBOEÛ3$ÛÛÛÛÛ"QBSUÛGSPNÛUIFÛDBTFTÛXJUIÛ�QSPUIFUJDÛWPXFM�ÛTFFÛUIFÛOFYUÛTFDUJPO ÛOPUF�ÛÛÛ�, � U��ÛVU Û�Û�, � QU�Û�FJHIU��ÛUIFÛGPSNÛIBTÛUIFÛ[FSPÛHSBEFÛGSPNÛUIFÛPSEJOBMÛBOEÛ �Q�Û GSPNÛ �TFWFO��Û �PQU�Û �Û �PXU�Û �ÛVU �Û(S�Û±KTW� Û -BU�Û PDU Û 4LU�Û B V Û(PUI�ÛBIUBV�ÛÛÛÛ�, � S IJ��ÛPSKJL Û �UFTUJDMFT��Û.*S�ÛVJSHHF Û"MC�ÛIFSEIÄ�Û"W�Û S [J�Û5IFÛHFOFSBMÛP�WPDBMJTNÛ-JUI�ÛFS JMBTÛDBOÛCFÛGPSÛPMEFSÛP�ÛQPJOUTÛ UPÛ, � � ÛCVUÛCPUIÛ[FSPÛHSBEFÛBOEÛP�HSBEFÛBSFÛQPTTJCMF�ÛÛÛÛÛÛÛÛÛÛÛÛ4QFDJBMÛQSPCMFNTÛÛÛÛÛ*OÛNBOZÛDBTFT Û JUÛ JTÛ JNQPTTJCMFÛBTÛZFUÛ UPÛEFUFSNJOFÛUIFÛDPMPVSÛPGÛ UIFÛMBSZOHFBM Û BOEÛ PGUFOÛ BMTPÛ XIJDIÛ WPDBMJTN Û TPÛ UIBUÛ XFÛ PGUFOÛ DBOOPUÛEFNPOTUSBUFÛXIBUÛIBQQFOFE�ÛÛÛÛÛÛ4PNFÛDBTFTÛBSFÛSBUIFSÛDPNQMJDBUFE�ÛÛÛ�, � F, � N S�ÛBXSÛ�EBZ��Û(S�Û£MAR��H@ME�R �Û'PSÛUIFÛMPOHÛ ÛXFÛDBOÛIBWFÛ, � FÛ�ÛBÛ MBSZOHFBM Û CVUÛ UIBUÛ XPVMEÛ HJWFÛ I��Û 8FÛ NVTUÛ TUBSUÛ GSPNÛ �F, � �Û XJUIÛ BÛQSFDFEJOHÛMBSZOHFBMÛUIBUÛEPFTÛOPUÛHJWFÛI� ÛXIJDIÛDBOÛPOMZÛCFÛ, � ��ÛÛÛÛ"SN�ÛBUFBNÛ �IBUF� Û-BU�ÛPEJVN Û0&ÛBUPMÛ �UFSSJCMF�ÛXJMMÛIBWFÛ, � �ÛCFDBVTFÛPGÛJUTÛ HFOFSBMÛ P�WPDBMJTN Û JOÛXIJDIÛ DBTFÛ JUÛ DBOÛ CFÛ �, � PE�Û PSÛ �, � E� Û CVUÛXFÛIBWFÛOPÛEFGJOJUFÛQSPPG�Û, � P�ÛPSÛ, � P�ÛBSFÛBMTPÛQPTTJCMF�ÛÛÛ"ÛQSPCMFNÛ JTÛ QSPWJEFEÛ CZÛ BZDÛ �HPBU� Û(S�ÛAI��X Û BTÛ UIJTÛ DPVMEÛ POMZÛ IBWFÛCFFOÛ�, � FJ � ÛCVUÛUIJTÛGPSNÛXPVMEÛIBWFÛHJWFOÛ"SN�ÛI��Û"ÛGVSUIFSÛDPHOBUFÛIBTÛCFFOÛTVTQFDUFEÛJOÛ"W�ÛJ� [B OB�Û�PGÛMFBUIFS� ÛCVUÛUIJTÛNBZÛCFÛVOSFMBUFE�Û5IFSFÛIBTÛCFFOÛNVDIÛEPVCUÛBCPVUÛUIJTÛXPSEÛBOEÛPUIFSTÛGPSÛ�HPBU� ÛBOEÛUIFÛXPSEÛNBZÛCFÛBÛDPNNPOÛMPBO�ÛÛÛÛÛÛ4JODFÛPHJ�IPHJÛ �CSFBUI Û TQJSJU Û TPVM�ÛEPFTÛOPUÛIBWFÛBÛ HFOFSBMMZÛBDDFQUFEÛFUZNPMPHZ Û XFÛ DBOÛ TBZÛ OPUIJOHÛ BCPVUÛ JU�Û *UÛ NBZÛ CFÛ SFMBUFEÛ UPÛ IFXBNÛ�QBOU� ÛXIJDIÛNVTUÛIBWFÛIBEÛ�Q��Û5IFÛIÛ�Û�QÛXBTÛMPTUÛCFGPSFÛ�P� ÛTPÛUIFÛI�ÛNVTUÛ IBWFÛ CFFOÛ JOUSPEVDFEÛ GSPNÛ SFMBUFEÛ GPSNTÛXJUIÛI�Û GSPNÛ �Q�Û CFGPSFÛPUIFSÛWPXFMÛXIJDIÛNVTUÛCFÛ�F��ÛÛÛÛÛÛ"OPUIFSÛDBTFÛ JTÛBHBOJNÛ �TQFOEÛ UIFÛOJHIU� Û(S�Û I�AU�W�Û5IFÛBCTFODFÛPGÛI�ÛNBZÛCFÛFYQMBJOFEÛUISPVHIÛJOGMVFODFÛPGÛBXU Û �QMBDFÛUPÛTQFOEÛUIFÛOJHIU�Û JGÛGSPNÛ�, � PVUJ�ÛDG�Û(S�Û�ULH�Û�DPVSUZBSE�ÛXIJDIÛNVTUÛIBWFÛ�, � FV��ÛÛÛÛÛÛ'PSÛIBHBOJNÛ�QVUÛPOÛDMPUIFT�ÛGSPNÛ�, � FV��, � PV�ÛDG�Û-JUI�ÛBVOÒ �B UJÛ�QVUÛPOÛGPPUXFBS��Û


ÛÛ ��ÛÛÛ8PSEÛJOJUJBMÛMBSZOHFBMÛCFGPSFÛDPOTPOBOU Û)$�ÛÛ o"SNFOJBOÛJTÛQFSIBQTÛCFTUÛLOPXOÛUPÛTDIPMBSTÛPGÛ*OEP�&VSPQFBOÛMBOHVBHFTÛBOEÛ$MBTTJDJTUTÛBTÛUIFÛQSJODJQBMÛ*OEP�&VSPQFBOÛMBOHVBHFÛPUIFSÛUIBOÛ(SFFLÛUPÛTIPXÛ�QSPUIFUJDÛWPXFMT�pÛ$MBDLTPOÛ��Û��ÛÛÛÛÛ8FÛEJTDVTTÛ IFSFÛ BMTPÛ ��Û)3��BOEÛ ��Û)8�ÛXIFSFÛ UIFTFÛ HSPVQTÛBSFÛGPMMPXFEÛCZÛBÛWPXFMÛTPÛUIBUÛUIFÛ3 �8ÛBSFÛDPOTPOBOUBM�ÛÛÛÛÛ, � � �, � � �, � �ÛBSFÛWPDBMJ[FEÛUPÛF� �B� �P��SFTQFDUJWFMZ�Û$G�Û,PSUMBOEUÛ��B�ÛÛÛÛ5IFSFÛJTÛOPÛQSPCMFNÛXJUIÛ, � ��B��ÛÛÛÛ�, � O S�ÛBZSÛ�NBO�ÛWJBÛ�BOJS��BZOS�Û(S�Û�NH�R Û4LU�ÛOºS�ÛFUD�ÛÛÛÛ�, � TUFS�M��ÛBTU ��TUBS��Û(S�Û�STH�R Û4LU�ÛTUºS� Û(PUI�ÛTUBJSOP�ÛÛÛ�, � SFV��ÛBSFXÛ�TVO��Û4LU�ÛSºWJ��ÛÛÛÛ�, � ��Û BTF OÛ �OFFEMF��Û (S�Û �KROJ�Û -BU�Û BDVTÛ �OFFEMF��Û 5IFÛ B�Û DBOÛ OPXÛ CFÛJEFOUJGJFEÛBTÛBÛ�QSPUIFUJDÛWPXFM�ÛBTÛGVMMÛHSBEFÛXPVMEÛIBWFÛHJWFOÛIB��ÛÛÛÛÛÛ'PSÛ, � ��F�ÛDG��ÛÛÛÛ�, � SFH PT�ÛFSFLÛ�FWFOJOH��Û(S�Û|REBOJ �4LU��SºKBT� Û(PUI�ÛSJRJT�ÛÛÛÛ�, � OFVO�ÛJOOÛ �OJOF��Û(S�ÛyNNE�A �4LU�ÛOºWB �-BU�ÛOPWFN�Û1*&Û�, � OFVOÛUPPLÛPWFSÛ UIFÛ [FSPÛHSBEFÛ GSPNÛ UIFÛPSEJOBMÛ �, � OVO�PT�Û DG�Û QM�Û JOVO�L �Û"TÛPOMZÛ(SFFLÛ BOEÛ "SNFOJBOÛ IBWFÛ BOÛ JOJUJBMÛ WPXFM Û UIFSFÛ JTÛ OPÛ CBTJTÛ GPSÛSFDPOTUSVDUJOHÛBÛGVMMÛHSBEFÛ�FOVO�ÛOPUFÛUIBUÛUIJTÛGPSNÛXPVMEÛIBWFÛ, � ��ÛÛÛ�, � NPT�ÛJN��NJOF��Û(S�ÛyMO�J �"W�ÛNB��Û$G�ÛBMTPÛ�, � NFÛ�NF� ÛJOFX��XJUIÛNF�Û�Û�JNF�CIJ Û(S�ÛyME� �-BU�ÛNFÛFUD�Û��ÛÛÛÛÛÛ�, � MFVEI��Û FMV[BOFNÛ �FYUSBDU��Û (S�Û yLEº EROJ � 4LU�Û SÌEIBUJ Û (PUI�Û MJVEBOÛ�HSPX��Û*GÛPOFÛEPVCUTÛUIJTÛMPOHÛFTUBCMJTIFEÛFUZNPMPHZ ÛUIFÛSPPUÛTUSVDUVSFÛJUTFMGÛJTÛTVGGJDJFOUÛUPÛQSPWFÛBÛMBSZOHFBM�ÛFMV[��)$7$��UIFÛMBSZOHFBMÛNVTUÛCF�, � � �BTÛUIFÛPUIFSÛMBSZOHFBMTÛXPVMEÛIBWFÛHJWFOÛB��ÛÛÛÛÛÛÛ8JUIÛF��P��CFGPSFÛPÛPSÛV�JOÛUIFÛGPMMPXJOHÛTZMMBCMF�ÛÛÛÛ�, � SFVH��Û PSDBNÛ �WPNJU�Û �Û �PSVD��Û (S�Û E�REU�GOMAI Û -JUI�Û SJºVH UJ � 0&ÛSPDFUUBO�Û4PNFÛTDIPMBSTÛBTTVNFÛBÛSFBMÛ"SNFOJBOÛQSPUIFUJDÛWPXFMÛCFGPSFÛS��Û 5IFZÛ TFFNÛ OPUÛ UPÛ IBWFÛ SFBMJ[FEÛ UIBUÛ UIJTÛ JTÛ JNQPTTJCMF�Û 1*&Û EJEÛ OPUÛIBWFÛ BÛXPSE�JOJUJBMÛ S��Û BQQBSFOUÛ JOTUBODFTÛIBEÛ BOÛ JOJUJBMÛ MBSZOHFBM�Û 5IJTÛMBSZOHFBMÛPSÛJUTÛSFGMFYÛJOÛ"SNFOJBOÛBOEÛ(SFFLÛXBTÛBMXBZTÛUIFSF�Û UIFSFÛXBTÛ OPÛ QFSJPEÛ XIFSFÛ BÛ SFBMÛ QSPUIFUJDÛ WPXFMÛ DPVMEÛ IBWFÛ CFFOÛ BEEFEÛCFGPSFÛUIFÛS�ÛBTÛUIFSFÛXBTÛOPÛJOJUJBMÛS��Û8FÛLOPXÛUIBUÛUIFÛWPXFMÛGSPNÛBOÛJOJUJBMÛMBSZOHFBMÛJTÛNVDIÛPMEFSÛUIBOÛUIFÛ"SNFOJBOÛQSPUIFTJT�Û"TÛJTÛTIPXOÛCZÛ BZS� �� , � O S � UIFÛ WPXFMÛ XBTÛ UIFSFÛ XIFOÛ UIFÛ TUSFTTÛ TIJGUFEÛ UPÛ UIFÛQFOVMUJNBUFÛTZMMBCMF ÛCVUÛFSJTÛBDD�ÛQM�Û�Û�USJOTÛTIPXTÛUIBUÛUIJTÛQSPUIFUJDÛWPXFMÛXBTÛOPUÛZFUÛUIFSFÛBUÛUIBUÛUJNFÛPUIFSXJTFÛXFÛXPVMEÛIBWFÛIBEÛ�FST�Û4PÛ JGÛ UIFSFÛ XBTÛ BOÛ JOJUJBMÛ MBSZOHFBM Û UIJTÛ IBEÛ BMSFBEZÛ CFDPNFÛ BÛ WPXFMÛCFGPSFÛ UIFÛ"SNFOJBOÛQSPUIFTJT�Û Û *GÛ POFÛEPVCUTÛ UIBUÛ FWFSZÛ S�Û JOÛ1*&ÛXBTÛQSFDFEFEÛCZÛBÛ MBSZOHFBM Û UIFOÛUIJTÛ JTÛ JOÛBOZÛDBTFÛUSVFÛ JOÛUIPTFÛ JOTUBODFTÛ

186� 5��6��3��%HHNHV��

XIFSFÛ(SFFLÛQPJOUTÛUPÛBÛWPXFMÛGSPNÛBOÛJOJUJBMÛMBSZOHFBM�Û4PÛBOÛ"SNFOJBOÛWPXFMÛCFGPSFÛBOÛS�ÛXIFSFÛPUIFSÛMBOHVBHFTÛIBWFÛS�ÛBOEÛ(SFFLÛIBTÛBÛWPXFMÛCFGPSFÛ UIJTÛ S�ÛNVTUÛ DPOUJOVFÛ BOÛ JOJUJBMÛ MBSZOHFBM�ÛsÛ *UÛ TIPVMEÛ GVSUIFSÛ CFÛOPUJDFEÛ UIBUÛ BMMÛ QPTTJCMFÛ JOTUBODFTÛ PGÛ, � �Û IBWFÛ F� Û PSÛ P�� F��Û 1SFTVNFEÛJOTUBODFTÛXJUIÛ B�� DBOÛ BOEÛNVTUÛ CFÛ SFDPOTUSVDUFEÛXJUIÛ, � ��Û BOVOÛ BOEÛBUBNOÛTFFÛCFMPX�ÛÛÛÛÛÛÛÛ'PSÛ, � ��P��OPUFÛUIBUÛVOTUSFTTFEÛP�ÛCFDBNFÛB�ÛJOÛ"SNFOJBOÛXJUIÛTPNFÛFYDFQUJPOT�ÛÛÛÛ�, � OFJE��ÛBO D�L �BOJDBOFN��DVSTF��Û(S�ÛŃEIDOJ Û4LU�ÛOJE� �(PUI�ÛHB�OBJUKBO�ÛÛÛ�, � CIFM��ÛBXFM��CSPPN��Û(S�Û´FELMAÛ�CSPPN��ÛÛÛÛ�, � CIFM��Û Z�BXFMVNÛ �BEEÛ UP��Û (S�Û ±FE�LLW� �JODSFBTF��Û 5IFSFÛ JTÛ EJTDVTTJPOÛXIFUIFSÛ UIFTFÛ UXPÛ SPPUTÛXFSFÛ JOÛ GBDUÛ POF Û BOEÛPOÛ UIFÛ SFMBUJPOÛXJUIÛ(S�ÛO�FEI�LWÛ�PXF��ÛÛÛÛ�, � EOU�N�BDD��ÛBUBNO��UPPUI�ÛQSPCBCMZÛWJBÛ�PEON��PENO��PUBNO�Û(S�Û±DW�N � ±DOU�J��"FPMJDÛ |DONTEJÛ DBOÛ FBTJMZÛ IBWFÛ JUTÛ F�Û GSPNÛ |DW� �FBU��Û PMEÛ P�WPDBMJTNÛ JTÛ QSPWFEÛ CZÛ NWDO�JÛ BOÛ BSDIBJDÛ GPSNÛ CFTJEFÛ �NO�DWNÛ BOEÛAI@MWDE�WÛ �IBWFÛ UIFÛ UFFUIÛ TFUÛ POÛ FEHF��Û 4LU�Û EºOU� � -BU�Û EFOT�� -JUI�Û ÓPEBT �-BUW�ÛVÍETÛ �HOBU�Û �Û�, � F�PE�ÛIBWFÛUIFÛTBNFÛSPPUÛ �CJUF� ÛBMTPÛGPVOEÛJOÛ(S�ÛO�DU�NH��QBJO�ÛGSPNÛXIJDIÛNW�DUNOJ�ÛBÛUPPUIÛEPFTÛOPUÛFBU ÛCVUÛCJUFT�ÛÛÛ�, � OF, � NO�Û BOVOÛ �OBNF�Û WJBÛ �PO NO� �Û �BOVNO�Û (S�Û ŃOMA��Û�, � O, � NO Û -BU�Û O NFOÛ FUD�Û 3FDPOTUSVDUJPOÛ PGÛ, � �Û SFTUTÛ POMZÛ POÛ UIFÛ(SFFLÛ OBNFÛ NUMAKRATI�DAJ ÛXIJDIÛ JTÛ JOTVGGJDJFOUÛ FWJEFODF�Û 5IFÛ GPSNTÛ(S�ÛNW�NUMNOJ��O�, � OVNOPT ÛZPVOHFSÛ�NW�NUMOJ ÛQSPWFÛUIFÛBOUJRVJUZÛPGÛUIFÛ , � ��Û "TÛ XFÛ OPXÛ LOPXÛ UIBUÛ , � �� SFTVMUFEÛ JOÛ F�Û JOÛ "SNFOJBO Û UIFÛDPSSFTQPOEFODFÛ"SN�ÛB�� (S�ÛO Û DBOÛCFÛVTFEÛ UPÛQSPWFÛ, � ��Û"UÛ UIFÛTBNFÛUJNF ÛUIJTÛDPSSFTQPOEFODFÛQSPWFTÛUIFÛQSFTFODFÛPGÛBOÛJOJUJBMÛMBSZOHFBM�ÛJUÛJTÛNPTUÛ JNQSPCBCMFÛ UIBUÛ SFTIVGGMJOHÛ PGÛ BOMBVUÛ WBSJBOUTÛ XPVMEÛ JOÛ CPUIÛMBOHVBHFTÛ IBWFÛ SFTVMUFEÛ JOÛ QSFDJTFMZÛ UIJTÛ DPSSFTQPOEFODF�Û 5IFÛSFDPOTUSVDUJPOÛPGÛUIFÛSFNBJOEFSÛPGÛUIFÛXPSEÛJTÛJSSFMFWBOUÛIFSF�ÛÛÛÛ "ÛQSPCMFNÛJTÛGPSNFEÛCZÛUIFÛXPSETÛXJUIÛBO��GSPNÛ�, � O� ÛGPSÛPOFÛXPVMEÛFYQFDUÛ�PO�Û�Û�VO��ÛBOVO �BOVS �BOJDBOFN��5IFÛTPMVUJPOÛJTÛOPUÛDMFBSÛZFU�Û1FSIBQTÛP��B�XBTÛFBSMJFSÛUIFOÛPO��VO�Û"OPUIFSÛQPTTJCJMJUZÛJTÛUIBUÛUIFÛP� �� , � Û IBEÛ OPUÛ ZFUÛ DPJODJEFEÛXJUIÛ UIFÛ FYJTUJOHÛ PÛ BUÛ UIFÛ UJNFÛXIFOÛ POÛCFDBNFÛVO�ÛÛÛ�, � O CI�P�Û �ÛBOJX��XIFFM��ÛUIFÛGPSNÛXPVMEÛCFÛBÛW EEIJ�GPSNBUJPOÛGSPNÛ�OBWF��ÛUIFÛFUZNPMPHZÛJTÛVODFSUBJO�Û*UÛDBOOPUÛCFÛTIPXOÛUIBUÛUIFÛWPXFMÛXBTÛP��PSJHJOBMMZ�Û��Û0O�, � 3$��P3$�ÛTFFÛUIFÛOFYUÛTFDUJPO Û��ÛÛÛÛÛÛ*GÛ UIFÛ GPMMPXJOHÛ UXPÛ GPSNTÛ IBWFÛ [FSPÛ HSBEFÛ CVUÛ , � P�Û DBOOPUÛ CFÛFYDMVEFE ÛXFÛGJOEÛUIFÛP�ÛQSFTFSWFE�ÛÛÛ�, � TU�VFS��ÛPTLSÛ�CPOF��Û4LU�ÛºTUIJ Û(S�Û±STzON�Û


ÛÛ�, � TEP��ÛPTUÛ�CSBODI��Û(S�Û´ZOJ Û(PUI�ÛBTUT��ÛÛÛÛÛÛÛ,PSUMBOEUÛUIJOLTÛUIBUÛVU Û�Û�PQU ÛNBZÛBMTPÛIBWFÛ[FSPÛHSBEF�ÛJUÛNVTUÛIBWFÛIBEÛP�ÛUPÛQSPEVDF�PX��V��Û�ÛÛÛÛÛÛÛÛÛ5IFÛ XPSEÛ GPSÛ �FZF� Û BLOÛ GSPNÛ UIFÛ SPPUÛ �, � FL � Û NVTUÛ BMTPÛ IBWFÛ [FSPÛHSBEFÛ CFDBVTFÛ JUÛ IBTÛ OPÛ I��Û 5IVTÛ JUÛ JTÛ BOPUIFSÛ FYBNQMFÛ PGÛ BÛ QSPUIFUJDÛWPXFM�Û�, � L ��ÛBL�OÛ�FZF�ÛUIFÛB�ÛXBTÛUBLFOÛGSPNÛUIFÛPCMJRVFÛDBTF ÛDG�ÛHFO�ÛBLBO�Û#VUÛ, � PL �ÛDBOOPUÛCFÛFYDMVEFE�Û�� *UÛ JTÛ HFOFSBMMZÛ BTTVNFEÛ UIBUÛ CFGPSFÛ X�Û BÛ MBSZOHFBMÛ EJEÛ OPUÛEFWFMPQÛJOUPÛBÛWPXFM�Û5IJTÛBTTVNQUJPOÛSFTUTÛFTQFDJBMMZÛPOÛHPNÛ�CF� ÛXIJDIÛXBTÛEFSJWFEÛGSPNÛ�, � VFT��)PXFWFS Û,PSUMBOEUÛ��CÛJTÛPQQPTFEÛUPÛUIJTÛFUZNPMPHZÛ POÛ UIFÛ HSPVOEÛ UIBUÛ JUÛ JTÛ OPUÛ DMFBSÛ XIZÛXFÛXPVMEÛ IBWFÛ UIFÛWPDBMJTNÛPGÛUIFÛQFSGFDUÛIFSF�Û5IFSFÛBSFÛUIFÛGPMMPXJOHÛPUIFSÛGPSNT�ÛÛÛÛ�, � VFH I��ÛHPHÛ�TQFBL��Û-BU�ÛWPWF �4LU�ÛW HIºU��Û5IFÛMBSZOHFBMÛEFQFOETÛPOÛUIFÛDPOOFDUJPOÛXJUIÛ(S�ÛEU��COMAI�Û�Û�, � FVH I� ÛXIJDIÛJTÛEJTQVUFE��ÛÛÛÛÛÛÛÛÛÛ�, � VFM)��ÛHF NOÛ �XPPM Û GMFFDF��Û -BU�ÛWFMMVT�Û)PXFWFS Û UIFÛ SFDPOTUSVDUJPOÛPGÛUIJTÛSPPUÛJTÛEJTQVUFE��ÛÛÛÛ�ÛÛÛ5IVT ÛUIFÛNBUFSJBMÛJTÛJODPODMVTJWF ÛCVUÛUIFÛPMEÛWJFXÛNBZÛCFÛDPSSFDU�Û*OÛ(SFFLÛ)V��7X��JTÛGSFRVFOU�ÛÛÛ 5IFÛ MBSZOHFBMTÛ SFNBJOÛ EJTUJODUÛ JOÛ UIJTÛ DBTF Û BTÛ JTÛ QSPCBCMZÛ UIFÛDBTFÛ JOÛ)3$Û BTÛXFÛ TIBMMÛ TFFÛ JOÛ UIFÛ OFYUÛ TFDUJPOÛXIJMFÛ FMTFXIFSFÛ UIFZÛDPJODJEFÛ JOÛ"SNFOJBO�Û 5IJTÛ IBTÛ BÛ QBSBMMFMÛ JOÛ -BUJO ÛXIFSFÛ �)/$�Û HJWFTÛ�F/$� � �B/$� � �P/$�� BDDPSEJOHÛ UPÛ UIFÛ MBSZOHFBM Û XIFSFBTÛ JOÛ BMMÛ PUIFSÛJOTUBODFTÛUIFÛMBSZOHFBMTÛIBWFÛGBMMFOÛUPHFUIFS��ÛÛÛÛÛ 'JOBMMZÛJUÛNBZÛCFÛNFOUJPOFEÛUIBUÛFYDFQUÛJOÛ(SFFLÛUIFÛMBSZOHFBMTÛTFFNÛ BMTPÛ UPÛ IBWFÛ BÛ WPDBMJDÛ SFGMFYÛ JOÛ 1ISZHJBOÛ BOEÛ JOÛ UIFÛ "OBUPMJBOÛMBOHVBHFTÛJOÛUIJTÛQPTJUJPO ÛXIJDIÛDBOÛIBSEMZÛCFÛBDDJEFOUBM�Û*UÛ JTÛBOÛBSFBMÛGFBUVSF�ÛÛ��Û ��ÛÛ8PSE�JOJUJBMÛMBSZOHFBMÛCFGPSFÛSFTPOBOU Û)3�ÛÛ 5IFÛ TFRVFODFTÛ XJUIÛ BÛ WPXFMÛ GPMMPXJOHÛ )37�Û BSFÛ EJTDVTTFEÛBCPWF Û JOÛ ��Û )FSFÛ XFÛ HJWFÛ UIFÛ TFRVFODFÛ CFGPSFÛ DPOTPOBOU � )3�$��,PSUMBOEUÛ��C Û��C Û��C�ÛÛÛÛÛÛÛÛÛ�, � 3$�ÛÛÛÛÛÛ5IFSFÛJTÛOPÛQSPCMFNÛXJUIÛ, � $��ÛUIJTÛHJWFTÛBS$��Û/PUFÛUIBUÛJUÛZJFMETÛOPÛI��ÛÛÛ�, � SU P��ÛBS Û�CFBS��Û4LU�Û L B� Û(S�Û�RKTOJ Û)JUU�ÛIBSUBHHB�ÛFUD��ÛTFFÛ��E�ÛÛÛ�, � JQJP��ÛBSDVJÛ�FBHMF��Û4LU�Û KJQZº��ÛÛÛ�, � CIJ��ÛBNC�PM Û�XIPMF��Û4LU�ÛBCIJ Û(S��MFI� Û-BU�ÛBNC� Û0)(ÛVNCJ�ÛÛÛÛÛÛ�, � UJ��ÛBSEÛ�KVTUÛOPX��Û(S�Û�RTI�Û-JUI�ÛBSUÅÛ�OFBS� ÛXJUIÛGVMMÛHSBEF�Û

188� 5��6��3��%HHNHV��

ÛÛ�, � UV��Û BSEÛ �PSEFS��Û 4LU�Û UÓ�Û �UIFÛ SJHIUÛ UJNF� Û (S�Û�RTU�WÛ �BSSBOHF� Û -BU�ÛBSUVTÛ�KPJOU ÛMJNC��ÛÛÛ�, � O ��ÛIBTBOFNÛ�BSSJWF��Û4LU�ÛB OÌUJ Û(S�ÛPOD HNEKH�JÛ�SFBDIJOHÛUPÛUIFÛGFFU��ÛÛÛ�, � IV��ÛBOKV�LÛ�OBSSPX��Û4LU�ÛBNIÓ� Û0$4ÛP[ L Û(PUI�ÛBHHXVT�Û5IFÛMBTUÛGPSNTÛBMMÛIBWFÛGVMMÛHSBEF ÛCVUÛUIFÛ"SNFOJBOÛGPSNÛNVTUÛIBWFÛ[FSPÛHSBEF ÛBTÛJUÛXPVMEÛPUIFSXJTFÛIBWFÛI��ÛBCMBVUÛXJUIÛV�TUFNTÛJTÛOPSNBM Û[FSPÛHSBEFÛJOÛV�TUFNÛBEKFDUJWFTÛJTÛGSFRVFOU�ÛÛÛ�, � ��Û BSDBU Û TJMWFS��Û "W�Û S [BUB� Û (S�Û A��RGUROJ Û -BU�Û BSHFOUVN�Û UIFSFÛ JTÛEPVCUÛBCPVUÛUIFÛFUZNPMPHZ ÛCFDBVTFÛPGÛUIFÛFOEJOHÛ�BU �ÛÛÛ�, � L��ÛBSHFMÛ�PCTUBDMF��Û(S�ÛA�RKE�WÛ�XBSEÛPGG� Û-BU�ÛBSDF Û�LFFQÛPGG��ÛÛÛÛÛÛ�, � 3$� ��, � 3$�ÛÛÛÛÛÛ8JUIÛUIFÛPUIFSÛUXPÛMBSZOHFBMTÛUIFÛRVFTUJPOÛJTÛXIFUIFSÛUIFZÛCPUIÛHJWFÛB� ÛPSÛF�ÛBOEÛP�ÛSFTQFDUJWFMZ�Û*ÛLOPXÛKVTUÛPOFÛJOTUBODFÛGPSÛXIJDIÛ, � $�ÛIBTÛCFFOÛ BTTVNFE�Û B LB LÛ �QPPS�Û DPNQBSFEÛ XJUIÛ -JUI�Û FMHFUºVUJÛ �CFH�Û I KIX BÛ�QPWFSUZ� Û0)(ÛJMLJÛ�IVOHFS��Û)PXFWFS ÛUIFÛ-JUIVBOJBOÛJOUPOBUJPOÛQPJOUTÛUPÛ�HI ÛBOEÛ'SBFOLFMÛHJWFTÛBTÛ JUTÛDPSSFDUÛNFBOJOHÛ �UPÛCFIBWFÛPOTFMG ÛCFH�ÛBOEÛ DPOOFDUTÛ UIFÛ XPSEÛ XJUIÛ BMH¹� �TBMBSZ�Û BOEÛ GVSUIFSÛ XJUIÛ 4LU�Û BSHIº�Û�WBMVF� Û (S�� A�LFH�Û �QSPEVDF Û HBJO��Û IFÛ EPFTÛ OPUÛ NFOUJPOÛ UIFÛ "SNFOJBOÛXPSE�Û "MMÛ UIJTÛ QPJOUTÛ JOÛ BÛ EJGGFSFOUÛ EJSFDUJPO�Û -JUI�Û ºMLUJÛ �CFÛ IVOHSZ�ÛSFRVJSFTÛ)7M)L� Û3VTT�Û MºLPNZKÛ �OBTDIIBGU�ÛQPJOUTÛ JOÛBOPUIFSÛEJSFDUJPO�Û JUÛJTÛ DPOOFDUFEÛ XJUIÛ 0)(Û JMHJ Û XIJDIÛ TUBOETÛ CFTJEFÛ JMLJÛ DJUFEÛ BCPWF�Û 5IFÛXIPMFÛHSPVQÛJTÛUPPÛVODMFBSÛGPSÛBOZÛDPODMVTJPOT�ÛÛÛÛÛ'PSÛ, � 3$�Û ,PSUMBOEUÛNFOUJPOTÛ TFWFSBMÛ DBOEJEBUFT�Û P C � PSC � PSKJL � BOEÛF VOHO�Û*ÛXJMMÛTIPSUMZÛEJTDVTTÛUIFN�ÛÛÛÛ�, � MCI��ÛP CÛ�MBNFOUBUJPO��Û(S�ÛO�LOF ROMAI Û-JUI�ÛVMCÓPUJ�Û5IFÛ(SFFLÛGPSN ÛGSPNÛ �, � MPCI� Û NBLFTÛ UIFÛ FYJTUFODFÛ PGÛ BOPUIFSÛ GVMMÛ HSBEFÛ XJUIÛ4DIXFCFBCMBVUÛ JNQSPCBCMF�Û TPÛ UIJTÛ DBTFÛ TFFNTÛ HPPE�Û 5IPVHIÛ JUÛJOEJDBUFTÛTPVOET ÛUIFÛXPSEÛJUTFMGÛJTÛOPUÛPOPNBUPQPFJD�ÛÛÛ�, � I��Û PSKJL Û �UFTUJDMFT��Û "W�Û S [J� Û (S�Û ´RCIJ Û "MC�� IFSEIFÛ �, � FS I� Û,PSUMBOEUÛ��B Û��Û<UIJTÛWPM� Û��> Û-JUI�ÛFS JMBTÛ �TUBMMJPO��ÛUIFÛ(SFFLÛXPSEÛDBOÛIBWFÛGVMMÛPSÛ[FSPÛHSBEF ÛUIFÛ"SNFOJBOÛXPSEÛDPVMEÛIBWFÛ�, � PS I��ÛÛÛ�, � SCI��Û PSCÛ �PSQIBO��Û (S�Û O�RFANO�J Û -BU�Û PSCVT Û (PUI�Û BSCJ Û )JUU�Û IBSQ[JÛ�TFQBSBUF��ÛÛÛ�, � H I��Û JOÛF VOHO� �OBJM��Û(S�ÛŃUXÛ�Û�, � OPH I� Û-JUI�ÛO¼HBT Û0)(ÛOBHBM�Û,PSUMBOEUÛ��B Û��Û<UIJTÛWPM� Û��>ÛBTTVNFTÛUIBUÛJOÛ"SNFOJBOÛ�POPH�ÛBOEÛ�POH�ÛXFSFÛVOJUFEÛJOÛ�POPOH� ÛXIJDIÛXBTÛEJTTJNJMBUFEÛUPÛ�FOPOH� ÛBOEÛUIJTÛUPÛ F VOH�Û UIFÛ �OÛ CFJOHÛ BÛ TFDPOEBSZÛ BEEJUJPO Û BTÛ JUÛ TPÛ PGUFOÛ JT�Û *UÛ JTÛDFSUBJOMZÛNPTUÛBUUSBDUJWFÛUPÛFYQMBJOÛ�VOH�ÛJOÛUIJTÛXBZ�ÛÛÛÛÛÛ5IVT ÛUIFSFÛBSFÛTFWFSBMÛQSPCBCMF ÛBOEÛPOFÛBQQBSFOUMZÛDFSUBJOÛJOTUBODFÛHJWJOHÛP3$��ÛÛÛÛÛÛ"TÛ UPÛ UIFÛ BQQBSFOUÛ BCTFODFÛ PGÛ F3$� Û JUÛ JTÛ QPTTJCMFÛ UIBUÛ , � �Û JOÛ UIJTÛ


TFRVFODFÛ EJEÛ OPUÛ JOGMVFODFÛ UIFÛ DPMPVSÛ PGÛ UIFÛ WPXFM Û, � Û CFJOHÛ UIFÛ OPO�DPMPVSJOHÛ MBSZOHFBM�Û )PXFWFS Û BTÛ XFÛ TBX Û OPÛ DFSUBJOÛ JOTUBODFÛ PGÛ B3$�ÛGSPNÛ, � 3$�ÛJTÛLOPXO�ÛÛÛÛÛÛÛ"TÛ UIFÛ P�DPMPVSJOHÛ JOGMVFODFÛ PGÛ , � �Û JOÛ UIJTÛ HSPVQÛ IBTÛ OPXÛ CFFOÛFTUBCMJTIFE Û UIFSFÛ JTÛOPÛMPOHFSÛSFBTPOÛ UPÛEPVCUÛUIBUÛ, � �Û JTÛSFGMFDUFEÛBTÛBÛQSPUIFUJDÛWPXFMÛP��ÛÛÛ ��ÛÛ8PSE�JOJUJBMÛMBSZOHFBMÛCFGPSFÛTFNJWPXFM Û)8��ÛÛ 5IJTÛ DPOGJHVSBUJPOÛ JTÛ EJGGJDVMUÛ UPÛ EFNPOTUSBUF�Û /PUFÛ UIBUÛ J� � V�ÛXFSFÛ MPTUÛ SFEVDFEÛ CFGPSFÛ UIFÛ TUSFTT�Û "Û HPPEÛ JOTUBODFÛ TFFNTÛ UIFÛGPMMPXJOH�ÛÛÛ�, � VT�O��ÛVO�LOÛ �FBS��Û-BU�ÛBVSJTÛ�, � FVT� Û(S�ÛOÂJÛ�, � PVT Û"W�ÛVu Û0$4ÛVYP�Û 5IFÛ , � �Û QFSIBQTÛ MJWFTÛ POÛ JOÛ QM�Û BL�BO L Û XIPTFÛ GVSUIFSÛ PSJHJOÛ JTÛVODMFBS�ÛÛÛÛÛÛ5IFSFÛ JTÛ OPÛ USBDFÛ PGÛI�Û GSPNÛ)J�V$� Û TPÛ UIFÛ MBSZOHFBMÛXBTÛ BQQBSFOUMZÛMPTUÛJOÛUIJTÛQPTJUJPO�Û�ÛÛ� �� ;SVH�JMREP�PEV]RKIEPÛÛ ��ÛÛ8PSE�GJOBMÛMBSZOHFBMÛBGUFSÛWPXFM Û�7)ÛÛ 5IJTÛ TFRVFODFÛ HBWFÛ BÛ MPOHÛ WPXFM Û XIJDIÛ EJTBQQFBSFEÛ JOÛ"SNFOJBO Û TPÛ JUÛ JTÛ EJGGJDVMUÛ UPÛ EFNPOTUSBUF�Û CVUÛ JOÛ NPOPTZMMBCMFTÛ UIFÛWPXFMÛJTÛNBJOUBJOFE�ÛÛÛ�NF, � �ÛNJÛ�UIBUÛOPU��Û(S�ÛMH� Û4LU�ÛN �ÛÛÛ ��ÛÛ8PSE�GJOBMÛMBSZOHFBMÛBGUFSÛÛDPOTPOBOU Û�$)�ÛÛ 0OFÛ TVQQPTFTÛ UIBUÛ UIFÛ MBSZOHFBMÛ CFDBNFÛ �B Û CVUÛ BTÛ JUÛ XBTÛ MPTU ÛUIJTÛJTÛEJGGJDVMUÛUPÛQSPWF�Û*UÛNBZÛCFÛBTTVNFEÛJOÛ�� PNU�, � Û�Û�TVOÛ�UFOT� ÛF�H�ÛFSFTVOÛ�UIJSUZ��ÛÛÛ�NF , � ��ÛNFDÛ�HSFBU��Û4LU�ÛNºI OU� Û(S�ÛME�GAJ�Û%PFTÛUIFÛMBZOHFBMÛMJWFÛPOÛJOÛUIFÛB�TUFN ÛNFDBD � Û)PXFWFS ÛXFÛDBOOPUÛBTDFSUBJOÛXIBUÛUIFÛFOEJOHÛPGÛUIFÛOPNJOBUJWFÛXBT�ÛNBZCFÛJUÛXBTÛ�NF �F�P, � �T�ÛÛÛÛÛ ��ÛÛ8PSE�GJOBMÛMBSZOHFBMÛBGUFSÛSFTPOBOU Û�3)�ÛÛÛÛÛÛ *GÛ BÛ WPXFMÛ QSFDFEFE Û UIFÛ EFWFMPQNFOUÛ XBTÛ UIFÛ TBNFÛ BTÛ JOÛ UIFÛQSFDFEJOHÛTFDUJPO�Û*GÛBÛDPOTPOBOUÛQSFDFEFE ÛUIFÛEFWFMPQNFOUÛJTÛVOLOPXOÛBTÛUIFSFÛJTÛOPÛFWJEFODF�ÛÛÛ ��ÛÛ8PSE�GJOBMÛMBSZOHFBMÛBGUFSÛTFNJWPXFM Û�8)ÛÛ *UÛIBTÛCFFOÛQSPQPTFEÛUIBUÛ�J) ��V)ÛCFDBNFÛ�ZB ��XB ÛBTÛJOÛ(SFFL�Û0GÛDPVSTF ÛUIFÛGJOBMÛTZMMBCMFÛEJTBQQFBSFE ÛXIJDIÛNBLFTÛJUÛWFSZÛEJGGJDVMUÛUPÛEFDJEFÛ UIFÛ NBUUFS�Û 5ISFFÛ UZQFTÛ PGÛ FWJEFODFÛ IBWFÛ CFFOÛ QSFTFOUFE�Û ��Û

190� 5��6��3��%HHNHV��

QMVSBMTÛXJUIÛUIFÛEVBMÛFOEJOHÛ�J, � �Û��ÛDPNQPVOEÛOVNFSBMT�ÛBOEÛ��ÛJTPMBUFEÛGPSNT�ÛÛÛÛÛÛ�ÛUIFÛQMVSBMTÛBSFÛUIFÛGPMMPXJOH�ÛÛÛDVOH�L Û�LOFFT�Û�Û� PO �J, � �ÛÛÛB �L Û�FZFT�Û�Û�, � L �J, � �ÛÛÛMBO �L Û�CSFBTUT�Û�Û�M H I�J, � �ÛÛÛBLBO �L Û�FBST�Û�Û�, � FVT�O�J, � �ÛÛÛ5IFÛGJSTUÛGPSNÛXPVMEÛIBWFÛHJWFOÛUIFÛTBNFÛSFTVMUÛJGÛ�J, � ÛIBEÛCFDPNFÛ� ÛBOEÛJTÛUIFSFGPSFÛJSSFMFWBOU�Û5IFÛGPMMPXJOHÛUXPÛGPSNTÛTIPXÛQBMBUBMJ[BUJPOÛPGÛ BÛ WFMBSÛ XIJDIÛ DPVMEÛ BMTPÛ IBWFÛ CFFOÛ DBVTFEÛ CZÛ �Û 5IFÛ MBTUÛ GPSN ÛIPXFWFS ÛSFRVJSFTÛ�ZBÛ UPÛHFUÛ� ��Û0OMZÛUIJTÛ � JTÛGPVOEÛJOÛBMMÛTUFNTÛPGÛUIFÛQMVSBMÛBOEÛJUÛJTÛEPVCUGVMÛUIBUÛJUÛPSJHJOBUFEÛPOMZÛJOÛUIFÛOPNJOBUJWF�Û5IBUÛJT ÛUIFÛSFDPOTUSVDUJPOÛHJWFOÛNBZÛCFÛXSPOH�Û5IFÛEFWFMPQNFOUÛPGÛ�VTÛUPÛLÛJTÛJNQSPCBCMF�ÛÛÛÛÛÛÛ�Û5IFÛOVNFSBMTÛJOÛEJTDVTTJPOÛBSFÛDG�Û,PSUMBOEUÛ��C�ÛÛÛBÛNF�UBTBOÛ�FMFWFO�Û�Û�NJB�UBTBO�ÛÛÛCÛGPSNTÛJOÛ�BTVO�ÛL B �BTVOÛ��ÛBOEÛFXU BO�BTVOÛÛ��ÛÛÛÛDÛFSFTVOÛ�UIJSUZ�Û�Û�FSJ�BTVO�ÛÛÛ5IFÛGPSNTÛTIPXÛSFMFWBOUÛEJGGFSFODFT�Û5IFÛGJSTUÛJTÛTVQQPTFEÛUPÛDPOUBJOÛBÛMBSZOHFBM Û �TNJ, � ��Û UIFÛ UXPÛ GPMMPXJOHÛ CÛ IBWFÛ �BTVO Û XIJDIÛ DPOUBJOTÛ�E PNUB�ÛUIFÛMBTUÛIBEÛCPUIÛBÛMBSZOHFBMÛUSJ, � ÛBOEÛ�E PNUB�ÛÛÛÛÛBÛ*UÛJTÛOPUÛFWJEFOUÛUIBUÛUIJTÛGPSNÛTIPVMEÛDPOUBJOÛUIFÛGFNJOJOFÛFOEJOH�Û*ÛQSPQPTFÛJUÛIBTÛ�TNJ)�P�ÛXJUIÛ�P��B�ÛNJÛJTÛJOGMFDUFEÛBTÛBOÛP�TUFN�ÛÛÛÛÛÛCÛ/PUFÛUIBUÛ�FXU BO�TVOÛXPVMEÛIBWFÛHJWFOÛ�FXU BTVO�Û3FJOUSPEVDUJPOÛPGÛ�BO�Û NJHIUÛ IBWFÛ HJWFOÛ UIFÛ BDUVBMÛ GPSN�Û ,PSUMBOEUÛ BSHVFTÛ UIBUÛ UIFÛQSFHMPUUBMJ[FEÛEÛ MPTUÛ JUTÛEFOUBMÛ FMFNFOUÛ UISPVHIÛEJTTJNJMBUJPOÛBOEÛ UIBUÛUIFÛTFRVFODFÛ� � �ÛHBWFÛ�BOB� ÛBOEÛ� � �Û�Û�BSB�ÛJOÛ��ÛU �Û�ÛL BS��ÛÛÛÛÛÛDÛ"TÛJUÛJTÛJNQSPCBCMFÛUIBUÛ�FSJ�TVOÛXBTÛDIBOHFEÛUPÛ�FSJBTVO ÛUIFÛ�B�ÛIFSFÛQSPCBCMZÛ BSPTFÛ QIPOFUJDBMMZ�Û *GÛ XFÛ BTTVNFÛ �USJ, � � Û JUÛ NBZÛ QPJOUÛ UPÛ BÛEFWFMPQNFOUÛJOÛBVTMBVUÛPGÛUIJTÛGPSNÛUPÛ�USJB ÛBTÛJOÛ(SFFLÛTRI�A�Û#VUÛPOFÛNJHIUÛ BMTPÛ BTTVNFÛ UIBUÛ UIFÛ SFHVMBSÛ EFWFMPQNFOUÛ XBTÛ UPÛ �USJ Û XJUIÛ UIFÛOFVUFSÛ QMVSBMÛ FOEJOHÛ �BÛ TFDPOEBSJMZÛ BEEFE�Û rÛ 5IFÛ SJTFÛ BOEÛ UIFÛEJTUSJCVUJPOÛ PGÛ UIFÛ �B� Û CFGPSFÛ �TVO Û NVTUÛ CFÛ FYQMBJOFEÛ BOEÛ UIFSFGPSFÛQPJOUTÛUPÛEFWFMPQNFOUÛGSPNÛ��E PNU�ÛÛÛÛÛÛÛ�Û TUFS � �TUFSJMF�Û IBTÛ CFFOÛ FYQMBJOFEÛ GSPNÛ GFN�Û �TUFS�J, � �Û 5IFÛ WBSJBOUÛSFBEJOHÛTUFSEÛJTÛJTPMBUFE�Û5IJTÛTFFNTÛBÛHPPEÛQPTTJCJMJUZ�Û#VU ÛBQBSUÛGSPNÛBÛTVGGJYÛ� P� ÛHFOFSBMJ[BUJPOÛPGÛUIFÛPCMJRVFÛTUFNÛ�TUFS�JF, � �ÛJTÛBÛQPTTJCJMJUZ�ÛÛÛÛÛ4VSWFZJOHÛ UIFÛ UISFFÛ UZQFTÛ PGÛ BSHVNFOUÛ POFÛ DPODMVEFTÛ UIBUÛ JUÛ JTÛJNQSPCBCMFÛUIBUÛUIFÛEFWFMPQNFOUÛPGÛ�J) ��V)ÛXBTÛUPÛ�ZB Û�XB�Û5IFÛGPSNTÛXJUIÛ��E PNUBÛIBWFÛ�BTVOÛEFSJWJOHÛGSPNÛUIJTÛGPSN�ÛÛÛ


��0EV]RKIEP�MR�MRPEYX��Û Û��ÛÛ-BSZOHFBMÛJOÛJOMBVU ÛBGUFSÛWPXFM Û�7)��ÛÛÛÛÛÛ ��ÛÛ-BSZOHFBMÛJOÛJOMBVU ÛBGUFSÛWPXFM ÛCFGPSFÛWPXFM Û�7)7��ÛÛÛÛÛÛ 5IFÛ MBSZOHFBMÛ EJTBQQFBSFEÛ BGUFSÛ DPMPVSJOHÛ BOÛ BEKBDFOUÛ �F� ÛUIPVHIÛJUÛJTÛEJGGJDVMUÛUPÛGJOEÛDFSUBJOÛFWJEFODF�ÛÛÛÛÛÛÛ ��Û Û -BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ WPXFM Û CFGPSFÛ DPOTPOBOU ��7)$�ÛÛ 8FÛ GJOEÛ UIFÛ XFMM�LOPXOÛ USFBUNFOUÛ F, � � F, � � F, � � �� ÛXIJDIÛCFDBNF�J �B �V�JOÛ"SNFOJBO�ÛÛÛ�NF, � U S�ÛNBZSÛ�NPUIFS��Û4LU�ÛN UºS �(S�ÛM THR �-BU�ÛN UFS�ÛÛÛ�CIF, � �OJ��ÛCBO��XPSE��Û0*D�ÛCÌO �0&ÛC O��(S�ÛFWNH�Û�Û�CIP, � OF, � �ÛÛÛÛ�H SF, � �O��ÛFSLBO� �NJMMTUPOF��Û0$4Û S OZ Û0*D�ÛLXFSO�ÛEJGGFSFOUMZÛ GPSNFEÛ4LU�ÛHS WBO��ÛÛÛÛÛÛÛÛ�OF, � CI��ÛOBXU JÛ�GBTUJOH��Û(S�ÛN FW��ÛÛÛÛÛ�QSF, � �U��ÛFSBTU�BO�L Û �BSTF��Û(S�ÛPRWKTO�J��QSP, � �U��5IFÛGPSNÛJTÛNVDIÛEFCBUFE�Û *OÛ UIFÛ XBZÛ JOEJDBUFEÛ CPUIÛ GPSNTÛ DBOÛ CFÛ FYQMBJOFEÛ XJUIPVUÛEJGGJDVMUZ�Û 'PSÛ UIFÛ BCMBVUÛ DG�Û OPN�Û �OPL �U�T Û HFO�Û OFL �U�T�Û .ZÛ FBSMJFSÛQSPQPTBMÛ �QFS) �UÛ IBTÛ UIFÛ EJTBEWBOUBHFÛ PGÛ BÛ SPPUÛ FOEJOHÛ JOÛ UISFFÛDPOTPOBOUT ÛXIJDIÛJTÛWFSZÛSBSF��ÛÛÛÛ�QF, � �USP��Û IBXSBO� �IFSE��Û "W�Û Q SB�WBOU�Û �HJWJOHÛ QSPUFDUJPO��Û DG�Û -BU�ÛQ CVMVN��QBTUVSF��Û(S�ÛPW>UÛ�IFSE�Û�Û�QP, � �JV��ÛÛÛÛÛ�, � F, � �NS�ÛBXSÛ �EBZ� ÛWJBÛ� N S��BVS�Û(S�Û£MAR�Û'PSÛUIFÛMBSZOHFBMTÛTFFÛ��ÛÛÛÛÛÛ�QMF, � �UPT�ÛMJÛ�GVMM��Û4LU�ÛQS Uº� �-BU�ÛDPNQM UVT��ÛÛÛÛÛ�NF, � EPT�ÛNJUÛ �NJOE Û JOUFMMJHFODF��Û (S�Û MH>DOJ� �QMBO��Û #VUÛ �N E�Û JTÛ BMTPÛQPTTJCMF�ÛÛÛÛÛÛÛÛÛ�CISF, � �VS�ÛB CJXS��TQSJOH��Û(S�ÛFRE� R��QIS XBS��ÛÛÛÛÛ�, � MF, � �VS�ÛBMJXS��GMPVS��Û(S�Û�LEURON��ÛÛÛÛÛ�EP, � SPN�ÛUVSÛ�HJGU��Û(S��DW>RON Û0$4ÛEBS ��ÛÛÛÛÛ�, � OF, � �NO�ÛBOVO� �OBNF�ÛWJBÛ�PO NBO��4LU�ÛO NB �-BU�ÛO NFO�Û(S�ÛŃOMA��, � O, � �NO��5IFÛ FUZNPMPHZÛ PGÛ FSLBSÛ �MPOH�Û GSPNÛ �EVF, � SP�� NVTUÛ CFÛ SFKFDUFEÛ,PSUMBOEUÛ��ÛÛÛÛÛÛ ��ÛÛ-BSZOHFBMÛJOÛJOMBVU ÛBGUFSÛWPXFM ÛCFGPSFÛSFTPOBOU Û�7)3ÛÛ #FGPSFÛWPXFMÛ UIFÛEFWFMPQNFOUÛNVTUÛ CFÛ UPÛ 37�Û *GÛ BÛ DPOTPOBOUÛGPMMPXT Û JUÛ EFQFOETÛ POÛ XIFUIFSÛ UIFÛ SFTPOBOUÛ CFDPNFTÛ WPDBMJDÛ BTÛ JOÛ4BOTLSJU ÛW UB�Û�Û �WBBUB�Û�Û�, � VF, � OU�ÛPSÛOPUÛ BTÛ JOÛ(PUI�ÛXJOET�Û"SN�ÛBNJTÛ�NPOUI�Û�Û��N OT�Û�Û��NF)OTÛQPJOUTÛUPÛUIFÛMBUUFSÛEFWFMPQNFOU�ÛÛ

192� 5��6��3��%HHNHV��

Û ��Û Û -BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ WPXFMÛ CFGPSFÛ TFNJWPXFM Û�7)8�ÛÛÛÛÛ"ÛHPPEÛFYBNQMFÛJT�ÛÛÛ�EF, � JV S�ÛUBZHSÛ�CSPUIFS�JO�MBX�ÛWJBÛ�EBJX S�Û(S�ÛD H�R Û-JUI�ÛEJFWFSÅT Û0$4ÛE WFS �ÛÛÛ ��0EV]RKIEP�MR�MRPEYX��EJXIV�GSRWSRERX�ÛÛÛÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ DPOTPOBOU Û CFGPSFÛ WPXFM Û�$)7��ÛÛ 4FFÛ��ÛPOÛ TUPQÛ�Û MBSZOHFBM�Û *GÛ UIFÛ DPOTPOBOUÛ JTÛ OPUÛ BÛ TUPQ Û UIFÛMBSZOHFBMÛEJTBQQFBSTÛXJUIPVUÛUSBDF ÛFYDFQUÛGPSÛUIFÛQPTTJCMFÛDPMPVSJOHÛPGÛBÛGPMMPXJOHÛ�F��ÛÛÛ� FO, � PT�ÛDJOÛ�CJSUI��Û(S�ÛGzNOJ Û4LU�ÛKºOBT Û-BU�ÛHFOVT�ÛÛÛ� FS, � ��ÛDFSÛ�PMEÛNBO��Û4LU�ÛKºSBOU� Û(S�ÛGzRWN�ÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ DPOTPOBOUÛ CFGPSFÛ DPOTPOBOU Û�$)$��ÛÛ 5IFÛMBSZOHFBMÛJTÛWPDBMJ[FEÛUPÛ�B�ÛJOÛUIFÛGJSTUÛTZMMBCMFÛBOEÛCFGPSFÛBÛDMVTUFS ÛBOEÛEJTBQQFBSTÛJOÛPUIFSÛDBTFT�Û5IFÛNBUFSJBMÛJTÛMJNJUFE�ÛÛÛÛ�Q, � U S�ÛIBZSÛ�GBUIFS��Û(S�ÛPATH�R �4LU�ÛQJUºS� Û-BU�ÛQBUFS�ÛÛÛ�CI, � UJT�ÛCBZÛ�XPSE��Û(S�ÛF�TIJ��ÛÛÛÛÛÛÛ�E, � ��ÛUBN��*ÛHJWF��Û(S�ÛDO TO�J��HJWFO� Û-BU�ÛEB�UVT�JE�Û1SFTFOUÛCFTJEFÛBPSJTUÛFUV��EF, � ��ÛÛÛÛÛ�QU, � L��ÛU BL JNÛ�CFÛIJEEFO��Û(S�ÛPTH�SSW �BPS�Û PTAKW�N�ÛÛÛÛ�E, � Q�OJ��Û UBXO� �GFBTU��Û 0*D�Û UBGO�Û -BU�Û EBQTÛ �PGGFSJOHÛ NFBM� Û 5PDI�Û U Q��FBU��ÛDG�Û(S�ÛDAPA�NH��DPTUT� Û-BU�ÛEBNOVNÛJE�ÛÛÛ�T, � E��ÛBU�PL Û�GVMM ÛGBU��Û(S�Û�DRO�J��Û� FO, � �UMPT�ÛDOBX ��QBSFOU��ÛDG�Û4LU�ÛKBOÆUSBNÛ�CJSUIQMBDF��ÛÛÛ�, � FS, � �V��ÛIBSBXVOL ��GJFME��Û(S�Û�ROURA�Û4FFÛCFMPXÛPOÛUIJTÛGPSN��ÛÛÛÛÛ)FSFÛBMTPÛTB Û�TBMU� ÛJGÛJUÛEFSJWFTÛGSPNÛ�T, � M��ÛÛÛ��0OÛFSBTUBOL ÛTFFÛ��ÛÛÛÛÛÛÛÛÛÛÛ8JUIÛ [FSPÛ GSPNÛ MBSZOHFBMÛ JOÛ OPO�JOJUJBMÛ TZMMBCMFÛ BOEÛ CFGPSFÛ TJOHMFÛDPOTPOBOUÛXFÛGJOE�ÛÛÛÛ�EIV , � �U S�ÛEVTUSÛ�EBVHIUFS��Û(S�Û UGA�THR Û4LU�ÛEVIJUºS��ÛÛÛÛÛ��P�N, � OPT�Û�VOÛQBSUJDJQJBMÛBEKFDUJWF Û�EPJOHÛ��Û(S�Û O MENOJ�ÛÛÛ�JFO, � U S�Û �<J>OEJSÛ �TJTUFS�JO�MBX�Û CFMPOHTÛ IFSFÛ BDDPSEJOHÛ UPÛ ,PSUMBOEUÛ��Û CVUÛ *Û GJOEÛ UIFÛ SFPSHBOJ[BUJPOTÛ UPÛ BSSJWFÛ BUÛ UIFÛ BDUVBMÛ GPSNÛ O SÛDPNQMJDBUFE�ÛÛÛÛÛÛ*OÛGJSTUÛTZMMBCMF ÛUIFO ÛUIFÛMBSZOHFBMÛXBTÛWPDBMJ[FE ÛKVTUÛBTÛJOÛ(FSNBOJD�ÛÛÛÛÛÛ*OÛ UIFÛTFDPOEÛTZMMBCMFÛ JUÛ MPPLTÛBTÛ JGÛ UIFÛ MBSZOHFBMÛXBTÛPOMZÛWPDBMJ[FEÛCFGPSFÛ UXPÛ DPOTPOBOUT�Û Û )BSBXVOL Û TFFNTÛ UPÛ CFÛ BOÛ FYDFQUJPO�Û UIFÛ


TFRVFODFÛ�BSB�ÛNVTUÛDPNFÛGSPNÛUIFÛPCMJRVFÛDBTFT ÛXIJDIÛIBEÛ�, � , � �ÛTFFÛ��Û(F NO� �XPPM Û GMFFDF�Û JTÛ FRVBUFEÛXJUIÛ -BU�ÛWFMMVT Û CVUÛ UIJTÛXPSEÛIBEÛOPÛMBSZOHFBM�ÛÛÛÛÛ5IFSFÛ BSFÛ TPNFÛXPSETÛ UIBUÛ IBWFÛ JOJUJBMÛ)3)�$��5IFTFÛXPSETÛ DBOÛ CFÛUSFBUFEÛ VOEFSÛ3)$� ��Û PSÛ IFSF Û JGÛ3Û JTÛ DPOTJEFSFEÛ BÛ DPOTPOBOU�Û *ÛQSFGFSÛ UPÛ USFBUÛ UIFNÛ IFSFÛ CFDBVTFÛ UIFZÛ BHBJOÛ TIPXÛ UIBUÛ UIFÛ TFDPOEÛMBSZOHFBMÛJTÛWPDBMJ[FEÛCFGPSFÛBÛEPVCMFÛDPOTPOBOU�ÛÛÛÛ�, � S, � �USPN�ÛBSBXSÛ�QMPVHI��Û(S�Û�ROTRON��ÛÛÛÛ�, � S, � �USJ��ÛB BXSJÛÛ�NJMM��Û(S�Û�LETR I�J��XPNBOÛXIPÛHSJOETÛDPSO��ÛÛÛÛ�, � S)N��ÛBSNVLO��FMCPX��Û4LU�Û SNº��BSN� Û-BU�ÛBSNVT �(PUI�ÛBSNT�ÛÛÛ�, � O)U��Û ES�BOEÛ �EPPS�QPTU��Û 4LU�Û U � �-BU�Û BOUBF�Û /PUFÛ UIBUÛ UIJTÛ GPSNÛNBZÛ DPOUBJOÛ �IBOE Û XJUIÛ MPTTÛ PGÛ I�Û JOÛ DPNQPVOET�Û JUÛ XPVMEÛ SFRVJSFÛ�, � FO)U��ÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ DPOTPOBOU Û CFGPSFÛ SFTPOBOU Û�$)3�ÛÛ /PÛ FWJEFODFÛ LOPXO�Û 5IFÛ EFWFMPQNFOUÛ XJMMÛ EFQFOEÛ POÛ XIBUÛGPMMPXT ÛBÛDPOTPOBOUÛPSÛBÛWPXFM�ÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ DPOTPOBOUÛ CFGPSFÛ TFNJWPXFM Û�$)8�ÛÛ /PÛFWJEFODFÛLOPXO�ÛÛÛ ��0EV]RKIEP�MR�MRPEYX��EJXIV�VIWSRERX��6,��ÛÛÛÛÛ ��ÛÛ-BSZOHFBMÛJOÛJOMBVU ÛBGUFSÛSFTPOBOUÛCFGPSFÛWPXFM Û�3)7��$� )7��ÛÛÛÛÛÛÛ"GUFSÛBÛWPXFMÛUIFÛ MBSZOHFBMÛCFIBWFTÛBTÛBÛ MBSZOHFBMÛBGUFSÛBÛDPOTPOBOU�ÛTFFÛ ��Û 4QFDJBMÛ EFWFMPQNFOUTÛ PDDVSÛ BGUFSÛ BÛ DPOTPOBOU Û XIFOÛ UIFÛSFTPOBOUÛCFDBNFÛTZMMBCJD Û$� )7��5IJTÛDBUFHPSZÛXJMMÛCFÛEJTDVTTFEÛIFSF�ÛÛÛÛ5IFÛ TZMMBCJDÛ SFTPOBOUÛ CFDPNFTÛ B3Û BTÛ VTVBM Û UIFÛ MBSZOHFBMÛEJTBQQFBSTÛBTÛVTVBMÛCFGPSFÛWPXFM�ÛÛÛÛÛ�TN)�S�Û BNÛ �ZFBS� Û BNB OÛ �Û �TN, � FS�N Û DG�Û KNF OÛ �XJOUFS�Û �Û � IJNFS�N��TVNNFS��Û4LU�ÛTºN Û0)(ÛTVNBS�ÛÛÛÛÛ�TN, � ��Û BN O � BNFO�FL �JO� QM�Û �BMM� Û BNFO�B�� PGÛ BMM��Û 4LU�Û TBNB� � (S�Û�M² EN��GSPNÛTPNFXIFSF� ÛOU�D AM O� ��OPÛPOF� Û(PUI�ÛTVNTÛ�TPNF��ÛÛÛÛÛÛ�VS, � �FO��ÛHB OÛ �MBNC��Û(S�ÛA�RH�N Û4LU�ÛVSºO��Û5IFÛ�OÛ JTÛTFDPOEBSZ�Û/PUFÛUIBUÛ UIFÛ [FSPÛ HSBEFÛ PGÛ UIFÛ TVGGJY Û �O� � XPVMEÛ IBWFÛ HJWFOÛ �HBSBOÛ TFFÛ�� XIJDIÛJTÛBMTPÛBÛSFHVMBSÛGPSNÛPGÛUIFÛQBSBEJHN��ÛÛÛÛÛ� M, � �PT�� T�Û DB S� �MBVHI� Û JGÛ UIFÛ [FSPÛ HSBEFÛ SPPUÛ GPSNÛXBTÛ HFOFSBMJ[FE�ÛBMUFSOBUJWFMZÛ� M, � �T��DBMB� �PGÛXIJDIÛUIFÛ�B��XBTÛDPOUSBDUFEÛXJUIÛPSÛMPTUÛ CFGPSFÛ GPMMPXJOHÛ WPXFMÛ ,PSUMBOEUÛ ��B�Û TFFÛ ��Û (S�Û GE�LWJ�

194� 5��6��3��%HHNHV��

�Û� FM, � �� T �GELASTO�JÛ�MBVHIBCMF��ÛÛÛÛ� S, � ��Û TBS � �PZ� � �UPQ Û TVNNJU Û QFBL��Û 4LU�Û ÆSBT� �(S�Û KA�RH � KE�RAJ � -BU�ÛDFSFCSVN��LFSBT�SP��Û5IFÛGPSNÛNVTUÛIBWFÛBMUFSOBUFEÛXJUIÛUIFÛGVMMÛHSBEF ÛPUIFSXJTFÛUIFÛ� �XPVMEÛIBWFÛCFFOÛEFQBMBUBMJ[FEÛCFGPSFÛS��ÛÛÛÛ� S)��ÛTB OÛ�JDF��-JUI�ÛuºSNBT��IPBSGSPTU� Û3VTT�ÛTFSFO ��4BNFÛSFNBSLÛBTÛUIFÛQSFDFEJOHÛXPSEÛBTÛUPÛUIFÛQBMBUBM��ÛÛÛÛ�QO)��ÛIBOVN��XFBWF��Û-JUI�ÛQÅOUJ��UXJTU� ÛTQ TUJ��TFUÛUSBQT� Û(PUI�ÛTQJOOBO�ÛÛÛ�VM)��ÛHB U��TFDSFUMZ��Û-JUI�ÛQSB�WÅMUJ��EFDFJWF��ÛÛÛÛÛ�VS)��Û WB FN� �LJOEMF�Û JGÛ DPHOBUFÛ XJUIÛ -JUI�Û WÅSUJ Û 0$4Û WBSJUJ� �DPPL��ÛQSPCMFNÛJTÛUIFÛW�ÛPGÛ"SNFOJBOÛTFFÛ��D�ÛÛÛÛ� M, � �PV��ÛUBM��IVTCBOE�TÛTJTUFS�ÛXJUIÛU�ÛGSPNÛUBZHSÛ�IVTCBOE�TÛCSPUIFS��Û(S�ÛGA�LWJ Û0$4Û[ M WB�ÛÛÛÛÛ.VDIÛ EJTDVTTFEÛ JTÛ UIFÛ DPOOFDUJPOÛ CFUXFFOÛ LB JO� �BDPSO�Û BOEÛ (S�ÛBA�LANOJ��5IFÛ(SFFLÛXPSEÛ SFRVJSFTÛ �H M, � �FO� �XIJDIÛXPVMEÛ HJWFÛ"SN�Û�LBMBO� Û�FOP��BGUFSÛ, � ÛCFDPNJOHÛ�BOP��Û5IFÛFOEJOHÛ�JOÛNVTUÛCFÛTFDPOEBSZ��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ SFTPOBOU � CFGPSFÛ DPOTPOBOU Û�3)$�ÛÛÛÛÛ 5IJTÛ DPOGJHVSBUJPOÛ JTÛ POMZÛ JOUFSFTUJOHÛ BGUFSÛ DPOTPOBOU ÛXIFSFÛ3ÛDBOÛCFÛTZMMBCJD�Û5IFÛ"SNFOJBOÛSFGMFYÛJTÛ$�B3B�$Û,PSUMBOEUÛ��ÛÛÛ� O, � �TL�Z��Û BOB FNÛ �LOPX�Û BTTJNJMBUFEÛ GSPNÛ �DBOB FN�Û (S�ÛGI GNW�SKW��ÛÛÛÛÛ�TM, � �TL�Z��ÛB B FN��BTL��Û(S�Û L SKOMAI��TJ�TM, � �TL��ÛÛÛ� O, � �UPT�ÛDBOBXU Û �LOPXO�ÛGSPNÛ�DBOBXÛXJUIÛUIFÛTVGGJYÛ�UP��*OÛUIFÛTBNFÛXBZÛB BXU L Û�QSBZFS��ÛÛÛÛÛÛ�L UXS�) PNU�� GSPNÛ rE PNU�Û XIFSFÛ �E�Û �Û ��E��Û L B BTVOÛ �GPSUZ��Û (S�ÛTETRW�KONTA �-BU�ÛRVBES HJOUB�Û$G�ÛUIFÛEJTDVTTJPOÛJOÛ��ÛÛÛÛÛ�EIM, � �SP��ÛEBMBS��HSFFO ÛGSFTI��Û(S�Û ALERO�JÛ�CMPPNJOH ÛGSFTI ÛBCVOEBOU�Û�Û�EIM, � �FSPT�� 5IJTÛ FUZNPMPHZÛ IBTÛ CFFOÛ NVDIÛ EJTDVTTFE Û XJUIPVUÛDPOWJODJOHÛSFTVMU�Û5PÛNZÛNJOEÛJUÛDBOÛCFÛVOEFSTUPPEÛJOÛUIFÛXBZÛJOEJDBUFEÛIFSF�Û�ÛÛÛÛ� M, � �T��ÛDB S��MBVHI�ÛWJBÛ�DBMB� ÛPGÛXIJDIÛUIFÛ�B�ÛDPOUSBDUFEÛXJUIÛPSÛXBTÛMPTUÛCFGPSFÛBÛGPMMPXJOHÛWPXFMÛ,PSUMBOEUÛ��B�Û1FSIBQTÛUIFÛ[FSPÛHSBEFÛSPPUÛXBTÛHFOFSBMJ[FE�ÛTFFÛ��5IJTÛEFWFMPQNFOUÛJTÛNVDIÛEFCBUFE�Û5PÛNZÛNJOE Û BOB FN�DBOÛIBSEMZÛCFÛEPVCUFE�Û"ÛTVGGJYÛ�BL ��JTÛNPTUÛJNQSPCBCMF ÛBTÛ1*&ÛEJEÛOPUÛIBWFÛTVDIÛBÛTVGGJYÛJUÛXPVMEÛIBWFÛUPÛCFÛ�, � FL ��ÛUIFÛPOMZÛFWJEFODFÛJTÛGSPNÛ(SFFLÛA�LA�SSW ÛXIFSFÛJUÛEFSJWFTÛGSPNÛOPO�*&ÛOPVOT�Û5IFÛ�X�ÛJOÛDBOBXU ÛFUD�ÛJTÛ DPOWJODJOHMZÛFYQMBJOFEÛCZÛ,PSUMBOEUÛ JOÛ UIFÛXBZÛ JOEJDBUFEÛBCPWF�Û *OÛFXU BOBTVOÛ UIFÛ �O�Û NVTUÛ IBWFÛ CFFOÛ UBLFOÛ GSPNÛ UIFÛ DBSEJOBMÛ FXU O�Û 'PSÛEBMBSÛOPÛFYQMBOBUJPOÛIBEÛCFFOÛHJWFOÛVQÛUPÛOPX�ÛJGÛUIFÛPOFÛHJWFOÛIFSFÛJTÛ


DPSSFDU ÛJUÛDPOGJSNTÛ�B3B��BTÛUIFÛSFHVMBSÛSFGMFY�ÛÛÛÛÛ0UIFSÛFWJEFODFÛ JTÛVOSFMJBCMF�Û0UIFSÛQSFTVNFEÛEFWFMPQNFOUTÛNVTUÛCFÛFYQMBJOFEÛEJGGFSFOUMZ�ÛFSBTUBOL ÛFSLBOÛBOEÛOBXU JÛIBWFÛGVMMÛHSBEFÛ�F, � �ÛTFFÛ��Û DOBX �HPFTÛCBDLÛPOÛ � FO, � UMPTÛ TFFÛ��Û DOBXU� �KBX�Û DBOOPUÛCFÛ DPOOFDUFEÛXJUIÛ(S�ÛGN� OJ� �KBX� Û CFDBVTFÛ UIFÛ(SFFLÛXPSEÛDBOOPUÛ CFÛSFEVDFEÛ UPÛ BÛ 1*&Û GPSN�Û $B S � HB O� BOEÛ LB JO� IBWFÛ QSFWPDBMJDÛ �3)��Û TFFÛ��Û0OÛBSNVLOÛBOEÛESBOEÛTFFÛ��ÛÛ 4PÛUIFÛPOMZÛSFGMFYÛUIBUÛDBOÛCFÛFTUBCMJTIFEÛJTÛ�B3B��Û5IJTÛGJUTÛJOÛWFSZÛXFMMÛ XJUIÛ XIBUÛ XFÛ LOPXÛ PGÛ "SNFOJBO�Û 4ZMMBCJDÛ 3Û CFDPNFTÛ B3Û JOÛ"SNFOJBO Û BOEÛ UIFOÛ UIFÛ MBSZOHFBMÛ JTÛ WPDBMJ[FEÛ UPÛ B�Û Û "Û DPNQMJDBUFEÛJOTUBODFÛTFFNTÛUPÛCFÛUIFÛGPMMPXJOH�ÛÛÛÛ�, � FS, � �VS�Û IBSBXVOL Û �GJFME��Û (S�Û �ROURA Û -BU�� BSWVN Û 0*S�Û BSCPS�Û 5IFÛXPSEÛNVTUÛIBWFÛJUTÛI�ÛGSPNÛUIFÛOPNJOBUJWF�Û*GÛJUÛJTÛDPSSFDUÛUIBUÛBÛMBSZOHFBMÛJOÛ JOMBVUÛ JTÛ OPUÛ WPDBMJ[FEÛCFGPSFÛ BÛ TJOHMFÛ DPOTPOBOU Û UIFÛ TFDPOEÛ BÛNVTUÛQSPCBCMZÛ CFÛ FYQMBJOFEÛ GSPNÛ UIFÛPCMJRVFÛ DBTFT ÛXIJDIÛIBEÛ �, � S, � �VFO�TÛDG�Û0*S�ÛBSCBF ÛXIFSFÛUIFÛ[FSPÛHSBEFÛHBWFÛBSB�� 0OÛ)3)�$�TFFÛTFDUJPOÛ��ÛÛÛÛÛÛÛÛÛÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ SFTPOBOUÛ CFGPSFÛ SFTPOBOU Û�3)3ÛÛ /PÛFWJEFODFÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ SFTPOBOUÛ CFGPSFÛ TFNJWPXFM Û�3)8�ÛÛ /PÛFWJEFODFÛÛÛ ��0EV]RKIEP�MR�MRPEYX��EJXIV�WIQMZS[IP��;,�ÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ TFNJWPXFMÛ CFGPSFÛ WPXFM Û�8)7�ÛÛ /PÛFWJEFODF�ÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ TFNJWPXFMÛ CFGPSFÛ DPOTPOBOU Û�8)$�ÛÛ 5IFÛ TFRVFODFTÛ �J)� � �V)�Û SFTVMUFEÛ JOÛ � � � � �� J � V�Û "UUFNQUTÛ UPÛTIPXÛEFWFMPQNFOUTÛMJLFÛZB �XBÛIBWFÛGBJMFE�Û"ÛTQFDJBMÛEFWFMPQNFOU Û�V)��VL� ÛJTÛQPTJUFEÛCFGPSFÛGJOBMÛOBTBMÛCZÛ,PSUMBOEUÛ��C�ÛÛÛ�H IJ)T�MPN�Û JMÛ�TJOFX��Û-JUI�ÛHÖTMBÛ�WFJO� Û-BU�ÛG MVNÛ�UISFBE��ÛÛÛ�ULJ, � �OP��ÛD JOÛ�IBXL��Û(S�ÛI�KTkNOJ�Û*GÛUIFÛGPSNÛNVTUÛCFÛSFDPOTUSVDUFEÛBTÛ�ULJ)�JOP� Û BTÛ IBTÛ CFFOÛ QSPQPTFE Û UIFOÛ JUÛ JTÛ OPUÛ SFMFWBOUÛ IFSFÛ CVUÛ JOÛ��ÛÛÛ�QV)S��ÛIVSÛ�GJSF��Û(S�ÛPU>R Û)JUU�ÛQBIIVS�Û

196� 5��6��3��%HHNHV��

ÛÛ�V)O��ÛVO�BZOÛ �FNQUZ�Û JGÛ UIJTÛCFMPOHTÛUPÛ4LU�Û Oº�Û �EFGJDJFOU� Û-BU�ÛW OVT Û(PUI�ÛXBOTÛ�Û�V, � O� ÛÛCVUÛ(S�ÛEÂNIJÛQSFTFOUTÛBÛQSPCMFN�ÛÛÛÛ�EI IV)�NÛBDD��ÛKVL�OÛ�GJTI��Û(S�ÛI�C U>J Û-JUI�Û VWÅT�ÛÛÛ�NV)T�NÛBDD��ÛNVLOÛ �NPVTF��Û (S�Û MU>J Û 4LU�ÛN � Û -BU�ÛN T�Û )FSFÛ UIFÛEFWFMPQNFOUÛXPVMEÛIBWFÛPDDVSSFEÛBGUFSÛUIFÛMPTTÛPGÛ�T�ÛÛÛÛÛTSVOL Û�TIJO�CPOF�ÛNBZÛCFÛBÛMPBOÛGSPNÛ*SBOJBO ÛCVUÛTFFÛ,PSUMBOEUÛ��C�ÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ TFNJWPXFMÛ CFGPSFÛ SFTPOBOU Û�8)3�ÛÛ /PÛFWJEFODFÛLOPXO�ÛÛÛÛÛÛ ��ÛÛ-BSZOHFBMÛ JOÛ JOMBVU Û BGUFSÛ TFNJWPXFMÛCFGPSFÛ TFNJWPXFM Û�8)8�ÛÛ )FSFÛNBZÛCFMPOHÛ �U J)�JOP��Û D JOÛ �IBXL��Û(S�� I�KTkNOJ Û 4LU�Û ZFOº��ÛTFFÛ��ÛBOEÛ��E�ÛÛÛ ��8[S�PEV]RKIEPWÛÛ ��ÛÛ5XPÛDPOTFDVUJWFÛMBSZOHFBMT Û))�ÛÛÛÛÛ /PÛFWJEFODF�ÛÛÛ ��ÛÛ5IFÛTFRVFODFÛ)3)ÛÛÛÛÛÛÛÛ 'PS�)3)�$ÛTFFÛPOÛ��Û3)$�ÛÛÛ ��ÛÛ5IFÛTFRVFODFÛ)8)�ÛÛÛÛÛ /PÛFWJEFODF��ÛÛÛÛÛÛÛÛ ��'0978)67�ÛÛÛÛ *Û EJTDVTTÛ BTÛ DMVTUFSTÛ HSPVQTÛ PGÛ DPOTPOBOUTÛ UIBUÛ BSFÛ SFGMFDUFEÛ CZÛPOFÛ QIPOFNFÛXIJDIÛ JTÛ OPUÛ JEFOUJDBMÛXJUIÛ POFÛ PGÛ UIFÛ DPOTPOBOUTÛ PGÛ UIFÛPSJHJOBMÛDMVTUFS�ÛCVUÛ*ÛBEEFEÛ�TQ ��TUÛBOEÛ�QTÛ UIFÛGJSTUÛBOEÛ UIFÛ MBTUÛCFJOHÛEJTQVUFE Û UIFÛ TFDPOEÛ GPSÛ UIFÛ TBLFÛ PGÛ DPIFSFODF�Û 4PÛ *Û EPÛ OPUÛ USFBUÛ BTÛ BÛDMVTUFSÛ�QO��XOÛTFFÛPOÛ�QÛPSÛ�US��S�ÛTFFÛPOÛ�U�Û*OÛDBTFÛPGÛEPVCU Û*ÛUBLFÛUIFÛHSPVQÛBTÛBÛDMVTUFS�ÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛS�Û�STÛ�Û ÛÛÛ�, � PSTPT�ÛP ÛÛ�CBDL�ÛCPEZQBSU�Û)JUU�ÛBSSBT Û(S�Û´RROJ ��PALI�N ORSOJ Û0)(ÛBST�ÛÛÛÛ�UST��Û U B BNJNÛ BMTPÛ U BSuBNJN�Û TFFÛ ��Û �XJUIFS��Û 4LU�Û U ZBUJ Û (S�ÛTE�RSOMAI Û-BU�ÛUPSSF Û(PUI�ÛHB× BVSTBO�ÛÛÛÛ


Û ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛ�T�ÛÛÛÛÛÛBÛ�TZÛ�ÛZ�ÛÛÛÛÛ��PTJP�Û�PZ ÛUIFÛHFO�ÛTH�ÛFOEJOHÛPGÛUIFÛP�TUFNT�Û5IJTÛJTÛUIFÛPOMZÛJOTUBODFÛPGÛUIFÛDMVTUFS�Û/PUFÛUIBUÛOPXIFSFÛFMTFÛJTÛ�ZÛQSFTFSWFEÛBTÛZ�ÛÛÛÛ��FTJP�Û � SÛ HFO�Û PGÛ �TPNFUIJOH� ÛXJUIÛ BEEFEÛ �SÛ GSPNÛ �SPÛ GPVOEÛ JOÛNBOZÛQSPOPVOT�Û5IFÛOPNJOBUJWFÛJTÛ[�J�ÛJ�FÛBTÛJOÛ1*&Û�L JE ÛHFO�Û�L FTP �(S�ÛTI ÛTEO��5IFÛQFSTPOBMÛGPSNÛJTÛPZ�S�Û�Û�PTJP�ÛÛÛÛÛÛÛÛÛÛÛÛCÛ�TXÛ�ÛL �ÛÛÛÛÛÛÛ5IJTÛJTÛPOFÛPGÛUIFÛTUSBOHFTUÛTPVOEÛMBXTÛPGÛ"SNFOJBO�Û*UÛJTÛFYQMBJOFEÛJOÛ,PSUMBOEU�TÛ DISPOPMPHZÛ CZÛ BTTVNJOHÛ �T� �� I� �� L Û Û JTÛ UIFÛ SFHVMBSÛTPVSDFÛ PGÛ L Û BOEÛ NPOPQIPOFNJDJ[BUJPOÛ �IX� �� I Û GPMMPXFEÛ CZÛ MPTTÛ PGÛMBCJBMJ[BUJPO�ÛÛÛ�TVPQOPT�ÛL VOÛ �TMFFQ�Û XJUIÛVÛ GSPN��PX��PQ�Û4LU�Û TWºQOB� Û-BU�ÛTPNOVT�Û(S�Û»PNOJÛ�Û�TVQOPT�ÛÛÛÛÛ�TVFT S�ÛL PZSÛ�TJTUFS��Û4LU�ÛTWºTBS� Û(PUI�ÛTXJTUBS�ÛÛÛÛÛÛÛ�TVJES�P��ÛL JSUOÛ�TXFBU��Û4LU�ÛTWJE� Û(S�ÛcDRW�J�Û�TVFJE�ÛJOÛ-BUW�ÛTWJÃESJ Û0)(ÛTXFJ[�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛDÛ�TSÛÛ�Û ÛÛMJLFÛ�ST ÛDG�Û��Û�ÛÛÛÛÛ�TVFT�S�PT�ÛL F Û�TJTUFS�ÛHFO�ÛTH��Û(PUI�ÛTXJTUST�ÛÛÛ�TVFTS��ÛL F JÛÛ�NBUFSOBMÛVODMF��ÛDG�Û4LU�ÛTWBTS ZB�Û�TJTUFS�TÛTPO� Û-BU�ÛTPCS OVTÛ�DPVTJO��ÛÛÛ�TSV��ÛB VÛ�DBOBM��Û4LU�ÛTSV�UÆ��Û/PUFÛUIFÛQSPUIFUJDÛWPXFM�ÛÛÛÛÛÛÛEÛ�TQÛ�ÛTQÛÛÛÛÛ�TQFS�O��Û TQB OBNÛ �UISFBUFO��Û 4LU�Û TQ OÌUJÛ �LJDLÛ BXBZ� Û -BU�Û TQFSO Û �TFWFS ÛEFTQJTF� Û0*D�ÛTQFSOBÛ�LJDLÛBXBZ��Û5IFÛFUZNPMPHZÛJTÛTFNBOUJDBMMZÛEJGGJDVMU�ÛÛÛ�TQFM��ÛB B�TQFMÛ�NZUI ÛGBCMF��Û(PUI�ÛTQJMMÛ�GBCMF� Û0&ÛTQFMM�ÛÛÛÛÛÛÛÛ5IJTÛ JTÛ UIFÛ EFWFMPQNFOUÛ XFÛ FYQFDUÛ QBSBMMFMÛ UPÛ �TU� �� TUÛ CFMPX�Û)PXFWFS ÛUIFSFÛBSFÛTFWFSBMÛFUZNPMPHJFTÛUIBUÛTFFNÛUPÛQPJOUÛUPÛQ Û�Û�TQ�Û*OÛDBTFÛTPNFÛPGÛUIFNÛBSFÛDPSSFDU ÛOPÛEJTUSJCVUJPOÛJTÛLOPXO�Û$G��ÛÛÛ�TQFMH��ÛQ F LÛ�MPOHÛQJFDFÛPGÛXPPE��Û0&ÛTQFMD Û0*D�ÛTQKBMLÛ�TQMJOU� ÛTQFMLKB�ÛÛÛ�TQFS�U��ÛÛQ FSU Û�BÛUPSOÛPGGÛQJFDF��Û0*D�ÛTQJPSSÛÛJE�ÛÛÛÛÛ$G�ÛBMTPÛQ BZDB OÛ �TQMFFO��Û4LU�ÛQM IºO� Û"W�ÛTQ S [BO� Û(S�ÛSPL�N ÛFUD�Û5IFÛXPSEÛ DBOOPUÛ CFÛ SFDPOTUSVDUFE Û CVUÛ BOÛ JOJUJBMÛ �TQ�Û TFFNTÛ QSPCBCMFÛ �Q�ÛXPVMEÛOPUÛTPMWFÛBOZUIJOHÛGPSÛ"SNFOJBO�ÛÛÛ�TQPVE��ÛQ PZU Û�[FBM��Û(S�ÛSPOUDH� Û-JUI�ÛTQºVE JV�Û5IFÛGJOBMÛDPOTPOBOUÛEPFTÛOPUÛSFQSFTFOUÛ�E ÛTPÛUIFÛFUZNPMPHZÛIBTÛCFFOÛEPVCUFE�ÛQFSIBQTÛUIFÛXPSEÛJTÛOPO�*OEP�&VSPQFBO�Û5IFÛFUZNPMPHZÛDBOÛIBSEMZÛCFÛSFKFDUFE ÛIPXFWFS�ÛÛÛÛÛÛ"Û RVJUFÛ EJGGFSFOUÛ SFMBUJPOÛ JTÛ TFFOÛ JOÛ "SN�Û TVOH�LÛ �NVTISPPN� Û (S�Û

198� 5��6��3��%HHNHV��

SPO�GGOJÛ�TQPOHF��ÛUIJTÛJTÛOPÛEPVCUÛBÛOPO�*&ÛXPSE�ÛÛÛÛÛ0OÛPQ JÛÛTFFÛPOÛ�QT� Û��ÛÛÛÛÛÛÛFÛ�TU��TUÛÛÛBMTPÛBGUFSÛTÛ�Û� ÛÛÛ�TUFJC��ÛTUJQFNÛ�VSHF ÛDPNQFM��Û(S�ÛSTEI�BW�ÛÛÛÛÛÛ�TUFS�JP��ÛTUFS ��TUFSJMF��Û4LU�ÛTUBS � Û(S�ÛSTEI>RA Û-BU�ÛTUFSJMJT Û(PUI�ÛTUBÆSP�ÛÛÛ�, � TUFS��ÛBTU ÛÛ�TUBS��Û4LU�ÛTUºS� Û(S�ÛA�STH�R Û-BU�ÛTU MMB�ÛÛÛ�VFT�UV��Û[�HFTUÛ�DMPUIFT��Û-BU�ÛWFTUJT Û(PUI�ÛXBTUJ�ÛÛÛ�EIVH, � U S�Û EVTUSÛ �EBVHIUFS��Û 4LU�Û EVIJUºS� Û (S�� UG�THR Û (PUI�Û EBVIUBS�Û/PUFÛ UIBUÛ UIFÛ MBSZOHFBMÛEJTBQQFBSFE ÛBOEÛ UIBUÛPOMZÛ UIFOÛ UIFÛ �U�Û DBNFÛ UPÛTUBOEÛBGUFSÛUIFÛTÛ�Û� �Û5IFÛQBMBUBMÛJOÛ"SNFOJBOÛJTÛEVFÛUPÛUIFÛQSFDFEJOHÛV�ÛÛÛÛÛÛÛÛGÛ�TL Û�T Û�TL ��D �ÛÛÛÛÛÛÛ"GUFSÛ�TÛOPÛ ÛDPVMEÛPDDVSÛJOÛ1*& ÛPOMZÛ�TLÛPSÛ�TL �Û"TÛ�L ÛCFDBNFÛL ÛJUÛJTÛQSPCBCMFÛUIBUÛ�TL ÛHBWFÛUIFÛTBNFÛSFTVMUÛBTÛ�TL�ÛÛÛ�, � FTL��ÛIBD JÛÛ�BTIÛUSFF��Û0)(ÛBTD�Û(S�ÛO�XU�HÛXJUIÛNFUBUIFTJT�ÛÛÛ�QS �TL��Û IBSD �BOFNÛ �BTL��Û 4LU�Û Q DDIºUJ Û -BU�Û QPTD Û �Û �QPSD�TD��Û *UÛ JTÛQSPCBCMFÛ UIBUÛ JOÛ UIFÛ HSPVQÛ �LTL�Û UIFÛ GJSTUÛ LÛ EJTBQQFBSFEÛ UISPVHIÛEJTTJNJMBUJPOÛBTÛJOÛ(SFFLÛBOEÛ-BUJOÛBOEÛBMSFBEZÛJOÛ1*& �ÛÛÛ�QSFJTLV��Û FS D Û �FMEFS Û QSJFTU��Û -BU�Û QS TDVTÛ �BODJFOU��Û (S�Û PRE�SBUJ Û $SFUBOÛPREI>GUJ�ÛÛÛÛ�TLFM)��ÛD FMVNÛ�TQMJU��Û-JUI�ÛTLÂMUJ�ÛÛÛ�TLJE�Û QFSIBQTÛ <�TLIJE�>Û �Û ��THIJE��Û D UJNÛ �TDSBUDI� Û �Û �D JU��Û 4LU�Û DIJE� ÛDIJOºUUJ Û(S�ÛSCI�ZW Û-BU�ÛTDJOE �ÛÛÛÛÛÛÛ0OÛQBMBUBMJ[FEÛGPSNTÛTFFÛ��ÛÛÛÛÛÛÛÛÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛMBCJBMTÛÛÛÛÛÛ�QTÛ�ÛT��ÛBMTPÛ�Q �ÛIBTÛCFFOÛQSPQPTFEÛÛÛ"TÛJOÛUIFÛDBTFÛPGÛ�TQ ÛUIFÛEFWFMPQNFOUÛJTÛOPUÛXFMMÛEPDVNFOUFE�Û5IFÛCFTUÛFWJEFODFÛJT�ÛÛÛ�QSFQ�T��ÛFSFTÛ�GBDF��Û(S�ÛPRE�PWÛ�CFÛDPOTQJDVPVT��ÛÛÛ�QTFVE��Û TVUÛ �MJF��Û(S�ÛYEU�DOMAI ÛYUDRO�J Û 4MPWBL�Û uVEJU��Û5IFÛXPSEÛNBZÛCFÛOPO�*OEP�&VSPQFBO ÛDG�Û(S�ÛYU� OJ�ÛÛÛÛÛÛ5IJTÛ EFWFMPQNFOUÛ JT Û JGÛ OPUÛ DPOGJSNFE Û BUÛ MFBTUÛ OPUÛ DPOUSBEJDUFEÛ CZÛ�QTU��TU��ÛÛÛÛ�QTUFOP�Û� ��ÛTUJOÛ�XPNBO�TÛCSFBTU��Û"W�ÛGuU OB� Û(S�ÛSTHNI�ON�ÛÛÛÛÛÛ5IFÛFWJEFODFÛGPSÛ�Q �ÛJT�ÛÛÛ�TFQT��Û FQ FNÛ �DPPL��Û (S�Û {YW�Û 5IFÛ XPSEÛ IBTÛ OPÛ GVSUIFSÛ SFMBUJWFTÛ BOEÛNBZÛCFÛBÛMPBOÛGSPNÛBÛOPO�*OEP�&VSPQFBOÛMBOHVBHF�ÛÛÛ�)PQT��Û PQ JÛ �XIJUFÛ QPQMBS��Û 0&Û¿QTF Û 3VTT�Û PTÆOBÛ �Û �PQT� Û DG�Û 5VSLJTIÛBQTBL�Û (FSNBOJDÛ GPSNTÛ MJLFÛ 0)(Û BTQBÛ NBZÛ IBWFÛNFUBUIFTJT�Û JUÛ JTÛNPSFÛ


QSPCBCMFÛ UIBUÛ UIFÛ "SNFOJBOÛ GPSNÛ HPFTÛ CBDLÛ UPÛ UIFÛ FBTUFSOÛ GPSNTÛXJUIÛ�QT��Û5IFÛXPSEÛEPFTÛOPUÛMPPLÛ*OEP�&VSPQFBO�ÛÛÛÛÛÛÛÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛBÛEFOUBMÛÛÛÛÛÛBÛ�EIZÛ�Û ÛÛÛ�NFEI�JP��ÛN � �NJEEMF��Û4LU�ÛNºEIZB� Û(S�ÛME�S�SOJ Û-BU��NFEJVT�Û0OÛ ÛDG�Û��ÛÛÛ�TF, � EI�JP��ÛB Û�SJHIU��Û4LU�ÛT EIÓ��ÛÛÛÛÛÛÛ$G�Û,PSUMBOEUÛ��B�Û5IFSFÛBSFÛOPÛDFSUBJOÛJOTUBODFTÛPGÛUÛPSÛEÛ�ÛZ�ÛÛÛÛÛ��ÛÛÛÛÛÛÛÛCÛ�UX��L ��ÛÛ�UVF�P��ÛL F[Û�ZPV�ÛEBU��BDD� ÛL PÛHFO��Û4LU�ÛUW N �UWB��Û(S�ÛSE�ÛBDD�Û�Û�UVF�ÛÛÛ�UXS��Û L BS�Û JOÛ L B BTVOÛ �GPSUZ��Û UIFÛ GPSNÛ JTÛ UIFÛ [FSPÛHSBEFÛ GSPNÛ �L FUV S ��L FUVS��Û 4LU�Û UVS ZB� Û (S�Û TRA� PEZAÛ �UBCMF�Û �Û �GPVS�GPPUFS� Û .ZD�� UPQF[BÛÛ�UPSQF[B��ÛÛÛÛÛÛÛÛÛ5IFÛEFWFMPQNFOUÛXFOUÛUISPVHIÛ�L ÛBÛEFWFMPQNFOUÛLOPXOÛGSPNÛPUIFSÛMBOHVBHFTÛBOEÛMPTTÛPGÛMBCJBMJ[BUJPOÛXFMMÛLOPXOÛGSPNÛ"SNFOJBO�ÛTFFÛUIFÛDISPOPMPHZ�ÛÛÛÛÛÛÛDÛ �EX� �� LÛ � 5IFÛ EFWFMPQNFOUÛ PGÛ UIJTÛ DMVTUFSÛ JTÛ NVDIÛ EFCBUFEÛ TJODFÛ.FJMMFUÛ BTTVNFEÛ UIBUÛ JUÛ SFTVMUFEÛ XPSEÛ JOJUJBMMZÛ JOÛ FSL�Û XJUIÛ QSPUIFUJDÛWPXFM�Û5IFÛNBUFSJBMÛPOÛXIJDIÛIFÛCBTFEÛUIJTÛJTÛUIFÛGPMMPXJOH�ÛÛÛ�EV �ÛFSLVÛ�UXP��Û4LU�ÛEW Û(S�ÛDU�W �DU�O Û0$4ÛE WB�ÛÛÛ�EVF, � SP��ÛFSLBSÛ�MPOH��Û(S�ÛDHRO�J ÛDG�Û4LU�ÛE Sº��ÛÛÛ�EVFJ��ÛFSLF�BZÛÛBPS�Û�CFÛBGSBJE�Û�Û�FSLJ�BZ ÛQSFT�ÛFSLO JN ÛFSLJX Û�GFBS��Û"W�ÛEWB Û�UISFBU� Û(S�ÛDEI�DWÛ�Û�EF�EVPJ�B�ÛÛÛ�FEV O�ÛFSLOÛ�CJSUIÛQBOHT��Û(S�ÛO�DU�NH�ÛÛÛÛÛ5IFSFÛ IBTÛ BMXBZTÛ CFFOÛ PQQPTJUJPOÛ UPÛ .FJMMFU�TÛ UIFPSZ�Û 5IFÛ FYQFDUFEÛPVDPNFÛPGÛEXÛJTÛL ÛQBSBMMFMÛUPÛUXÛ�ÛL ÛBTÛ.FJMMFUÛIJNTFMGÛCFMJFWFEÛFBSMJFS�Û5IJTÛEFWFMPQNFOUÛJTÛTIPXOÛCZ�ÛÛÛ�NFMEVJ��ÛNF LÛ�TPGU��Û4LU�ÛN EÓ� ÛGFN�ÛN EV Û-BU�ÛNPMMJTÛ�Û�NMEVJ��ÛÛ5IJTÛDBOÛCFÛ SFDPODJMFEÛ XJUIÛ .FJMMFU�TÛ UIFPSZÛ BTÛ CFJOHÛ UIFÛ JOUFSOBMÛ EFWFMPQNFOUÛBOE�ÛBGUFSÛSFTPOBOU�Û5IFSFÛBSFÛUIFÛGPMMPXJOHÛPCKFDUJPOT�Û��Û UIFÛQIPOFUJDÛBTQFDU�Û ��Û PCKFDUJPOTÛ UPÛ UIFÛ FUZNPMPHJFT Û BOEÛ��Û DPVOUFS�FWJEFODF�Û"EÛ��Û1SPQPOFOUTÛ PGÛ .FJMMFU�TÛ UIFPSZÛ IBWFÛ OPUÛ TVDDFFEFEÛ JOÛ HJWJOHÛ BÛTBUJTGBDUPSZÛ QIPOFUJDÛ FYQMBOBUJPOÛ PGÛ UIFÛ EFWFMPQNFOU Û BTÛ UIFZÛ NPTUMZÛBENJUÛXIJMFÛUIBUÛUPÛLÛJTÛVOQSPCMFNBUJD�Û"EÛ��Û"TÛSFHBSETÛFSLO ÛUIFÛ(SFFLÛGPSNÛ QPJOUTÛ UPÛ , � ��Û UIFÛ XPSEÛ QSPCBCMZÛ CFMPOHTÛ XJUIÛ O��DWN Û GPSÛ XIJDIÛOPUBCMZÛNWDO�JÛQSPWFTÛ, � ��Û5IFÛFUZNPMPHZÛNVTUÛUIVTÛCFÛHJWFOÛVQ�Û"TÛUPÛFSLBS Û UIFSFÛ JTÛ BOÛ"SNFOJBOÛ TVGGJYÛ �BS Û BOEÛ UIFÛ FYJTUFODFÛPGÛ FSL�BZOÛXJUIÛUIFÛTBNFÛNFBOJOHÛTVHHFTUTÛ UIBUÛ UIFÛXPSE� FSLBSÛDPOUBJOTÛUIJTÛTVGGJY�Û UIJTÛ

200� 5��6��3��%HHNHV��

BOBMZTJTÛ SFGVUFTÛ UIFÛ FUZNPMPHZ�Û 5IFÛ DPOOFDUJPOÛ XJUIÛ -JUI�Û FSEWBTÛ�TQBDJPVT�ÛNVTUÛQSPCBCMZÛ CFÛHJWFOÛVQ Û BTÛ JUTÛ BDDFOUÛQPJOUTÛ UPÛ BÛ �EI�Û UIFÛEFWFMPQNFOUÛPGÛ�EIXÛ JTÛVOLOPXO�ÛPOFÛFYQFDUTÛH�Û5IFÛXPSEÛGPSÛ �UXP�Û JTÛQSPCMFNBUJDÛ UPP Û BTÛ FS�Û DBOÛPOMZÛIBWFÛCFFOÛBEEFEÛBGUFSÛ UIFÛ MPTTÛPGÛ GJOBMÛWPXFMTÛ PUIFSXJTFÛ �FSLVÛXPVMEÛ IBWFÛ CFDPNFÛ �FSL�Û"OEÛ,PSUMBOEUÛ QVUTÛUIFÛ MPTTÛPGÛ GJOBMÛWPXFMTÛTUBHFÛ��ÛBGUFSÛ UIFÛSJTFÛPGÛQSPUIFTJTÛ ��ÛBOEÛUIFÛNFUBUIFTJTÛ ��Û GPSÛ �SL�ÛNVTUÛIBWFÛ BSJTFOÛ UISPVHIÛNFUBUIFTJTÛ GSPNÛ LS�Û �ÛEX��Û4PÛUIFÛFUZNPMPHZÛPGÛFSLVÛTFFNTÛJNQPTTJCMF�ÛÛÛÛÛ"EÛ��Û.PTUÛJNQPSUBOUÛJTÛUIFÛDPVOUFS�FWJEFODF ÛFTQFDJBMMZÛSFHBSEJOHÛUIFÛXPSEÛ GPSÛ �UXP��Û ,PSUMBOEUÛ GPMMPXTÛ 1FEFSTFOÛ JOÛ BTTVNJOHÛ UIBUÛ UIFÛXPSEÛXBTÛ �LV Û XJUIÛ FS�Û MBUFSÛ BEEFEÛ GSPNÛ FSFL Û �UISFF��Û.PTUÛ DPOWJODJOHÛ JTÛ LJD Û�DPOKPJOFE Û DPOUJHVPVT Û BEKPJOJOH Û UPHFUIFS�Û CFTJEFÛ FSLJD �TÛ �UXJDF��Û UIFÛXPSEÛDBOÛCFÛ DPNQBSFEÛXJUIÛ0)(Û[XJTLÛ �EPVCMF�Û �Û �EVJ�TL��Û5IFÛ TFDPOEÛGPSNÛXBTÛ�NPEFSOJ[FE�ÛBGUFSÛUIFÛOFXÛGPSNÛPGÛUIFÛXPSEÛGPSÛ�UXP� ÛUIFÛGJSTUÛUIFSFGPSFÛ EFWFMPQFEÛ BÛ NPSFÛ SFNPUFÛ NFBOJOHÛ GSPNÛ ��UXPÛ UPHFUIFS��Û2VJUFÛDPOWJODJOHÛJTÛBMTPÛLS�LJOÛ�EPVCMF�ÛXJUIÛLS�ÛGSPNÛ�LJS ÛXIJDIÛJTÛGPVOEÛJOÛFSLJSÛ �TFDPOE�Û DG�ÛNFLJOÛ �TJOHMF�Û/PUFÛBMTPÛLV Û �EPVCMF� ÛNPTUÛQSPCBCMZÛDPOOFDUFEÛXJUIÛ�UXP� ÛQFSIBQTÛGSPNÛ�EVP�QM�P��ÛÛÛÛÛÛÛ0OFÛ DPODMVEFTÛ UIBUÛ UIFÛ QIPOFUJDÛ EFWFMPQNFOUÛ PGÛ FSL�Û JTÛ JNQSPCBCMF ÛUIBUÛ UIFSFÛBSFÛPCKFDUJPOTÛ UPÛ UIFÛFUZNPMPHJFT ÛBOEÛ UIBUÛ GPSÛ �UXP�Û UIFSFÛBSFÛTFWFSBMÛGPSNTÛQPJOUJOHÛUPÛL�Û�Û�EX��ÛÛÛÛÛÛÛEÛ�5 �ÛÛÛÛÛÛ5IJTÛ JTÛ UIFÛ QSPCMFNÛ PGÛ UIFÛ DPSSFTQPOEFODFÛ 4LU�Û L Û �Û (S�Û KTÛ FUD�Û 5IFÛEJTDPWFSZÛPGÛ)JUU�ÛU LBOÛIBTÛTIPXOÛUIBUÛNPTUÛPGÛUIFTFÛGPSNTÛHPÛCBDLÛUPÛBÛEFOUBMÛ GPMMPXFEÛCZÛBÛWFMBS�Û"SNFOJBOÛIBTÛUXPÛPSÛUISFFÛ JOTUBODFTÛPGÛ UIJTÛDMVTUFS�ÛÛÛ�EI IV)��Û KVLOÛ �GJTI��Û(S�Û I� C U>J Û -JUI�Û VWÅT�Û *OÛ"SNFOJBOÛ UIFÛ EFOUBMÛXBTÛMPTUÛBOEÛUIFÛWFMBSÛEFWFMPQFEÛSFHVMBSMZ�Û0OÛ�L�ÛTFFÛ��ÛÛÛ�, � SU P��ÛBS Û�CFBS��Û4LU�Û L B� Û(S�Û�RKTOJ Û)JUU�ÛIBSUBHHB��Û'PSÛ� ÛXFÛFYQFDUÛ�D ÛJOTUFBEÛPGÛT�JOÛUIFÛTFSJFTÛT��D��K�ÛXJUIÛUIFÛQSFDFEJOHÛEFOUBMÛUIJTÛNBZÛIBWFÛ HJWFOÛ D �Û 5IJTÛ TPVOEÛNVTUÛ IBWFÛ CFDPNFÛ WPJDFEÛ BGUFSÛ SFTPOBOU Û BTÛVTVBMÛJOÛ"SNFOJBO ÛHJWJOH�K��5IFO��, � SU �Û�Û�BSKÛHPUÛUIFÛTVGGJYÛ�JP� ÛHJWJOHÛ�BSEZ�Û�ÛBS �ÛÛÛ�U J)�JO��ÛD JOÛ�LJUF��Û(S��gKTI>NOJ�Û4LU�Û ZFOº� Û"W�ÛTB OB��Û8FÛIBWFÛUIFÛTBNFÛEFWFMPQNFOUÛJOÛ"SNFOJBOÛBTÛBTTVNFEÛJOÛUIFÛQSFDFEJOHÛJUFN�Û5IFÛGPSNTÛHPÛCBDLÛUPÛBÛDPNQMJDBUFEÛQBSBEJHN ÛF�H�Û�U F)�J O ��U J)�JO�PT�ÛÛÛÛ ��ÛÛ$MVTUFSTÛCFHJOOJOHÛXJUIÛWFMBSÛÛÛÛÛÛÛBÛ�L Z Û�LZÛ�Û ÛÛÛÛÛ5IFÛSFTVMUÛJTÛUIFÛTBNFÛBTÛCFGPSFÛF �J ÛTFDUJPOÛ��Û


ÛÛ�LJPV��Û PH�BZÛ�*ÛXFOU��Û4LU�ÛDZºWBUF Û(S�ÛSEU�W�ÛÛÛÛ� O, � �TL� F�P��Û BOB �FNÛ�LOPX�ÛGSPNÛ�DBOB ��Û(S�ÛGI GNW� SKW Û-BU�ÛO TD �Û/PUFÛUIBUÛUIFÛ�T�ÛBMTPÛNFSHFEÛXJUIÛUIFÛSFGMFYÛPGÛUIFÛDMVTUFS�ÛÛÛ�, � L �J, � �ÛB �L Û�FZFT�ÛJGÛ�J, � ÛEFWFMPQFEÛUPÛ�ZB�BOEÛOPUÛUP�� ÛÛÛ�VPL � F�P��ÛHP �FNÛ�DSZ� ÛJGÛUIFÛWFSCÛJOEFFEÛIBEÛ� F�P��ÛÛÛÛÛÛ/PÛJOTUBODFTÛPGÛ�H ZÛ�Û Û ÛPSÛH IZÛ�Û � ÛBSFÛLOPXO�ÛÛÛÛÛÛÛCÛ� XÛ�ÛuÛÛÛ� V O �� VPO�N�ÛuVOÛ�EPH��Û4LU�Û VW Û(S�ÛKU�WN�ÛÛÛ�, � F VP��Û uÛ �EPOLFZ��Û JOÛ 1*&Û �IPSTF� Û 4LU�Û º WB� Û -BU�� FRVVT�Û 5IFÛ Û �Û �FJÛDPNFTÛGSPNÛ�FÛCFGPSFÛBÛQBMBUBMÛDPOTPOBOU�ÛTFFÛ��ÛÛÛÛÛÛÛDÛ� TÛ�ÛD ÛÛ5IFSFÛJTÛPOMZÛPOFÛJOTUBODFÛLOPXO�ÛÛÛ�TVF T�ÛWFD Û�TJY��Û"W�ÛYuWBu Û(S�Û{X ÛðE�X Û8FMTIÛDIXFDI�ÛÛÛÛÛ/PUFÛUIBUÛXFÛIBWF�ÛÛÛÛ� TE�� TUÛ�Û� UÛ�Û� U��ÛuU�JO�ÛÛWFuUBTBOÛ�TJYUFFO�ÛGSPNÛ�TVF T�EF N�ÛÛÛÛÛÛÛÛEÛ�L T Û�LTÛ�ÛD ÛQBMBUBMJ[FEÛuÛTFFÛ��ÛÛÛ5IFSFÛ BSFÛ OPÛ EBUB Û TPÛ UIFÛ SFTVMUÛ JTÛ JOGFSSFEÛ GSPNÛ � TÛ ��DÛ BOEÛ GSPNÛTL Û��G�ÛÛÛ'PSÛ BÛ �QBMBUBMJ[FE�Û GPSNÛ DG�Û �VFL TFSPTÛ �ÛHJuFS�Û 5IFÛ FYBDUÛ DPOEJUJPOTÛ PGÛUIJTÛ EFWFMPQNFOUÛ BSFÛ OPUÛ ZFUÛ DMFBS�Û Û 1FSIBQTÛ �L TQ� Û XIJDIÛ XBTÛ UIFÛPSJHJOBMÛ DMVTUFS Û HBWFÛ ��LuQ� Û DG�Û UIFÛ QSFDFEJOHÛ TFDUJPOÛ POÛ WFuUBTBO�Û $G�ÛBMTPÛPOÛ��ÛÛÛÛÛ ��ÛÛ4UPQÛ�ÛMBSZOHFBMÛÛ 5IJTÛJTÛPGÛDPVSTFÛUIFÛRVFTUJPOÛPGÛUIFÛ �UFOVFTÛBTQJSBUBF��Û4JODFÛUIFÛEJTDPWFSZÛ PGÛ UIFÛ MBSZOHFBMT Û JUÛ IBTÛ CFFOÛ QSPWFEÛ UIBUÛ 1*&Û EJEÛ OPUÛ LOPXÛWPJDFMFTTÛ BTQJSBUFTÛ CVUÛ UIBUÛ UIFÛ SFMFWBOUÛNBUFSJBMÛ HPFTÛ UPÛ BÛ MBSHFÛ FYUFOUÛCBDLÛ UPÛ DMVTUFSTÛ PGÛ TUPQÛ �Û MBSZOHFBM�Û 5IFÛ "SNFOJBOÛ NBUFSJBMÛ JTÛ WFSZÛEJGGJDVMU�Û8FÛTIBMMÛEJTDVTTÛUIFÛDMVTUFSTÛTFQBSBUFMZ�ÛÛÛÛÛÛÛBÛÛÛ�Q)Û�ÛQ Û ÛBMTPÛBGUFSÛTÛÛÛ�Q, � FM� �Û Q BZMÛ �MVTUFS��Û 4LU�Û QIBMHÓ�Û �SFEEJTI��Û /PUÛ WFSZÛ DPOWJODJOH�Û 'PSÛUIFÛ4BOTLJUÛXPSEÛGPSFJHOÛPSJHJOÛIBTÛCFFOÛDMBJNFE�ÛÛÛ�TQFS, � ��Û TQ J �L Û �EJTQFSTJPO� Û TQ FNÛ �EJTQFSTF� Û Q BSBUÛ �EJTQFSTFE�Û �Û�QS, � ��Û4LU�ÛTQIVSºUJÛ�Û�TQS, � �F�Û �LJDLÛBXBZÛXJUIÛUIFÛGPPU� Û-JUI�ÛTQÅSUJ�Û*ÛUIJOLÛ JUÛ OPUÛ WFSZÛ QSPCBCMFÛ UIBUÛ UIJTÛ SPPUÛ IBEÛ BÛ MBSZOHFBMÛ BGUFSÛ UIFÛ Q ÛCFDBVTFÛJUÛIBEÛPOFÛBUÛUIFÛFOEÛPGÛUIFÛSPPU�Û*ÛDPOTJEFSFEÛUIFÛQPTTJCJMJUZÛUIBUÛ(S�ÛA�SPAI�RWÛ�GMPVOEFS�ÛSFQSFTFOUTÛ�, � TQ�ÛBOEÛUIBUÛUIJTÛGJSTUÛMBSZOHFBMÛXBTÛJOÛTPNFÛMBOHVBHFTÛNPWFEÛUPÛUIFÛQPTJUJPOÛBGUFSÛUIFÛQ�ÛÛ

202� 5��6��3��%HHNHV��

ÛÛ�QF, � M� �ÛQ VMÛ�GBMM��Û-JUI�ÛQÓPMV �QÒMUJÛGSPNÛ�QM, � �ÛXJUIÛNFUBUIFTJT Û0)(ÛGBMMBO�Û*UÛTFFNTÛJNQSPCBCMFÛUIBUÛUIJTÛSPPUÛBMTPÛIBEÛBÛMBSZOHFBMÛBGUFSÛUIFÛQ�ÛÛÛ�QIV��ÛQ VL Û �CSFBUI��Û4LU��QI U�L �Û �CMPX� Û(S�ÛFU>SAÛ �CMPXJOH��Û5IFÛXPSEÛJTÛDMFBSMZÛPOPNBUPQPFUJDÛBOEÛDBOOPUÛCFÛVTFEÛBTÛFWJEFODFÛGPSÛ�Q)��Û/PUFÛUIBUÛJOÛ(SFFLÛUIFÛDMVTUFSÛEJEÛOPUÛHJWFÛQI��ÛÛÛÛÛÛÛ5IFÛ DPODMVTJPOÛ JTÛ UIBUÛ UIFSFÛ JTÛ OPÛ DPOWJODJOHÛ FWJEFODF�Û 4UJMMÛ JUÛ JTÛ OPUÛJNQSPCBCMFÛUIBUÛUIFÛPVUDPNFÛXPVMEÛIBWFÛCFFOÛQ �ÛÛÛÛÛÛÛCÛÛ�U)��U Û ÛÛÛ�QPSU)� �ÛPSU Û �DBMG��Û4LU�ÛQ UIVLB�Û �ZPVOHÛBOJNBM� Û(S�ÛPO�RTIJ �PO�RIJÛ.ZD�ÛJOTUS�ÛQM�ÛQPUJQJRFÛ�QPSUJQIJ�RF� �Û#VUÛUIFÛ4BOTLSJUÛXPSEÛJTÛWFSZÛMBUFÛBOEÛOPUÛGPVOEÛFMTFXIFSFÛJOÛ*OEP�*SBOJBO ÛBOEÛJOÛUIFÛ(SFFLÛXPSEÛUIFÛ�U�ÛEPFTÛOPUÛCFMPOHÛUPÛUIFÛSPPU�ÛÛÛ�QMU, � V��ÛZ�B U Û �MBSHF��Û4LU�ÛQ UIÓ� Û(S�ÛPLATU�J ÛCPUIÛ�CSPBE��Û5IFÛBOBMZTJTÛPGÛUIFÛ"SNFOJBOÛXPSEÛJTÛVODFSUBJO�ÛÛÛÛÛÛ4PÛIFSFÛBHBJOÛUIFSFÛJTÛOPÛDFSUBJOÛFWJEFODF�Û*GÛUIFÛPVUDPNFÛXPVMEÛCFÛU ÛJUÛ XPVMEÛ DPJODJEFÛ XJUIÛ U Û �Û �U Û CVUÛ BGUFSÛ BÛ SFTPOBOUÛ UIJTÛ XPVMEÛ IBWFÛCFDPNFÛE ÛTPÛPSU ÛBOEÛB U ÛXPVMEÛTIPXÛBÛTQFDJGJDÛSFGMFY�ÛÛÛÛÛÛÛDÛÛ�L ) �L)Û�ÛYÛÛBMTPÛBGUFSÛTÛÛÛ�L, � FE��YBDBOFNÛ�D�Û�Û�E�T��Û,PSUMBOEUÛ��B Û��G�Û<UIJTÛWPM� Û��G�>Û�CJUF��Û4LU�ÛLI EBUJÛ�DIFX�Û�L, � E� Û-JUI�ÛLºOEV �LºTUJ Û1PMJTIÛL TB Û�Û�L, � �FO�E� ÛXIJDIÛJTÛBÛTUSBOHFÛGPSNBUJPO�ÛJUÛDPVMEÛCFÛTJNQMZÛ�LPOE��Û5IFÛFUZNPMPHZÛJTÛEPVCUFE�ÛÛÛ�TL)FM��Û TYBMFNÛ �FSS Û TUVNCMF��Û 4LU�Û TLIBMBUFÛ WFSZÛ EPVCUGVMÛ (S�Û SFA�LLW ÛXIJDIÛXPVMEÛSFRVJSFÛ�L ��ÛÛÛÛÛ�T F, � LF)� �Û D BYÛ �CSBODI��Û 4LU�Û LI Û(PUI�ÛIPIB� �QMPVHI��Û -JUI�Û uBL¹ Û0$4ÛTPYB�Û5IFÛTUSVDUVSFÛPGÛUIFÛXPSEÛXPVMEÛCFÛSBUIFSÛTUSBOHF�Û5IFÛTIPSUÛ�BÛPGÛ4MBWJDÛDBOOPUÛCFÛFYQMBJOFEÛMBSZOHFBMÛCFUXFFOÛDPOTPOBOUTÛEPFTÛOPUÛWPDBMJ[F ÛPSÛJGÛXFÛBTTVNFÛ�P� ÛUIFÛ"SNFOJBOÛGPSNÛDBOOPUÛCFÛFYQMBJOFE�Û5IFÛGPSNTÛXJUIÛOBTBMÛ4LU�Û BOLÓ� Û0$4ÛTPL Û8FMTIÛDBJOD Û0*D�ÛI SÛQSPWFÛUIBUÛ UIFÛ XPSEÛ JTÛ OPO�*OEP�&VSPQFBO�Û "SN�Û D �Û DBOÛ SFQSFTFOUÛ �TL��Û T�NPCJMFÛJTÛGSFRVFOUÛJOÛ&VSPQFBOÛTVCTUSBUVNÛXPSET�ÛÛ 5IFÛ DPOOFDUJPOÛ PGÛ YB BNÛ �UPÛ QMBZ�Û PSÛ YB B FNÛ �SFBTTVSF�Û XJUIÛ (S�ÛCALA�WÛ�TMBDLFO ÛMPPTFO�ÛJTÛRVJUFÛEPVCUGVMÛBOEÛOPUÛSFMFWBOU�Û(S�ÛC ÛDBOOPUÛSFQSFTFOUÛ�L)��ÛrÛ5IBUÛHMVYÛ �IFBE�ÛJTÛDPHOBUFÛXJUIÛ-JUI�ÛHBMW¹ÛFUD�ÛJTÛRVJUFÛVODFSUBJO�Û5IFÛPSJHJOÛPGÛUIFÛ�YÛJTÛVOLOPXO�ÛÛÛÛ 5IFSFÛ JTÛOPÛ SFBTPOÛ UPÛ TVQQPTFÛBÛ MBSZOHFBMÛ JOÛ D UJNÛ �TDSBUDI�Û (S�ÛSCI�ZW�ÛTFFÛ��G�ÛÛ "HBJOÛ UIFÛ DPODMVTJPOÛ JTÛ UIBUÛ UIFÛNBUFSJBMÛ JTÛNFBHSFÛBOEÛNPTUMZÛVOSFMJBCMF�Û


Û ��38,)6�():)0341)287�� ÛÛJ�B��FB�ÛÛ 8IFOÛ JOÛ JOGMFDUJPOBMÛ GPSNTÛ PSÛ JOÛ XPSE�GPSNBUJPOÛ BOÛ BÛ DPNFTÛBGUFSÛBOÛJ ÛUIJTÛJÛJTÛMPXFSFEÛUPÛF�ÛÛÛUBSJÛ�ZFBS�ÛÛÛÛÛÛJOTUS�ÛUBSF�BXÛÛÛPSEJÛ�TPO�ÛÛÛÛ�Û�BL ÛEJNJOVUJWFÛTVGGJY�ÛPSEFBLÛÛÛÛÛÛÛ/PUFÛ UIBUÛ JGÛ �FB�Û JTÛ GPMMPXFEÛ CZÛ BOPUIFSÛ TZMMBCMF Û J�F�Û JTÛ QSFUPOJD Û JUÛ JTÛSFEVDFEÛUPÛ�F�ÛBDDPSEJOHÛUPÛUIFÛHFOFSBMÛSVMFTÛTFFÛ��ÛÛÛ�HJOJ�B�UVOÛ�XJOF�IPVTF�Û�Û�HJOFBUVOÛÛ�ÛHJOFUVOÛÛÛ*GÛUIFÛGPSNÛJOÛ�JÛJTÛNPOPTZMMBCJD ÛJUÛSFNBJOTÛVODIBOHFE�ÛÛÛKJÛ�IPSTF�ÛÛÛÛÛHFO�ÛKJPZ ÛÛÛÛKJ�B�XPSÛÛ�ÛIPSTFSJEFS�ÛÛÛ ��ÛÛ�FÛ�Û�FJ�� ÛCFGPSF�u � � ÛÛÛÛÛÛ5IFSFÛ BSFÛ BÛ GFXÛ JOTUBODFTÛ XIFSFÛ �FÛ CFDBNFÛ �FJ Û XIJDIÛ EFWFMPQFEÛOPSNBMMZÛUPÛ �ÛJUTÛQSFUPOJDÛGPSNÛJTÛJ�BTÛVTVBM�ÛÛÛ�, � F VP��Û uÛ�EPOLFZ��ÛÛÛ�NFEIJP��ÛN Û�NJEEMF��ÛÛÛ�VFL TFSP��ÛHJuFSÛ �OJHIU��Û0$4ÛWF FS Û(S�Û{SPEROJ�Û5IFÛEFWFMPQNFOUÛ �FÛ �Û�FJÛ�Û ÛSFEVDFEÛHSBEFÛJ ÛTPMWFTÛUIFÛWPDBMJDÛQBSU�Û4FFÛGVSUIFSÛ��E�ÛÛÛÛ�, � FH IJT�ÛJ Û�TOBLF��Û(S�Û´FIJ Û4LU�ÛºIJ� Û"W�ÛB J��Û"SN�Û ÛQPJOUTÛUPÛBÛQVSFÛWFMBSÛPSÛBÛMBCJPWFMBSÛXIJDIÛXBTÛQBMBUBMJTFEÛCFGPSFÛUIFÛJ�ÛJUÛNVTUÛUIFSFGPSFÛCFÛDPOOFDUFEÛXJUIÛ(S�Û´FIJÛFUD��ÛCVUÛUIFÛXPSEÛNVTUÛIBWFÛIBEÛ�FÛMJLFÛ(S�Û|CIJÛ XIJDI Û IPXFWFS Û IBTÛ � I�Û 5IFÛ �FÛ CFDBNFÛ �FJÛ �Û Û PGÛ XIJDIÛ UIFÛSFEVDFEÛHSBEFÛJÛDG�ÛHFO�ÛJ JÛXBTÛFYUFOEFEÛUPÛUIFÛOPNJOBUJWF�ÛÛÛÛ ��ÛÛ�JP��XPÛBOEÛJX��V�ÛUV O FBO�ÛÛÛ5IFÛ TVGGJYÛ ��JP�Û PGUFOÛOPUFEÛ ��J P��Û UIFÛHMJEFÛ JTÛ OPUÛQIPOFNJDÛ CFDPNFTÛ�XP�Û XIFOÛ BÛ TZMMBCMFÛ GPMMPXT�Û 5IJTÛ IBQQFOTÛ F�H�Û JOÛ UIFÛ JOGMFDUJPO�Û PSEJÛ�TPO� ÛHFO��PSEXPZ ÛJOTUS�ÛPSEXPW�ÛÛÛÛÛÛ5IFÛTFRVFODFÛ�JX�ÛGSPNÛXPSEÛGJOBMÛ�JXÛ�Û��JX� ÛPSÛGSPNÛ��JQ�ÛCFDPNFTÛ�V�ÛXIFOÛGPMMPXFEÛCZÛBÛTZMMBCMF ÛJ�F�ÛCFGPSFÛUIFÛTUSFTT�ÛÛÛÛBOJXÛ�XIFFM�Û HFO�ÛBOVPZÛÛÛÛUJXÛ�EBZ�ÛÛÛÛÛÛÛ HFO�ÛUV O FBOÛÛÛ5IJTÛEFWFMPQNFOUÛDBOÛBMTPÛCFÛTFFOÛPVUTJEFÛUIFÛJOGMFDUJPO�ÛÛÛ�, � S JQ�JPT�ÛBSDVJÛ�Û�BSDJX�JÛ�FBHMF��Û4LU�� KJQZº�Û�EBTIJOH��Û5IFÛGPSNÛBSDJXÛJTÛZPVOHFS�ÛÛÛ�TUJQ��ÛTUVBSÛ�TPMJE ÛIVHF�Û�Û�TUJX�BS�Û-JUI�ÛTUJQSÒTÛ�TUSPOH��ÛÛÛÛÛÛ8IBUÛ IBQQFOFEÛ JOÛ CPUIÛ DBTFTÛ JTÛ UIBUÛ JÛ XBTÛ SFEVDFEÛ UPÛ �Û 5IFÛEFWFMPQNFOUÛNBZÛ CFÛ FOWJTBHFEÛ BTÛ GPMMPXTÛ OPO�QIPOFNJDÛ FMFNFOUTÛ JOÛTVQFSTDSJQU�ÛÛÛJP�� P�� XP��XPÛÛÛ

204� 5��6��3��%HHNHV��

ÛÛJXP�� XP��V P��VPÛÛÛ0OÛUIJTÛEFWFMPQNFOUÛ,PSUMBOEUÛ�� Û��Û<UIJTÛWPM� Û�>�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ"TÛUPÛUV O FBO �UIFÛHFOJUJWFÛPGÛUJXÛ �EBZ� ÛUIFÛGPSNÛJTÛCBTFEÛPOÛ�UJX�JO� Û�JO��FO��Û)FSFÛXFÛGJOEÛCFTJEFÛJXÛ�ÛVÛBMTPÛ�J�� Û*UÛJTÛVOJRVFÛJOÛUIBUÛUIFÛSFEVDFEÛWPXFMÛJTÛXSJUUFOÛJOÛJOMBVUÛJOÛBOMBVUÛUIJTÛJTÛOPSNBM ÛUIPVHIÛSBSF�ÛOEFSL Û�Û�JOEFS�Û�Û�, � FOUFS��Û*UÛJTÛIFSFÛGPVOEÛBGUFSÛBÛWPXFM ÛPGÛXIJDIÛJUÛJTÛUIFÛPOMZÛJOTUBODF�Û*UÛIBTÛQIPOFNJDÛWBMVFÛIFSF ÛBOEÛJTÛUIFSGPSFÛJOEJDBUFE�ÛÛÛ ��ÛÛ�PQ$Û�Û�PX$Û�ÛV$ÛÛ "Û�QÛCFGPSFÛDPOTPOBOUÛEFWFMPQTÛJOUPÛX�Û *GÛ UIJTÛXÛ JTÛQSFDFEFEÛCZÛ�P ÛUIFÛTFRVFODFÛ�PXÛEFWFMPQTÛJOUPÛV�Û�TVPQOPT�ÛL VOÛ�TMFFQ� ÛWJBÛ�L PXO�Û4LU�ÛTWºQOB� Û-BU�ÛTPNOVT�ÛÛÛ ��ÛÛ5IFÛTFRVFODFÛ�O, �ÛÛ 5IJTÛ JTÛ UIFÛ BXDBOFNÛ �BOPJOU�Û QSPCMFN�Û *UÛ BQQFBSTÛ UIBUÛ UIFÛTFRVFODFÛ��POH � Û��BOH I�ÛCFDBNFÛ��BX � ��BX I�Û�Û�BXD� ÛSFTQ�Û�BXK��Û5IFÛGPSNTÛBSF�ÛÛÛ�, � OH ��ÛBXDBOFNÛ�BOPJOU��Û-BU�ÛVOHV ÛVOHVFO Û0*S�ÛJNCÛ�CVUUFS��ÛÛÛ�, � OH I�J��ÛBXKÛ �TOBLF��Û-BU�ÛBOHVJT Û0)(ÛVOD�Û-JUI�Û BOHÅT Û BDD�ÛBÊH Û IBTÛGVMMÛHSBEF�Û5IFÛ EFWFMPQNFOUÛ IBTÛ CFFOÛ NVDIÛ EJTDVTTFEÛ BTÛ JUÛ JTÛ QIPOFUJDBMMZÛVOFYQFDUFE�Û*UÛIBTÛCFFOÛTVHHFTUFEÛUIBUÛJUÛXBTÛBÛLJOEÛPGÛX�FQFOUIFTJT ÛUIFÛTFRVFODFÛ�XO,�Û MPTJOHÛJUTÛ �O��Û Û5IFÛGBDUÛ UIBUÛ UIFÛWFMBSÛCFDPNFTÛBÛQBMBUBMÛBGUFSÛUIFÛVÛTVHHFTUTÛBÛWFSZÛFBSMZÛEBUFÛ,PSUMBOEUnTÛTUBHFÛ��Û/PUFÛUIBUÛUIFÛEFWFMPQNFOUÛ QSPWFTÛ UIFÛ FYJTUFODFÛ PGÛ MBCJPWFMBSTÛ BUÛ POFÛ UJNFÛ JOÛ"SNFOJBO�ÛrÛ/PUÛSFMFWBOUÛJTÛBOLBOJNÛ �UPÛGBMM� ÛXIJDIÛJTÛOPUÛDPHOBUFÛXJUIÛ(PUI�ÛTJHRBOÛ�UPÛTJOL��ÛBOLJXOÛEPFTÛOPUÛIBWFÛBÛMBCJPWFMBSÛ0)(ÛFOLB�ÛÛÛ�, � FOH I�ÛSBUIFS��, � FVH I��ÛBXKJ�L Û�DPMMBS��ÛJUÛIBTÛCFFOÛFRVBUFEÛXJUIÛ(S�ÛAU�CH�N Û "FPM�Û AU��FHN Û �MFHNÛ �OFDL��Û IPXFWFS Û UIFÛ "FPMJDÛ GPSNTÛ BSFÛEPVCUGVMÛ BOEÛ VOFYQMBJOFE�Û (SFFLÛ EPFTÛ OPUÛ IBWFÛ TVDIÛ BÛ EFWFMPQNFOUÛBOEÛFWFSZUIJOHÛQPJOUTÛUPÛUIFÛGJSTUÛGPSNÛCFJOHÛUIFÛPSJHJOBM�Û5IFÛXPSEÛJTÛQSPCBCMZÛOPO�*OEP�&VSPQFBO�ÛÛÛÛ ��ÛÛ��BVH�Û�Û��BH Û�Û�BL�ÛÛÛÛ 5IJTÛEFWFMPQNFOUÛJTÛGPVOEÛJO�ÛÛÛ�, � FVH��ÛB FNÛ �HSPX�ÛXJUIÛQBMBUBMJ[FEÛ�LÛ�Û�H�Û-BU�ÛBVH SF Û(S�ÛAU�XA�NW�Û,PSUMBOEUÛ��C Û��Û<UIJTÛWPM� Û��>ÛTVHHFTUTÛUIBUÛJOÛUIJTÛXBZÛNBZÛBMTPÛCFÛFYQMBJOFEÛXJUIÛ�PVL�Û�Û�PL ��ÛÛÛ�PVLJEÛÛ�Û�, � PJV�L JE�ÛP Û�OPU�ÛXJUIÛQBMBUBMJ[FEÛL�ÛDG�Û(S�ÛOU��ÛÛÛ"HBJO ÛUIJTÛEFWFMPQNFOUÛJTÛVOFYQFDUFE�ÛÛÛ


Û ��ÛÛZ�FQFOUIFTJTÛÛ *OÛ��D ÛFQFOUIFTJTÛGSPNÛZÛIBTÛCFFOÛNFOUJPOFE ÛBTÛBMTPÛUIBUÛGSPNÛ�J��Û'VSUIFS ÛJÛ�Û ÛDPVMEÛDBVTFÛZ�FQFOUIFTJT ÛBTÛJTÛTIPXOÛCZ�ÛÛÛ�, � O S�ÛBZSÛ�NBO� ÛGSPNÛ�BZOS�Û4LU�ÛOºS� Û(S�Û�NH�R ÛFUD�ÛÛÛ5IJTÛ JTÛ UIFÛ POMZÛ FYBNQMF�Û BZSÛ �DBWF�Û DBOOPUÛ DPNFÛ GSPNÛ �BOU S�Û UIFÛDPOOFDUJPOÛXJUIÛ(S�Û�NTRONÛNVTUÛQSPCBCMZÛCFÛSFKFDUFE�ÛÛÛÛÛÛ ��ÛÛX�FQFOUIFTJTÛÛ *OÛBÛGFXÛDBTFTÛBOÛVÛTFFNTÛUPÛIBWFÛDBVTFEÛX�FQFOUIFTJTÛCFGPSFÛUIFÛQSFDFEJOHÛ DPOTPOBOU�Û *OÛ GBDUÛ UIFSFÛ JTÛ OPÛ DFSUBJOÛ FWJEFODF�Û 'PSÛ JOTUBODF ÛHJXUÛ �GJOEJOH�ÛDMFBSMZÛCFMPOHTÛUPÛUIFÛSPPUÛHJU� ÛBOEÛJUÛJTÛRVJUFÛQPTTJCMFÛUIBUÛUIFÛ FQFOUIFTJTÛ XBTÛ DBVTFEÛ CZÛ BÛ GPMMPXJOHÛ V Û CVUÛ JUÛ DBOOPUÛ CFÛEFNPOTUSBUFE�Û)JXU Û�NBUUFS ÛNPJTUVSF�ÛIBTÛCFFOÛDPOOFDUFEÛXJUIÛ4LU�ÛQJUÓ�Û�OPVSJTINFOU� Û -JUI�Û QJ U TÛ �MVODI� Û CVUÛ UIFÛ FUZNPMPHZÛ JTÛ TFNBOUJDBMMZÛEPVCUGVM�Û"ÛHPPEÛJOTUBODFÛTFFNTÛBX JÛ�TUSPOHÛESJOL�ÛCFTJEFÛ0*D�ÛPMÛ�CFFS�Û�Û�BMV��ÛÛÛÛÛÛÛ4PNFÛDBTFTÛBSFÛEPVCUGVM�Û5IVTÛBXSÛ �EBZ�Û�Û �, � F, � N S Û UISPVHIÛ�BNVSÛIBTÛ CFFOÛ FYQMBJOFEÛ BTÛ �Û �BNVSÛ �Û �BXNS Û CVUÛ PUIFSTÛ ,PSUMBOEUÛ QSFGFSÛ�BNVSÛ�Û�BVSÛXJUIÛMPTTÛPGÛUIFÛNÛCFGPSFÛV�Û"SUBXTSÛ�UFBS�ÛCFTJEFÛQM�ÛBSUBTVL ÛXIJDIÛ JTÛ OPÛ QSPCMFNÛ XIFOÛ UIFÛ DPOEJUJPOÛ JTÛ UIBUÛ POMZÛ BÛ MPTUÛ VÛ DBVTFTÛFQFOUIFTJTÛ JTÛ FYQMBJOFEÛ EJGGFSFOUMZÛ CZÛ ,PSUMBOEUÛ ��B�Û 0OÛ HFX Û TFFÛ��Û"XDBOFN �BXKÛBOEÛBXKJL �XFSFÛUSFBUFEÛTFQBSBUFMZ ÛJOÛ��ÛÛÛÛÛ ��ÛÛ�F Û�J ��FX Û�JX ÛÛÛÛ "ÛXPSE�GJOBMÛ� ÛQSPEVDFTÛBÛWFMBSÛHMJEFÛBGUFSÛFÛBOEÛJ ÛXSJUUFOÛ�FX ��JX ÛBMTPÛXIFOÛUIFÛ ÛXBTÛGPMMPXFEÛCZÛBOPUIFSÛDPOTPOBOU�ÛQJX DÛ�EJSUZ��Û*UÛPDDVSTÛ JOÛ JOMBVUÛPOMZÛXIFSFÛ JUÛ DBOÛCFÛFYQMBJOFEÛBTÛEVFÛ UPÛBOBMPHZÛ GSPNÛXPSE�GJOBMÛ QPTJUJPO�Û "TÛ UIFÛ GPSNTÛ XJUIPVUÛ �X�Û BSFÛ TUJMMÛ GPVOE Û UIFÛEFWFMPQNFOUÛXBTÛMBUF�ÛuJ �uF �uJX Û�CSBODI ÛTUBML��ÛDF �DJ �DJX ÛÛ�TUBML��ÛÛÛ 5IFÛ TFRVFODFÛ �FX Û CFDBNFÛ �JX �Û 5IVTÛ JUÛ CFDBNFÛ JEFOUJDBMÛ XJUIÛUIBUÛ XIJDIÛ IBEÛ PSJHJOBMÛ �JX �Û 5IFOÛ FÛ BOEÛ JÛ CFDBNFÛ NJYFEÛ VQ�Û 5IFÛDPOGVTJPOÛCFDBNFÛFWFOÛHSFBUFSÛXIFOÛ�X ÛXBTÛJOUSPEVDFEÛJOUPÛPUIFSÛDBTFÛGPSNT Û F�H�Û uJX JÛ GPSÛ UIFÛ PMEÛ GPSNÛ uMJÛ XIJDIÛ TIPXTÛ UIBUÛ UIFÛ GPSNÛ IBEÛPSJHJOBMMZÛBOÛ�J��Û0GÛDPVSTF ÛOPUÛBMMÛGPSNTÛBSFÛBUUFTUFE�Û5IVT ÛXFÛIBWFÛFX Û�PJM�Û GSPNÛ(S�Û |LAION Û HFO�Û JX PZ�Û UIFÛ OPNJOBUJWF Û PGÛ DPVSTF ÛNVTUÛ BMTPÛIBWFÛIBEÛUIFÛGPSNÛ�JX ��ÛÛÛÛÛÛÛÛÛÛ ��ÛÛD Û�ÛTÛCFGPSFÛD � ÛÛÛÛ 8IFOÛJOÛNPSQIPMPHZÛD ÛDBNFÛUPÛTUBOEÛCFGPSFÛD � ÛJUÛCFDBNFÛT�ÛÛÛTJSFD JD ÛÛÛÛ��ÛTH�ÛBPS�ÛTVCK�Û�*ÛXJMMÛMPWF� ÛCFTJEFÛÛÛTJSFTD FTÛÛ�ÛTH�ÛÛ�Û��D �D FTÛÛÛ

206� 5��6��3��%HHNHV��

ÛÛTJSFT JL ��ÛQM�ÛÛ�Û��D � JL ÛÛÛÛ0OÛUIFÛFOEJOHTÛTFFÛ,PSUMBOEUÛ��D�ÛÛÛ ��ÛÛ'JOBMÛ�ZÛBGUFSÛWPXFMÛÛ �ZÛBGUFSÛWPXFMÛJTÛOPUÛQSPOPVODFEÛJOÛ�BZ Û�PZ�ÛDG�ÛUIFÛHFO�ÛFOEJOHÛPGÛUIFÛ P�TUFNTÛ ��PTJPÛ �Û �PZ�Û 8FÛ EPÛ OPUÛ GJOEÛ ��FZ Û BTÛ �FJÛ IBEÛ CFDPNFÛ ÛQSPOPVODFEÛ�F�Û)PXFWFS ÛBGUFSÛ JÛBOEÛVÛ JUÛEJTBQQFBSFEÛTPÛFBSMZÛUIBUÛJUÛ JTÛOPÛMPOHFSÛXSJUUFO�ÛÛÛ�LJVUJT�Û VÛ�EFQBSUVSF ÛKPVSOFZ ÛXBZ�ÛGSPNÛ� VZ�Û4LU�ÛDZVUJ��ÛÛÛ� MVUPT�ÛMVÛ�GBNPVT� Û�Û�MVZ�Û4LU�Û SVUº� Û(S��KLUTO�J�Û,PSUMBOEUÛUIJOLTÛUIBUÛUIFÛJOUFSNFEJBUFÛGPSNÛXBTÛSBUIFSÛ�MVX��DG�Û��H�ÛÛÛ ��ÛÛ-PTTÛPGÛTUPQTÛCFUXFFOÛDPOTPOBOUTÛÛ 5IFÛWFSCTÛEB OBNÛ �SFUVSO� ÛCB OBNÛ �MJGU Û SBJTF� Û HJWFOÛ UIFJSÛ BPSJTUTÛEBSKBZ �CBSKJ ÛHPÛCBDLÛUPÛ�EBSK�OBNÛFUD�ÛXJUIÛMPTTÛPGÛUIFÛTUPQ�Û*OÛPUIFSÛDBTFTÛUIFÛ TUPQÛ NBZÛ IBWFÛ CFFOÛ SFTUPSFEÛ PSÛ IBWFÛ BÛ EJGGFSFOUÛ IJTUPSZ�Û FSEOVNÛ�TXFBS�ÛQSPCBCMZÛEFSJWFTÛGSPNÛ�FSEV�OVN ÛXIFSFÛUIFÛVÛCFDBNFÛ ÛBTÛVTVBM�ÛÛÛ ��ÛÛ.FUBUIFTJTÛÛÛ "Û DMVTUFSÛ TUPQÛ �Û SÛ XBTÛ NFUBUIFTJ[FE Û BUÛ BÛ TUBHFÛ XIFOÛ JOJUJBMÛWPJDFMFTTÛTUPQÛCFGPSFÛDPOTPOBOUÛIBEÛEJTBQQFBSFE�Û*OÛUIBUÛDBTFÛBÛSFNBJOJHÛS�Û HPUÛ BÛ QSPUIFUJDÛ WPXFM Û FSFXJN� �BQQFBS�Û �� QSFQ��Û *OÛ JOMBVUÛ UIFTFÛ TUPQTÛCFDBNFÛ �X�Û �QBUSPT� �� IBXS � HFO�Û TH��Û 4PÛ POMZÛ 1*&Û �C � �CIÛ FUD�Û BSFÛDPODFSOFEÛXIJDIÛSFHVMBSMZÛCFDBNFÛQ �CÛFUD�ÛJOÛ"SNFOJBO�Û5IFÛNFUBUIFTJTÛPDDVSSFEÛOPUÛPOMZÛJOÛJOMBVU ÛCVUÛBMTPÛJOÛBOMBVUÛQSPCBCMZÛBGUFSÛBÛTIXBÛIBEÛBSJTFOÛ CFGPSFÛ UIFÛ DMVTUFS�Û *OÛ UIFÛ MBUUFSÛ DBTFÛ UIFÛ OFXMZÛ JOJUJBMÛ S�Û HPUÛ BÛQSPUIFUJDÛ WPXFMÛ FÛ PSÛ B�Û TFFÛ DI�Û ��ÛsÛ *UÛ JTÛVOLOPXOÛXIBUÛIBQQFOFEÛXJUIÛHSPVQTÛPGÛUIFÛUZQFÛ�E Û�EI�GPMMPXFEÛCZÛ�M� Û�O�ÛJOÛUIFÛTBNFÛQPTJUJPO ÛBTÛOPÛJOTUBODFTÛBSFÛLOPXO�ÛÛÛÛ*OJUJBMÛDBTFTÛBSFÛHJWFOÛGJSTU�ÛÛÛÛÛÛÛ�CIS��SCÛÛÛÛ�CISF, � U S��F CBZS��CSPUIFS� ÛUISPVHIÛ�FSCBZS��ÛÛÛÛÛ�CISF, � VS��B CJXSÛ�TQSJOH� ÛWJBÛ�CSJXBS��CSJXS��BCSJXS��BSCJXS��ÛÛÛÛÛ�EIBCISP��EBSCJOÛ�TNJUI��ÛQSPCBCMZÛBÛOPO�*&ÛXPSE�ÛÛÛÛ� VCISP��TVSC��QVSF ÛIPMZ��ÛÛÛÛÛÛÛ�ES��SU��ÛÛÛÛÛ�ESB �� BSUBXTSÛ �UFBS��Û QM�Û BSUBTVL �Û0)(Û USBIBO �(S�ÛD�KRU��4FFÛPOÛ UIFÛXPSEÛ,PSUMBOEUÛ��B��ÛÛÛÛÛ�T JESP��L JSUOÛ�TXFBU��Û(S�ÛI@DRW�J �-BUW�ÛTWJÃESJ��-BU�ÛT EPS��TXPJE T�ÛÛÛ�CIJESP��CJSUÛ�SJHJE ÛSVEF��Û04ÛCJUUFSÛ�CJUUFS��ÛÛÛÛÛÛÛ�H S��SL��ÛÛÛ


ÛÛ�H SF, � �O��FSLBOÛ�NJMMTUPOF��Û-JUI�ÛHÅSOB Û4LU�ÛHS WBO��ÛÛÛÛÛÛÛÛ� ITS��SKÛÛÛÛÛÛ5IFÛPOMZÛJOTUBODFÛJT�ÛÛÛÛ�NF ITSJ��ÛNFSK��OFBS��Û(S�ÛME�CRIÛÛ�VQÛUP��Û4FFÛ��ÛÛÛÛÛÛÛ�HIS��SH��ÛÛÛÛÛ�HISPOEJ�Û�ÛBSHBOE��XPNC��Û0$4ÛHSPE ��CSFBTU��ÛÛÛÛÛ*OÛUIJTÛXBZÛ"SNFOJBOÛBWPJEFEÛDPOTPOBOUÛDMVTUFSTÛBUÛUIFÛCFHJOOJOHÛPGÛBÛTZMMBCMF�Û *UÛ IBEÛ OPUIJOHÛ UPÛ EPÛXJUIÛPQFOÛ TZMMBCMFTÛ TVSC�� EBS�CJO�Û �EBCSJO�DPVMEÛIBWFÛCFFOÛQSPOPVODFEÛEB�CSJO �MJLFÛQB�USPTÛJOÛ"UUJD�Û5IJTÛTFFNTÛUPÛCFÛ DPOOFDUFEÛXJUIÛ UIFÛ GBDUÛ UIBUÛ JOÛ"SNFOJBOÛ BMMÛ PMEÛ DPOTPOBOUÛ DMVTUFSTÛIBEÛCFDPNFÛTJOHMFÛQIPOFNFT�Û�TL��D ��LZ�� ÛFUD�Û4VDIÛMBOHVBHFTÛEPÛOPUÛOPSNBMMZÛIBWFÛHFNJOBUFT ÛXIJDIÛJTÛBMTPÛUSVFÛPGÛ"SNFOJBO�ÛÛÛÛÛ"QQBSFOUÛ DMVTUFSTÛ BTÛ JOÛ ESBOEÛ �EPPS�QPTU�Û IBWFÛ BOÛ � XIJDIÛ JTÛ OPUÛXSJUUFO�ÛIFSFÛE S��DPNFTÛGSPNÛEVS�Û�EPPS��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ �� 6)0%8-:)�',632303+=��Û Û��ÛÛ*OUFSNFEJBUFÛTUBHFTÛÛÛ *OÛUIFÛGPSFHPJOHÛXFÛIBWFÛNPTUMZÛQSFTFOUFEÛUIFÛ1*&ÛTUBSUJOH�QPJOUÛBOEÛUIFÛSFGMFYÛBUÛ UIFÛFBSMJFTUÛTUBHFÛPGÛ"SNFOJBOÛLOPXOÛUPÛVT�Û)PXFWFS ÛUIFTFÛ EFWFMPQNFOUTÛ XFOUÛ UISPVHIÛ TFWFSBMÛ JOUFSNFEJBUFÛ TUBHFT�Û *GÛ XFÛXBOUÛ UPÛ VOEFSTUBOEÛ XIBUÛ SFBMMZÛ IBQQFOFEÛ XFÛ IBWFÛ UPÛ TUVEZÛ UIFTFÛTVDDFTTJWFÛTUBHFT�Û"TÛUIFÛEFWFMPQNFOUTÛMFBEJOHÛUPÛ"SNFOJBOÛPGUFOÛXFSFÛOPUÛTJNQMF ÛUIFÛSFMBUJWFÛDISPOPMPHZÛPGÛUIFTFÛTUBHFTÛHJWFTÛVTÛBÛQPTTJCJMJUZÛUPÛ DIFDLÛ OFXÛ FUZNPMPHJDBMÛ QSPQPTBMT�Û UIFÛ TPVOEÛ MBXTÛNVTUÛ GPMMPXÛ UIFÛTBNFÛPSEFSÛFWFSZÛUJNF�ÛÛ "ÛDISPOPMPHZÛPGÛBÛMBSHFÛOVNCFSÛPGÛEFWFMPQNFOUTÛIBTÛCFFOÛHJWFOÛCZÛ ,PSUMBOEUÛ ��C Û BOEÛXFÛ TIBMMÛ HJWFÛ BÛ TVSWFZÛ PGÛ UIFNÛ IFSF�Û #FGPSFÛUIBUÛJUÛXJMMÛCFÛVTFGVMÛUPÛHJWFÛBÛTVSWFZÛPGÛUIFÛJOUFSNFEJBUFÛTUBHFTÛXIJDIÛIFÛQSPQPTFE�ÛÛÛ *UÛTIPVMEÛCFÛOPUFEÛUIBUÛOPUÛBMMÛEFWFMPQNFOUTÛIBWFÛCFFOÛMJTUFEÛJOÛUIJTÛ DISPOPMPHZ Û MBSHFMZÛ CFDBVTFÛ JUÛ XBTÛ VODMFBSÛ XIFSFÛ UIFZÛ IBEÛ UPÛ CFÛJOTFSUFE�ÛÛÛÛÛ�Q��F��Û�ÛI�Û��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ7F7��ÛXÛ��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛF$��Û�G$ �$��C�Û�G$Û�ÛX$Û��CÛÛÛ�UÛ�Û� Û��Û�ÛU Û��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ7 7Û�ÛZÛ��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ $ÛÛÛ�Û�G$ �$��C�Û�G$Û�ÛX$��CÛÛÛÛ�LÛ�Û� Û��Û�ÛL Û��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ7 7Û�ÛZÛ��Û

208� 5��6��3��%HHNHV��

ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ $ÛÛÛ�Û�G$ �$��C�Û�G$Û�ÛX$Û��CÛÛÛÛ� ��Û� Û�Û�ÛTÛ�ÛÛÛÛ�TÛ�Û�IÛ�Û�Û[FSPÛ��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ��IÛ�Û� Û��Û�ÛL Û��ÛÛÛ�XÛ�Û� Û��Û�Û� Û��Û�ÛHÛ��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ� Û�Û� ÛCFGPSFÛV �SÛÛ��DÛ�Û[FSPÛ��DÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ�� Û�ÛXÛ��BÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ�� SÛ�Û�XSÛ��CÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ�� Û�Û�� Û��DÛ�Û� ��Û�Û�HÛ��ÛÛÛÛ$MVTUFSTÛÛÛÛ�TXÛ�Û�IXÛ�Û�Û�I Û�DÛ�Û� ��Û�Û� Û��Û�ÛL Û��ÛÛÛÛÛ�TS �ST��Û�IS ��SIÛ�Û�Û ÛÛ��ÛÛÛÛ�TLÛ�Û� Û�Û�Û�DÛ�Û�ÛD Û��ÛÛÛ�UXÛ�Û�U ��Û�Û�L Û�Û�Û� Û��Û�Û� Û��Û�ÛL Û��ÛÛÛÛ�EXÛ�Û�E ��Û�Û�H Û�Û�Û�L ÛÛEFWPJDJOHÛPGÛUIFÛHMPUUBMJDÛTUPQTÛCFGPSFÛ��Û�ÛLÛ��ÛÛÛ� XÛ�Û�D Û�Û�Û�T Û�Û�ÛuÛÛ��DÛÛÛÛÛ ��ÛÛ3FMBUJWFÛDISPOPMPHZÛ *ÛHJWFÛ,PSUMBOEUnTÛDISPOPMPHZÛ ��C ÛXJUIÛUIFÛDIBOHFÛJOEJDBUFEÛJOÛ,PSUMBOEUÛ�� Û��Û<UIJTÛWPM� Û��> ÛO�Û�ÛBOEÛ�ÛÛ��CÛ�Û��C ÛXJUIÛIJTÛOVNCFSJOH�ÛÛÛÛÛ�� 4VSXS�-RHS�)YVSTIER�Û 5IFÛ VOBTQJSBUFEÛ WPJDFEÛ DPOTPOBOUT Û XIJDIÛXFSFÛQSFHMPUUBMJ[FE ÛXFSFÛSFUBJOFEÛBTÛTVDIÛCVUÛEFWPJDFEÛJOÛ"SNFOJBOÛVQÛ UPÛ IJTUPSJDBMÛ UJNFT�Û 5IFÛ WFMBSTÛ BGUFSÛ VÛ XFSFÛ OFVUSBMJ[FE Û HJWJOHÛQBMBUPWFMBSTÛ ÛFUD�ÛÛÛÛÛ��0SWW�SJ�EWTMVEXIH�WXSTW��ÛÛ4PÛ1*&Û�U��E��EIÛÛCFDBNFÛ�U��EÛÛÛ��E�ÛÛÛÛÛ*UÛ JTÛBTTVNFEÛUIBUÛ UIFÛ MPTTÛPGÛBTQJSBUJPOÛXBTÛBÛEJBMFDUBMÛ *OEP�&VSPQFBOÛEFWFMPQNFOU ÛXIJDIÛ"SNFOJBOÛ TIBSFEÛXJUIÛ(FSNBOJDÛBOEÛ#BMUP�4MBWJD ÛBOEÛGVSUIFSÛXJUIÛ"MCBOJBO Û*SBOJBOÛBOEÛ5PDIBSJBO�ÛÛÛÛÛÛÛ��6MWI�SJ�RI[�PEFMEPMÎH�WXSTW�ÛÛÛ5IFÛ GPMMPXJOHÛ DMVTUFSTÛ CFDBNFÛ NPOPQIPOFNJD Û BTÛ JOEJDBUFEÛ XJUIÛ UIFÛGJOBMÛSFTVMUÛBEEFEÛJOÛUIFÛUIJSEÛMJOF�ÛÛÛÛÛÛÛÛÛÛÛÛÛ�� X��UX��EXÛÛÛÛUISPVHIÛ��D ��U ��E ÛÛÛCFDBNFÛÛ��u��L ��L��MBUFSÛÛÛÛÛÛ5PÛUIFÛTBNFÛTUBHFÛBSFÛEBUFE�ÛÛBÛÛ��VH��V �� V �PN��MVDÛ�ZPLF��ÛÛ��FVL��FV ��MFV ��MPZTÛ�MJHIU�ÛÛCÛÛ�BVH��BH ��BVH��B FNÛ�HSPX�ÛÛÛÛÛÛÛ�PVL��PL ��, � P V�L F��PVL F��P Û�OPU�ÛÛ


ÛDÛÛ�BOH I��BX ��Û�BOH I�J��BXKÛ�TOBLF�ÛÛÛÛÛÛÛ�POH ��BX ��POH �Û��BXDBOFNÛ�BOPJOU�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ��%WWMFMPEXMSR�SJ�� "� ÛERH��WO�"� ��ÛÛÛÛÛÛ�4FFÛGVSUIFSÛVOEFSÛ��ÛÛÛÛÛÛ��Û"TTJNJMBUJPOÛPGÛ�TXF VS �� XF VS �Û�ÛTLFTVS�Û�TUFQNPUIFS� ÛTFFÛ��ÛÛÛÛÛÛ��W�"��L��FYDFQUÛBGUFSÛDPOTPOBOUÛPSÛCFGPSFÛPCTUSVFOU�ÛÛ*OUFSWPDBMJDÛIÛXBTÛMPTUÛMBUFSÛ�� ÛCVUÛGJOBMÛ�IÛCFDBNFÛ�� Û��Û�Û�L Û��ÛÛÛÛÛÛ�� JVSQ�� Û<�> �"�W��ERH�� JVSQ��WOÛ<�> �"��GÛMBUFSÛ�ÛD �ÛÛÛÛÛÛÛ��6IHMWXVMFYXMSR�SJ�PEFMEPMÊXMSR�ÛÛÛÛBÛ�K Û�ÛKÛÛ IXPO��K PO�Û<�>��KBZOÛÛÛÛCÛ�D �Û�Û�T ��Û� XPO�Û�Û�T VO��uVO��EPH�ÛÛÛÛDÛ�IX�GSPNÛ�TXÛ�ÛI ÛÛÛ�TXFT S��Û�I FIVS��ÛL PZS��TJTUFS�ÛÛÛÛÛÛ��4EPEXEPMÊXMSR��Û"MMÛWFMBSTÛXFSFÛQBMBUBMJ[FEÛCFGPSFÛGSPOUÛWPXFMT�Û�L ��H Û��H IÛ�Û Û� �� ÛÛÛÛÛÛ��X �"��O ��UX��L ��ÛL �F[��UIPV�ÛBDD��Û�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ�H �"��K ��Û�NFMEX��NFMH ��NF L��TPGU�ÛÛÛÛÛÛ��0IRMXMSR��ÛÛÛÛÛÛ�QÛÛ�U��L��L ��F�� ÛÛÛ5IJTÛ IBQQFOFEÛ BGUFSÛ UIFÛ WPJDJOHÛ PGÛ UIFTFÛ DPOTPOBOUTÛ BGUFSÛ SFTPOBOU�Û�N UP��NBSE� �NBO� Û�QFOL FÛ�ÛIJOH�Û5IFÛEFWPJDJOHÛPGÛUIFÛHMPUUBMJDTÛ�E��UÛFUD�Û XBTÛ QSPCBCMZÛ BOUFSJPSÛ UPÛ UIFÛ WPJDJOHÛ BGUFSÛ SFTPOBOUTÛ CFDBVTFÛ UIFÛMBUUFSÛ EJEÛ OPUÛ CMPDLÛ UIFÛ EFWPJDJOH�Û � SE� �� TJSUÛ �IFBSU� Û � PS �PN� �� HPSDÛ�XPSL��ÛÛÛÛÛÛ'JOBMÛOBTBMTÛBGUFSÛWPXFMÛXFSFÛMPTUÛWFSZÛFBSMZ ÛXIJMFÛTZMMBCJDÛOBTBMTÛJOÛBVTMBVUÛXFSFÛSFUBJOFE�ÛÛÛÛÛÛ/BTBMTÛCFGPSFÛGSJDBUJWFTÛXFSFÛMPTU�ÛÛÛÛÛÛÛÛÛÛ�PNTPT��VTÛ�TIPVMEFS�ÛÛÛÛÛÛ�I�GSPN��TÛXBTÛMPTU�Û"TTJNJMBUJPOÛPGÛ�I�UPÛSÛHJWJOHÛ �ÛÛÛÛÛÛÛ�IÛ�Û[FSPÛÛÛÛÛÛÛÛÛ�VFT�OV��[�HFOVN��ESFTT�ÛÛÛÛF/��ÛJ/ÛXBTÛMBUFS�ÛIFSFÛUIFÛ�FÛJTÛBOBMPHJDBM�ÛTFFÛ��E��BGUFSÛ� ��VÛCVUÛCFGPSFÛ�FV��PZ�ÛÛÛÛ�TXFT S��IXFIVS��L PZS��TJTUFS�ÛÛÛÛÛÛ�ZÛXBTÛMPTUÛBGUFSÛ� ��J�ÛCVUÛCFGPSFÛUIFÛMPTTÛPGÛQIPOFNJDÛRVBOUJUZ�Û�USF FT�� SF ��FSFL ��UISFF�ÛÛÛÛÛÛÛ��6MWI�SJ�RI[�ZIPEV�JVMGEXMZIW��ÛÛÛ�BÛÛ�X�� ÛÛXIJDIÛMBUFSÛCFDBNFÛH Û��ÛÛÛÛÛÛCÛÛ�I ��GSPNÛ�T Û�Û ÛÛÛÛÛ�TVFT S��I FIVS�� FVS��L PZSÛÛÛÛÛDÛÛ��I�� Û��Û�L �ÛBUÛ��ÛÛÛÛOPN�ÛQM�Û��FT��L �ÛÛÛÛÛÛÛ��7MQTPMJMGEXMSR�SJ�GSRWSRERXEPÛEVXMGYPEXMSR�ÛÛÛÛÛÛÛBÛÛ�F�Û�ÛI�Û �QS �TL� ��IBSD BOFNÛÛ�BTL�ÛÛÛÛÛÛCÛÛ�FÛÛÛ� ÛÛ� ÛÛ�Û�GÛÛCFGPSFÛDPOTPOBOUÛÛÛÛÛÛÛÛÛ�Q, � USPT�HFO�ÛTH�Û�Û�IB SPI��IBGSP Û��IBXSÛÛÛÛÛÛÛÛÛÛ ÛÛÛÛÛ ÛÛÛÛ ÛÛÛÛÛ�Û[FSPÛJOJUJBMMZÛCFGPSFÛDPOTPOBOUÛÛ�ÛÛÛ Û �QSFQ��SFF ��FSFXJN��BQQFBS�ÛÛ

210� 5��6��3��%HHNHV��

ÛÛÛDÛÛMPTTÛPGÛMBCJBMJ[BUJPOÛCFGPSFÛV �DPOTPOBOUBMÛS��ÛÛÛÛÛÛOPUÛCFGPSFÛP Û,PSUMBOEUÛQ�D�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛ�OFV�SP��OPXSP��OPSÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ�CISF, � �VS�PT��CSF S�PT�HFO�ÛTH�Û�Û�CSF S��B CFSÛ�TQSJOH� Û��CÛÛÛÛEÛÛ�T �GSPNÛ� Û�Ûu�� XPO��T VO��uVOÛÛÛÛÛÛ��0SWW�SJ�JVMGEXMZIW�ÛJOUFSWPDBMJDÛBOEÛBOUFDPOTPOBOUBMÛÛÛÛÛBÛJOUFSWPDBMJDÛ� �� Û�ÛZÛXIJDIÛJTÛMPTUÛCFUXFFOÛWPXFMTÛÛÛÛÛÛÛÛÛÛÛÛÛ�Q, � U S��IBZ<J>SÛÛÛÛÛÛÛÛÛÛÛÛÛ�L FUXPSFT�� F PS�� PSL Û�GPVS�ÛÛÛÛÛÛÛ5IJTÛEFWFMPQNFOUÛEJEÛOPUÛUBLFÛQMBDFÛBGUFSÛOPOTZMMBCJDÛSFTPOBOUT�ÛÛÛÛÛÛÛÛÛÛÛÛBXU Û�TMFFQJOHÛQMBDF� �FXU OÛ�TFWFO�ÛÛÛÛÛÛ� �� [FSPÛCFGPSFÛ� ÛÛÛÛÛÛÛÛÛÛÛÛÛ�Q, � U �CIJ��IB �CJ��IBSC�JOTUS�ÛTH�ÛÛÛÛÛCÛÛJOUFSWPDBMJDÛ�F ÛBOEÛ�GÛ�CFGPSFÛDPOTPOBOUÛGSPNÛ��CÛ�ÛXÛÛÛÛÛÛÛÛÛÛÛÛÛ�QSFQ��SFF ��SFG��ÛFSFXJN��BQQFBS�ÛÛÛÛÛÛÛÛÛÛÛÛÛHFO�ÛTH�Û�IB SP ��IBGSP ��IBXSÛÛÛÛÛÛ/FXMZÛBSJTJOHÛFX �PXÛEJEÛOPUÛNFSHFÛXJUIÛFBSMJFSÛFV �PVÛXIJDIÛCFDBNFÛPZ�ÛOFXÛPXÛ�ÛVÛÛÛÛÛÛÛÛÛ�TFQUN��FXU O��TFWFO�ÛÛÛÛÛÛÛÛÛ�, � F U ��PFUV��PX V��VU Û�FJHIU�ÛÛÛÛÛÛÛÛÛ�TXPQOPT�� PXO��L VOÛ�TMFFQ�ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ ��XÛCFGPSFÛGJOBMÛS�ÛÛÛÛÛÛÛÛB CJXSÛÛÛÛÛÛ5IJTÛ EFWFMPQNFOUÛ XBTÛ CMPDLFEÛ CZÛ UIFÛ QSFDFEJOHÛ SFTPOBOUÛ JOÛ�EF, � JX SÛ�Û�UBZ JSÛ�ÛUBZHSÛ�IVTCBOE�TÛCSPUIFS��ÛÛÛÛÛDÛÛ5IFÛEFMBCJBMJ[FEÛ��CÛ� ��[FSPÛCFUXFFOÛWPXFMT ÛBOEÛCFGPSFÛSÛ��CISF, � �VS�PTÛ�Û�CSF SPI��ÛB CFSÛÛÛÛÛÛÛÛÛÛÛÛÛÛ��0SWW�SJ�W]PPEFMG�VIWSRERXW��ÛÛÛÛÛÛÛÛÛÛÛÛÛÛÛ� �� BS �BM �BO �BNÛÛÛÛÛÛÛÛ*OÛUIJTÛXBZÛGJOBMÛOBTBMTÛBGUFSÛWPXFMÛCFDBNFÛQPTTJCMFÛBHBJO�ÛÛÛÛÛÛÛÛ4VSXLIXMG�I�ÛCFGPSFÛSÛBOEÛCFGPSFÛJOJUJBMÛDMVTUFST��ÛÛÛÛÛÛÛ��1IXEXLIWMWÛÛÛÛÛÛÛÛ*UÛJTÛBTTVNFEÛUIBUÛ�CISF, � U S��FCSBZS��F CBZS��CSPUIFS�ÛÛÛÛÛÛÛÛ,PSUMBOEUÛ BMTPÛ UBLFTÛ IFSFÛ UIFÛ UZQFÛ �, � FMJPT� �� BZMÛ �PUIFS�Û TFFÛ IJTÛSFNBSLÛPOÛTUBHFÛ��ÛÛ��%TSGSTI�MPTTÛPGÛGJOBMÛWPXFMTÛÛÛÛÛÛÛÛJÛPSÛVÛXBTÛQSFTFSWFEÛBTÛBOÛFQFOUIFUJDÛZÛ SFTQ�ÛXÛXIFOÛ UIFÛQSFDFEJOHÛWPXFMÛXBTÛB�ÛÛÛÛÛÛÛÛÛ�, � O S��BOJS��BZOS��BZSÛ�NBO�ÛÛÛÛÛÛÛÛÛBSUBXTSÛ�UFBS�Û�Û�ESB �V�Û


ÛÛÛÛ5IFÛMPTTÛPGÛ UIFÛGJOBMÛWPXFMÛXBTÛ MBUFSÛ UIBOÛUIFÛNFUBUIFTJTÛ JOÛ�, � FMJPT��BZM�ÛÛÛÛÛÛÛ��7MQTPMJMGEXMSR�SJ�GSRWSRERXEP�EVXMGYPEXMSR��ÛÛÛÛÛÛBÛÛ'JOBMÛ�� XÛÛÛÛÛÛÛ�OF, � �Û�Û�O ��OBX��TIJQ�ÛÛÛÛCÛÛOPXÛ��C�ÛÛÛÛDÛÛ'JOBMÛ�� Û�Û�� ÛPSÛ�� ÛXIJDIÛCFDBNFÛ�HÛJOÛ��ÛÛÛÛÛÛÛÛÛHFO�Û�BSF � ��BSFHÛ�TVO��ÛÛ��0SWW�SJ�PEFMEPMÊXMSR�ÛÛÛÛ5IJTÛBGGFDUTÛOPUÛPOMZÛ UIFÛPSJHJOBMÛ MBCJPWFMBST Û CVUÛ BMTPÛ UIPTFÛ UIBUÛBSPTFÛTFDPOEBSJMZ�Û�L ��U Û<�>ÛL F[ÛBOEÛ� Û��I Û<��C>ÛL PZS�ÛÛÛÛÛ��7LSVXIRMRK�SJ�JVMGEXMZIW�ÛÛÛÛÛÛÛÛ� �� U ��L ��HÛÛÛÛÛÛ"GUFSÛTÛUIFÛHÛXBTÛWPJDFMFTTÛJOÛTLFTVSÛ�NPUIFS�JO�MBX��ÛÛÛÛÛÛÛ/PUFÛUIBUÛ�FÛIBEÛEJTBQQFBSFEÛFBSMJFS�Û�� Û��CÛÛÛÛÛ��6MWI�SJ�RI[�\�ÛÛÛÛÛÛ5IFÛMBSZOHFBMTÛBGUFSÛTUPQÛ�Q) ��U) ��L)�CFDBNFÛBGGSJDBUFTÛPSÛBTQJSBUFEÛTUPQT�ÛUIFÛMBUUFSÛOPXÛCFDBNFÛY�ÛÛ�L)Û�Û�R Û��Û�ÛYÛ��ÛÛÛÛÛÛÛÛÛ�TL)��JO�TYBMJNÛÛÛÛÛÛ��-RJPY\�SJ�-VERMER�PSER[SVHW��ÛÛÛÛÛÛÛ��6IHYGXMSR�SJ�TVIXSRMG�ZS[IPW��ÛÛÛÛÛÛÛÛÛÛÛ ��ÛÛ&YBNQMFT�ÛÛ )FSFÛBSFÛTPNFÛFYBNQMFTÛXIFSFÛBMMÛ TUBHFTÛBSFÛHJWFOÛ TPNFÛIBWFÛOPUÛCFFOÛEBUFE�ÛÛÛÛ�USFJFT��USFJFI��Û�Û�USFFI�XIJDIÛ,PSUMBOEUÛOPXÛEBUFTÛUPÛTUBHFÛ� ÛQ�D�Û�Û� SFIÛ��Û�Û� SF Û��Û�Û�SF Û��CÛ�Û�FSF Û��Û�ÛFSFL Û��ÛÛÛÛ�TXPQOPTÛ�Û�IXPQOPIÛ�Û�Û�I PQOPI��DÛ�Û�I PFOPIÛ��Û�Û� PFOPIÛ��Û�Û� PGOPIÛ��CÛ�Û� PXOP�Û��CÛ�Û� VO�Û��Û�ÛL VOÛ��ÛÛÛÛ�L FUXPSFTÛ�Û�L FU PSFT��Û�Û�L FU PSFI��Û�Û� FU PSFIÛ�Û�Û FL PSFI��Û�Û

F PSFI� ��Û�Û� � F PSF � ��DÛ�Û�

� FPSF Û Û ��Û Û�Û � � PS � ��Û�Û� PS ��Û�Û� PSL ��Û�Û PSL �ÛÛÛÛÛ�Q, � U SÛÛ�Û�FB JS��Û�Û�IB JSÛ��BÛ�Û�IBZJSÛ��BÛ�ÛIBZSÛ��ÛÛÛÛ�Q, � USPTÛ �Û �QBUSPIÛ �Û �Û �FB SPIÛ ��Û �Û �FB SP Û ��DÛ �Û �IBGSP Û ��CÛ �Û�IBXSP Û��CÛ�ÛÛ�IBXS Û�� ÛBOBM�ÛIBXS�ÛÛÛ�TXFT SÛ �Û �IXFI SÛ �Û �Û �I FIVSÛ �Û �Û �I FVSÛ ��Û �Û � FVSÛ ��CÛ �Û � PZSÛ��Û�Û� PZSÛ��Û�ÛL PZSÛ��ÛÛÛÛ�TXF VS Û�Û�TXF VSBÛ�Û�Û� XF VSBÛ�Û�Û�TXFTVSBÛ�Û�Û�T FTVSB��BÛXJUIÛVOWPJDFEÛ ÛBGUFSÛTÛ�Û�T FTVSÛ��Û�Û�T FTVS��Û�ÛTLFTVS��ÛXJUIÛL ÛOPUÛL ÛBGUFSÛT�ÛÛÛÛ�, � F U ��Û�, � LU Û XJUIÛ[FSPÛHSBEFÛ GSPNÛUIFÛPSEJOBMÛ��Û �PQU � DG�Û&MFBOÛ(SFFLÛO�PTO��Û�PFU VÛ��Û�Û�PGU V��Û�Û�PXU VÛ��Û�Û�VU V��Û�Û�VU Û��Û�Û

VU Û��Û

212

REFERENCES Adjarian, Hracya H. 1909 &ODVVLILFDWLRQ�GHV�GLDOHFWHV�DUPpQLHQV. Paris: Champion. Agajan, È.B. 1960 O genezise armjanskogo konsonantizma. 9RSURV\� MD]\NR]QDQLMD

9/4, 37-52. Allen, William S. 1950 Notes on the phonetics of an Eastern Armenian speaker.

7UDQVDFWLRQV�RI�WKH�3KLORORJLFDO�6RFLHW\, 180-206. 1951 Phonetics and comparative linguistics. $UFKLYXP� /LQJXLVWLFXP

3/2, 126-136. Bailey, Harold W. 1987 Armenian KDULZU ‘hundred’. $$UP/ 8, 1-3. B OW FHDQX��0�)� 1980 Rapports entre l'arménien et le daco-mésien. In: 6WXGLD�

,QGR�(XURSDHD� DG� 'DFRURPDQRV� 3HUWLQHQWLD� ,,�� 6WXGLL� GH�WUDFRORJLH��OLQJYLVWLF . Bucureúti, 31-156.

Bauer-Manndorf, Elisabeth 1984 'DV�IU�KH�$UPHQLHQ. Wien: Mechitharisten Verlag. Beekes, Robert S.P. 1969 � 7KH� 'HYHORSPHQW� RI� WKH� 3URWR�,QGR�(XURSHDQ� /DU\QJHDOV� LQ�

*UHHN. The Hague: Mouton. 1981a The neuter plural and the vocalization of the laryngeals in

Avestan. ,QGR�,UDQLDQ�-RXUQDO 23/4, 275-287. 1981b The subjunctive endings of Indo-Iranian. ,QGR�,UDQLDQ� -RXUQDO

23/1, 21-27. 1985 7KH� RULJLQV� RI� WKH� ,QGR�(XURSHDQ� QRPLQDO� LQIOHFWLRQ. Innsbruck:

Institut für Sprachwissenschaft. 1988a $�*UDPPDU�RI�*DWKD�$YHVWDQ. Leiden: Brill. 1988b Laryngeal developments: A survey. In: A. Bammesberger (ed.)

'LH� /DU\QJDOWKHRULH� XQG� GLH� 5HNRQVWUXNWLRQ� GHV�LQGRJHUPDQLVFKHQ�/DXW��XQG�)RUPHQV\VWHPV. Heidelberg: Winter, 59-105.

Benveniste, Émile 1954 Problèmes sémantiques de la reconstruction. :RUG 10, 251-264. 1959 Sur la phonétique et la syntaxe de 1'arménien classique. %6/�54/1,

46-68.

5HIHUHQFHV 213

1965 Arménien DUHJDNQ «soleil» et la formation nominale en �DNQ. 5($UP 2, 5-19.

Bonfante, Giuliano 1942 The Armenian aorist. -$26 62, 102-105. B ga, K. 1922 .DOED�LU�VHQRY . I dalis. Kaunas: Švietimo Ministerijos leidinys. Bugge, Sophus 1893 Beiträge zur etymologischen erläuterung der armenischen sprache.

.= 32, 1-87. Burney, Ch. & Lang, D. Marshall 1971 7KH�3HRSOHV�RI�WKH�+LOOV. London: Leidenfeld and Nicolson. Çabej, Eqrem 1972 Über einige Lautregeln des Albanischen. 'LH� 6SUDFKH 18, 132-

154. ýHNPDQ��9DOHULM�1� 1974 O refleksax indoevropejskix * , * v balto-slavjanskom

jazykovom areale. %DOWR�VODYMDQVNLH� LVVOHGRYDQLMD. Moskva: Nauka, 116-135.

Chantraine, Pierre 1967 0RUSKRORJLH�KLVWRULTXH�GX�JUHF. Paris: Klincksieck. Clackson, James 1994 7KH�OLQJXLVWLF�UHODWLRQVKLS�EHWZHHQ�$UPHQLDQ�DQG�*UHHN. Oxford:

Blackwell. 2000 Arm. DULZQ�‘blood’. $$UP/ 20, 27–45. Considine, Patrick 1979 Discussion of Greppin 1973. 5($UP�13, 355-364. Djahukian ( ahukyan, Dåaukjan), Gevorg B. 1967 � 2þHUNL� SR� LVWRULL� GRSLVPHQQRJR� SHULRGD� DUPMDQVNRJR� MD]\ND.

Erevan. 1970 +D\HUHQ HY�KQGHYURSDNDQ�KLQ�OH]XQHU . Erevan. 1982 6UDYQLWHOQDMD� JUDPPDWLND� DUPMDQVNRJR� MD]\ND. Erevan: AN

ArmSSR. Dumézil, Georges 1938 Séries étymologiques arméniennes. %6/ 40/1, 48-54. 1964 1RWHV� VXU� OH� SDUOHU� GXQ� DUPpQLHQ� PXVXOPDQ� GH� +HPúLQ.

Bruxelles: Palais des Académies. 1979 Le Ciel et sa femme en arménien. In: )ORULOHJLXP� $QDWROLFXP��

0pODQJHV�RIIHUWV�j�(PPDQXHO�/DURFKH. Paris: Boccard, 107-110. Eichner, Heiner 1978 Die urindogermanische Wurzel *+ � UHX ‘hell machen’. 'LH�

6SUDFKH 24, 144-162.

214 5HIHUHQFHV

Frisk, Hjalmar 1966 .OHLQH� 6FKULIWHQ� ]XU� ,QGRJHUPDQLVWLN� XQG� ]XU� JULHFKLVFKHQ�

:RUWNXQGH. Göteborg: Almqvist & Wiksell. Garibjan, A.S. 1958 Novaja gruppa dialektov armjanskogo jazyka. 9RSURV\�

MD]\NR]QDQLMD 7/6, 95-101. 1959 Ob armjanskom konsonantizme. 9RSURV\� MD]\NR]QDQLMD 8/5, 81-

90. Georgiev, V.I. 1977 7UDNLWH� L� WHKQLMDW� H]LN. Sofija: Izdatelstvo na B lgarskata

Akademija na Naukite. Godel, Robert 1969 Les aoristes arméniens en -Fµ-. 6WXGLD� FODVVLFD� HW� RULHQWDOLD�

$QWRQLQR� 3DJOLDUR� REODWD. vol. 2. Roma: Istituto di Glottologia dell'Università di Roma, 253-258.

1970 Diachronic Armenian. In: &XUUHQW� 7UHQGV� LQ� /LQJXLVWLFV 6. The Hague: Mouton, 139-159.

1975 $Q� LQWURGXFWLRQ� WR� WKH� VWXG\� RI� &ODVVLFDO� $UPHQLDQ. Wiesbaden: Reichert.

1982 Une “loi phonétique” bien difficile à énoncer: *Z > arm. Z��Y� / J / zéro. 5($UP 16, 9-16.

Grammont, Maurice 1908 La métathèse en arménien. In: 0pODQJHV�GH�OLQJXLVWLTXH�RIIHUWV�j�

0��)HUGLQDQG�GH�6DXVVXUH. Paris: Champion, 229-243. 1918 Notes de phonétique générale. 06/ 20, 213-259. Greppin, John A.C. 1972 The Armenian reflexes of IE *Z and *\. 5($UP 9, 69-78. 1973 ,QLWLDO� YRZHO� DQG� DVSLUDWLRQ� LQ� &ODVVLFDO� $UPHQLDQ. Wien:

Mechitharisten-Buchdruckerei. 1975 Does IE *V9- give Arm. KY-? .= 89, 46-52. 1978 The sequence DZ& in Classical Armenian. .= 92, 282-293. 1982 Some comments on the chronology of secondary aspiration in

Classical Armenian, ms. 1993 The development of *7\� in Classical Armenian. 7KH� VHFRQG�

LQWHUQDWLRQDO� V\PSRVLXP� RQ� $UPHQLDQ� OLQJXLVWLFV�� 3URFHHGLQJV. Yerevan: Press of the Armenian Academy of Sciences, 15-24.

Grousset, René 1973 +LVWRLUH�GH�O$UPpQLH�GHV�RULJLQHV�j��. Paris: Payot. Hamp, Eric P. 1960 Palatal before resonant in Albanian. .= 76, 275-280. 1965 Evidence in Albanian. In: W. Winter (ed.) (YLGHQFH� IRU�

5HIHUHQFHV 215

ODU\QJHDOV. The Hague: Mouton, 123-141. 1966 The position of Albanian. In: H. Birnbaum & J. Puhvel (eds.)�

$QFLHQW� ,QGR�(XURSHDQ� GLDOHFWV. Berkeley: University of California Press, 97-121.

1972 Latin GDFULPD, ODFUXPD and Indo-European ‘tear’ . *ORWWD 50, 291-299.

1976 Armenian KDULZU again. .= 90, 128-130. Harkness, John 1996 Another approach to the “Great Armenian Puzzle”. $$UP/ 17, 11-

19. Hermann, E. 1907 Uber das Rekonstruieren. .= 41, 1-64. Hübschmann, Heinrich 1877 Ueber die stellung des armenischen im kreise der

indogermanischen sprachen. .= 23, 5-49. 1897 $UPHQLVFKH�*UDPPDWLN I. Leipzig: Breitkopf and Härtel. Huld, Martin E. 1978 Albanian L�WKHOOs and its eastern relations. .= 92, 297-299. 1979 Albanian YMHUU and IE. in-law terms. ,) 84, 196-200. 1984 %DVLF�$OEDQLDQ�HW\PRORJLHV. Columbus: Slavica. Illiþ-6YLW\þ��,OOLFK-Svitych), Vladislav M. 1963 ,PHQQDMD� DNFHQWXDFLMD� Y� EDOWLMVNRP� L� VODYMDQVNRP�� VXGED�

DNFHQWXDFLRQQ\[� SDUDGLJP. Moskva: Akademija Nauk SSSR, 1963. (Engl. transl.: 1RPLQDO� DFFHQWXDWLRQ� LQ� %DOWLF� DQG� 6ODYLF. Cambridge/Mass.: MIT Press, 1979).

Jasanoff, Jay H. 1979 Notes on the Armenian personal endings. .= 93, 133-149. Jensen, Hans 1959 $OWDUPHQLVFKH�*UDPPDWLN. Heidelberg: Winter. Job, Dieter Michael 1977 3UREOHPH� HLQHV� W\SRORJLVFKHQ� 9HUJOHLFKV� LEHURNDXNDVLVFKHU� XQG�

LQGRJHUPDQLVFKHU� 3KRQHPV\VWHPH� LP� .DXNDVXV. Frankfurt/M: Peter Lang.

1986 Les noms des parties du corps dans le vocabulaire de base arménien. /D� SODFH� GH� ODUPpQLHQ� GDQV� OHV� ODQJXHV� LQGR�HXURSpHQQHV. (= 0pPRLUHV� GH� O$FDGpPLH� 5R\DOH� GH� %HOJLTXH��&ODVVH� GHV� /HWWUHV�� )RQGV� 5HQp� 'UDJXHW, Tome III). Leuven: Peeters, 20–37.

1995 Zum Lautwandel im Armenischen: Probleme einer relativen Chronologie. $QDOHFWD�,QGRHXURSDHD�&UDFRYLHQVLD�,,��.XU\áRZLF]�0HPRULDO�9ROXPH�,. Cracow: Universitas, 291-311.

216 5HIHUHQFHV

Jokl, Norbert 1937 Ein Beitrag zur Lehre von der alb. Vertretung der idg.

Labiovelare. In: 0pODQJHV� OLQJXLVWLTXHV� RIIHUWV� j� 0�� +ROJHU�3HGHUVHQ. Aarhus: Universitetsforlaget; København: Levin & Munksgaard, 127-161.

1963 Die Verwandtschaftsverhältnisse des Albanischen zu den übrigen indogermanischen Sprachen. 'LH�6SUDFKH 9, 113-156.

Joseph, Brian D. 1984 Lindeman versus Kortlandt: summary and evaluation. $$UP/ 5,

45-50. Karaljunas [Karali QDV@��6LPDV 1966 K voprosu ob i.-e. V posle L, X v litovskom jazyke. %DOWLVWLFD 1/2,

113-126. Karst, Josef 1901 +LVWRULVFKH� *UDPPDWLN� GHV� .LOLNLVFK�$UPHQLVFKHQ. Strassburg:

Trübner. Khachaturian, Amalya 1983 The nature of voiced aspirated stops and affricates in Armenian

dialects. $$UP/ 4, 57-62. Klingenschmitt, Gert 1982 'DV�DOWDUPHQLVFKH�9HUEXP. Wiesbaden: Reichert. Kortlandt, Frederik H.H. 1972 0RGHOOLQJ� WKH�SKRQHPH��1HZ�WUHQGV� LQ�(DVW�(XURSHDQ�SKRQHPLF�

WKHRU\. The Hague: Mouton. 1975a A note on the Armenian palatalization. .=�89/1, 43-45. [This vol.,

10-12]. 1975b 6ODYLF�DFFHQWXDWLRQ��$� VWXG\� LQ� UHODWLYH� FKURQRORJ\. Lisse: Peter

de Ridder. 1976 Notes on Armenian historical phonology I. 6&DXF 3, 91-100. [This

vol., 1-9]. 1977 Historical laws of Baltic accentuation. %DOWLVWLFD 13/2, 319-330. 1978a Notes on Armenian historical phonology II: The second consonant

shift. 6&DXF 4, 9-16. [This vol., 20-25]. 1978b I.-E. palatovelars before resonants in Balto-Slavic. In: J. Fisiak

(ed.) 5HFHQW�'HYHORSPHQWV� LQ�+LVWRULFDO�3KRQRORJ\. The Hague: Mouton, 237-243.

1978c On the history of the genitive plural in Slavic, Baltic, Germanic, and Indo-European. /LQJXD 45, 281-300.

1979a Toward a reconstruction of the Balto-Slavic verbal system. /LQJXD 49, 51-70.

5HIHUHQFHV 217

1979b The Old Irish absolute and conjunct endings and questions of rel-ative chronology, eULX 30, 35-53.

1979c Proto-Indo-European obstruents. ,) 83, 107-118. 1980a Albanian and Armenian. .=�94, 243-251. [This vol., 13-19]. 1980b On the relative chronology of Armenian sound changes. )LUVW�

LQWHUQDWLRQDO� FRQIHUHQFH� RQ� $UPHQLDQ� OLQJXLVWLFV�� 3URFHHGLQJV. Delmar: Caravan Books, 97-106. [This vol., 26-33].

1980c + � R and R+ � . /LQJXD�3RVQDQLHQVLV 23, 127-128. 1981a 1st sg. middle *-+ � . ,) 86, 123-136. 1981b More evidence for Italo-Celtic, eULX 32, 1-22. 1981c On the Armenian personal endings. $$UP/ 2, 29-34. [This vol.,

34-38]. 1983a Demonstrative pronouns in Balto-Slavic, Armenian, and

Tocharian. 6WXGLHV� LQ� 6ODYLF� DQG� *HQHUDO� /LQJXLVWLFV 3: 'XWFK�FRQWULEXWLRQV� WR� WKH� �WK� LQWHUQDWLRQDO� FRQJUHVV� RI� VODYLVWV��/LQJXLVWLFV, 311-322. [This vol., 52-53].

1983b Notes on Armenian historical phonology III: K�. 6&DXF 5, 9-16. [This vol., 39-44].

1983c Greek numerals and PIE glottalic consonants. 066�42, 97-104. 1983d Naxahayereni holovakan ver avorut‘yunner . %DQEHU� (UHYDQL�

+DPDOVDUDQL�1/49, 64-70. 1984a Proto-Armenian case endings. ,QWHUQDWLRQDO� V\PSRVLXP� RQ�

$UPHQLDQ� OLQJXLVWLFV�� 5HSRUWV� >0HåGXQDURGQ\M� simpozium po armjanskomu jazykoznaniju: Doklady]. Erevan: AN Arm. SSR, 97-106. [This vol., 45-51].

1984b PIE. *+- in Armenian. $$UP/ 5, 41-43. [This vol., 54-56]. 1984c Old Irish subjunctives and futures and their Proto-Indo-European

origins. eULX 35, 179-187. 1985a Arm. DUWDZVU ‘tear’ . $$UP/ 6 [Fs. Djahukian], 59-61. [This vol.,

60-62]. 1985b Notes on Armenian historical phonology IV. 6&DXF 6, 9-11. [This

vol., 57-59]. 1985c The syncretism of nominative and accusative singular in

Armenian. 5($UP 19, 19-24. [This vol., 63-67]. 1985d Proto-Indo-European glottalic stops: The comparative evidence.

)ROLD�/LQJXLVWLFD�+LVWRULFD 6, 183-201. 1986a Armenian and Albanian. /D�SODFH�GH�ODUPpQLHQ�GDQV�OHV�ODQJXHV�

LQGR�HXURSpHQQHV. (= 0pPRLUHV� GH� O$FDGpPLH� 5R\DOH� GH�%HOJLTXH�� &ODVVH� GHV� /HWWUHV�� )RQGV� 5HQp� 'UDJXHW, Tome III). Leuven: Peeters, 38-47. [This vol., 68-74].

1986b The origin of the Slavic imperfect. In: R. Olesch & H. Rothe

218 5HIHUHQFHV

(eds.) )HVWVFKULIW� I�U�+HUEHUW� %UlXHU� ]XP� ��*HEXUWVWDJ. Köln: Böhlau, 253-258.

1987a Notes on Armenian historical phonology V. 6&DXF 7, 61-65. [This vol., 75-78].

1987b Sigmatic or root aorist? $$UP/ 8, 49-52. [This vol., 79-82]. 1987c PIE. *V in Albanian. 6WXGLHV�LQ�6ODYLF�DQG�*HQHUDO�/LQJXLVWLFV��

'XWFK�VWXGLHV�LQ�6RXWK�6ODYLF�DQG�%DONDQ�OLQJXLVWLFV, 219-226. 1988a The Thraco-Armenian consonant shift. %DONDQVNR� (]LNR]QDQLH /

/LQJXLVWLTXH�%DONDQLTXH�31, 71-74. [This vol., 83-87]. 1988b Proto-Germanic obstruents. $PVWHUGDPHU� %HLWUlJH� ]XU� lOWHUHQ�

*HUPDQLVWLN 27, 3-10. 1989 The making of a puzzle. $$UP/ 10, 43-52. [This vol., 88-95]. 1991 Arm. FDQDZWµ ‘known’ . $$UP/ 12, 1-4. [This vol., 96-97]. 1992a The Aeolic optative. In: R.S.P. Beekes et al. (eds.) 5HNRQVWUXNWLRQ�

XQG� UHODWLYH� &KURQRORJLH�� $NWHQ� GHU� �� )DFKWDJXQJ� GHU�,QGRJHUPDQLVFKHQ�*HVHOOVFKDIW [Leiden, 1987]. Innsbruck: Institut für Sprachwissenschaft, 235-239.

1992b The Germanic fifth class of strong verbs. 1RUWK�:HVWHUQ�(XURSHDQ�/DQJXDJH�(YROXWLRQ 19, 101-107.

1993 Intervocalic *-Z- in Armenian. $$UP/ 14, 9-13. [This vol., 102-103].

1994a Palatalization of dentals in Armenian. $$UP/ 15, 27-31. [This vol., 104-106].

1994b Proto-Armenian numerals. In: E. Rasmussen (ed.) ,Q� KRQRUHP�+ROJHU�3HGHUVHQ��.ROORTXLXP�GHU�,QGRJHUPDQLVFKHQ�*HVHOOVFKDIW [Kopenhagen, 1993], Wiesbaden: Reichert, 253-257. [This vol., 98-101].

1994c The fate of the sigmatic aorist in Tocharian. In: B. Schlerath (ed.) 7RFKDULVFK�� $NWHQ� GHU� )DFKWDJXQJ� GHU� ,QGRJHUPDQLVFKHQ�*HVHOOVFKDIW� [Berlin, September 1990]. (7RFKDULDQ� DQG�,QGR�(XURSHDQ�VWXGLHV� Supplementary series, vol. 4). Reykjavík: Málvísindastofnun Háskóla Íslands, 61-65.

1995a The sigmatic forms of the Armenian verb. $$UP/ 16, 13-17. [This vol., 107-109].

1995b Lithuanian verbs in -DXWL and -XRWL. /LQJXLVWLFD�%DOWLFD 4, 141-143. 1996a Arm. FDáU ‘laughter’ . $$UP/ 17, 55-59. [This vol., 117-119]. 1996b The Proto-Armenian verbal system. 3URFHHGLQJV� RI� WKH� �WK�

LQWHUQDWLRQDO� FRQIHUHQFH� RQ�$UPHQLDQ� OLQJXLVWLFV. Delmar, N.Y.: Caravan Books, 35-43. [This vol., 110-116].

1997 Arm. Q U ‘sister-in-law’ . $$UP/ 18, 7-9. [This vol., 120-121]. 1998a The development of *\- in Armenian. $$UP/ 19, 15-18. [This

5HIHUHQFHV 219

vol., 122-124]. 1998b Arm. JRP ‘am’ . $$UP/ 19, 19-20. [This vol., 125]. 1998c Armenian glottalization revisited. $$UP/ 19, 11-14. [This vol.,

126-128]. 1999 The Armenian causative. $$UP/ 20, 47-49. [This vol., 129-130]. 2001 Arm. DULZQ�‘blood’ . $$UP/ 21, 11-12. [This vol., 131-132]. 2003 An Indo-European substratum in Slavic? In: A. Bammesberger

(ed.) /DQJXDJHV� LQ�3UHKLVWRULF�(XURSH� Heidelberg: Winter, 253-260.

2004 Initial laryngeals in Anatolian. 2USKHXV 13-14, 9-12. Kuipers, Aert H. 1974 7KH� 6KXVZDS� ODQJXDJH�� JUDPPDU�� WH[WV�� GLFWLRQDU\. The Hague:

Mouton. Lamberterie, Charles de 1978 Armeniaca I-VIII: études lexicales. %6/ 73/1, 243-285. 1979 Le signe du pluriel en arménien classique. %6/ 74/1, 319-332. 1998 Sur la “Loi de Meillet”. &RPSWHV� UHQGXV� GH� O¶$FDGpPLH� GHV�

LQVFULSWLRQV��EHOOHV�OHWWUHV 1998, 881-903. Lehmann, Winfred P. 1951 The distribution of Proto-Indo-European /r/. /DQJXDJH 27/1,

13-17. 1965 Germanic evidence. In: W. Winter (ed.) (YLGHQFH� IRU� ODU\QJHDOV.

The Hague: Mouton, 212-223. Lejeune, Michel 1979 Regards sur les sonores i.e. en vieux phrygien. In: )ORULOHJLXP�

$QDWROLFXP�� 0pODQJHV� RIIHUWV� j� (PPDQXHO� /DURFKH. Paris: Boccard, 219-224.

Leskien, August 1962 +DQGEXFK� GHU� DOWEXOJDULVFKHQ� �DOWNLUFKHQVODYLVFKHQ�� 6SUDFKH��

*UDPPDWLN��7H[WH��*ORVVDU. Heidelberg: Winter. Lidén, Evald 1906 Baumnamen und Verwandtes. ,) 18, 485-509. Lindeman, Fredrik Otto 1982 7KH� WULSOH� UHSUHVHQWDWLRQ� RI� 6FKZD� LQ� *UHHN� DQG� VRPH� UHODWHG�

SUREOHPV�RI�,QGR�(XURSHDQ�SKRQRORJ\. Oslo: Universitetsforlaget. Lubotsky, Alexander 1981 Gr. P�GNUMI : Skt. SDMUi- and loss of laryngeals before mediae in

Indo-Iranian. 066 40, 133-138. 1988 The Old Phrygian Areyastis-inscription. .DGPRV 27, 9-26. Makaev, Ènver A. 1971 Problema indoiranskogo jazykovogo edinstva. 9RSURV\�

220 5HIHUHQFHV

MD]\NR]QDQLMD 3, 7-20. Mawet, Francine 1986 Les développements fonctionnels de arménien �H�WµH. /D�SODFH�GH�

ODUPpQLHQ� GDQV� OHV� ODQJXHV� LQGR�HXURSpHQQHV. (= 0pPRLUHV� GH�O$FDGpPLH� 5R\DOH� GH� %HOJLTXH�� &ODVVH� GHV� /HWWUHV�� )RQGV� 5HQp�'UDJXHW, Tome III). Leuven: Peeters, 76-89.

Meillet, Antoine 1894a Notes arméniennes. 06/ 8, 153-165. 1894b De quelques difficultés de la théorie des gutturales indo-

européennes. 06/ 8, 277-304. 1894c Varia. 06/ 8, 235-245. 1896 Varia. 06/ 9, 149-159. 1900a Étymologies arméniennes. 06/ 11, 390-401. 1900b Recherches sur la syntaxe comparée de l'arménien. II. Les règles

d'accord de l'adjectif. 06/ 11, 369-389. [= 1962: 39-59]. 1909 Armeniaca. 06/ 15, 555-557. 1911 Recherches sur la syntaxe comparée de l'arménien. V. Emploi des

formes du pluriel des substantifs. 06/ 17, 1-35. [= 1962: 123-157].

1913 $OWDUPHQLVFKHV� (OHPHQWDUEXFK. Heidelberg: Winter. (Reprint 1980).

1921 Arménien \DJ. %6/�22, 20. 1922 /HV�GLDOHFWHV�LQGR�HXURSpHQV. Paris: Champion. 1936 (VTXLVVH� GXQH� JUDPPDLUH� FRPSDUpH� GH� ODUPpQLHQ� FODVVLTXH.

Vienne: Mékhitharistes. 1962 eWXGHV�GH�OLQJXLVWLTXH�HW�GH�SKLORORJLH�DUPpQLHQQHV I: Recherches

sur la syntaxe comparée de l'arménien, suivies de la composition en arménien. Lisboa: Imprensa Nacional.

1977 eWXGHV�GH� OLQJXLVWLTXH� HW� GH�SKLORORJLH�DUPpQLHQQHV II. Louvain: Imprimerie Orientaliste.

Meyer, Gustav 1891 (W\PRORJLVFKHV� :|UWHUEXFK� GHU� DOEDQHVLVFKHQ� 6SUDFKH.

Straßburg: Trübner. 1892 Albanesische Studien III: Lautlehre der indogermanischen

Bestandtheile des Albanesischen. 6LW]XQJVEHULFKWH� GHU�SKLORVRSKLVFK�KLVWRULVFKHQ�&ODVVH�GHU�.DLVHUOLFKHQ�$NDGHPLH�GHU�:LVVHQVFKDIWHQ (Wien) 125/11.

Muller, G. 1984a The development of vocalized laryngeals in Old Armenian.

$3,/.8 4, 95-98. 1984b Notes on four Old Armenian etymologies. $3,/.8 4, 99-101.

5HIHUHQFHV 221

Oettinger, Norbert 1979 'LH� 6WDPPELOGXQJ� GHV� KHWKLWLVFKHQ� 9HUEXPV. Nürnberg: Hans

Carl. Ölberg, Hermann M. 1968 Idg. vor X im Albanischen. 6WXGLHQ�]XU�6SUDFKZLVVHQVFKDIW�XQG�

.XOWXUNXQGH��*HGHQNVFKULIW�I�U�:LOKHOP�%UDQGHQVWHLQ. Innsbruck: Institut für vergleichende Sprachwissenschaft, 109-118.

1969 Zur Neutralisierung von palatalem und velarem N vor Liguida. 6WXGL� OLQJXLVWLFL� LQ� RQRUH� GL� 9LWWRUH� 3LVDQL II. Brescia: Paideia, 683-690.

1972 Einige Überlegungen zur Laryngaltheorie (An Hand des Albanischen). .= 86, 121-136.

1976 Zwei oder drei Gutturalreihen? Vom Albanischen aus gesehen. In: 6FULWWL�LQ�RQRUH�GL�*LXOLDQR�%RQIDQWH II. Brescia: Paideia, 561-570.

Olsen, Birgit A. 1984 On the development of Indo-European prothetic vowels in

Classical Armenian. $3,/.8 4, 103-118. [reprinted in: 5($UP 19 (1985), 5-17].

1986 The Armenian continuations of Indo-European intervocalic *Z. $$UP/ 7, 51-56.

1988 The formation of the subjunctive paradigm in Classical Armenian. $$UP/ 9, 7-14.

1993 On the origins of Armenian Fµ. 7KH� VHFRQG� LQWHUQDWLRQDO�V\PSRVLXP� RQ� $UPHQLDQ� OLQJXLVWLFV�� 3URFHHGLQJV. Yerevan, 25-48.

1999 7KH�1RXQ� LQ�%LEOLFDO�$UPHQLDQ. Berlin / New York: Mouton de Gruyter.

Pedersen, Holger 1895 Das indogermanische V im Slavischen. ,) 5, 33-87. 1900a Die gutturale im Albanesischen. .=�36, 277-340. 1900b Albanesisch und Armenisch. .= 36, 340-341. [= Kl.S., 1-2]. 1900c Wie viel laute gab es im Indogermanischen? .= 36, 74-110. 1905a Zur armenischen Sprachgeschichte. .= 38, 194-240. [= Kl.S., 56-

240]. 1905b Les pronoms démonstratifs de l'ancien arménien. 'HW�.RQJHOLJH�

'DQVNH�9LGHQVNDEHUQHV�6HOVNDEV�6NULIWHU� Sjette Række. +LVWRULVN�RJ�ILORVRILVN�$IGHOLQJ VI. 3. [= Kl.S., 8-55].

1906 Armenisch und die nachbarsprachen. .= 39, 334-485. [= Kl.S., 112-263].

1938 +LWWLWLVFK� XQG� GLH� DQGHUHQ� LQGRHXURSlLVFKHQ� 6SUDFKHQ. København: Levin & Munksgaard.

222 5HIHUHQFHV

1982 [Kl.S.] .OHLQH� 6FKULIWHQ� ]XP� $UPHQLVFKHQ, herausgegeben von Rüdiger Schmitt. Hildesheim: Olms, 1982.

Perikhanian, Anahit G. 1966 8QH� LQVFULSWLRQ�DUDPpHQQH�GX� URL�$UWDã V� WURXYpH�à Zanguézour.

5($UP 3, 17-29. Peters, Martin 1980 8QWHUVXFKXQJHQ�]XU�9HUWUHWXQJ�GHU�LQGRJHUPDQLVFKHQ�/DU\QJDOH�

LP� *ULHFKLVFKHQ. Wien: Verlag der Österreichischen Akademie der Wissenschaften.

Pinault, Georges-Jean 1997 Remarque sur le pluriel tokh. B DNU QD, A NUXQW. Lubotsky, A.

(ed.) 6RXQG� ODZ� DQG� DQDORJ\�� 3DSHUV� LQ� KRQRU� RI� 5REHUW� 6�3��%HHNHV�RQ�WKH�RFFDVLRQ�RI�KLV��WK�ELUWKGD\. Amsterdam: Rodopi, 219-233.

Pisani, Vittore 1934 Armeniaca. .= 61, l80-189. 1944 Armenische Studien. .= 68, 157-177. 1950 Studi sulla fonetica dell'armeno. 5LFHUFKH�/LQJXLVWLFKH 1, 165-193. Pisowicz, Andrzej 1976a /H�GpYHORSSHPHQW�GX�FRQVRQDQWLVPH�DUPpQLHQ��:URFáDZ��=DNáDG�

,PLHQLD�2VROL VNLFK�� Wydawnictwo Polskiej Akademii Nauk. 1976b Matériaux pour servir à la recherche du consonantisme arménien.

)ROLD�2ULHQWDOLD 17, 197-216. 1997 Consonant shifts in Armenian dialects during the post-classical

period revisited. In: $UPHQLDQ� SHUVSHFWLYHV. Richmond, Surrey: Curzon, 215-230.

Pokorny, Julius 1959 Indogermanisches etymologisches Wörterbuch. I. Bern: Francke. Polomé, Edgar C. 1980 Armenian and the Proto-Indo-European laryngeals. )LUVW�

LQWHUQDWLRQDO� FRQIHUHQFH� RQ� $UPHQLDQ� OLQJXLVWLFV�� 3URFHHGLQJV. Delmar: Caravan Books, 17-33.

Price, Glanville 1971 7KH�)UHQFK�ODQJXDJH��3UHVHQW�DQG�SDVW. London: Edward Arnold. Ruijgh, Cornelis J. 1985 Prob1èmes de philologie mycénienne. 0LQRV, N.S. 19, 105-167. 1992 SR�NX�WD et SR�NX�WH�UR, dérivés de *SRNX ‘petit bétail’ . 0\NHQDwND

(%XOOHWLQ�GH�FRUUHVSRQGDQFH�KHOOpQLTXH, supplément XXV, École Française d'Athènes), 543-562.

Saussure, Ferdinand de 1892a [Sanscrit WK]. %6/ 7 (1888-1892), cxviii. [= Recueil, 603].

5HIHUHQFHV 223

1892b Varia. 06/ 7, 73-93. [= Recueil, 435-463]. 1922 5HFXHLO�GHV�SXEOLFDWLRQV�VFLHQWLILTXHV. Genève: Société Anonyme

des Éditions Sonor. Schindler, Jochem 1975 Armenisch HUNQ, griechisch ± E��, irisch LGX. .= 89/1, 53-65. Schmitt, Rüdiger 1972/74 Die Erforschung des Klassisch-Armenischen seit Meillet (1936).

.UDW\ORV�17, 1-68. 1975 Von Bopp bis Hübschmann: das Armenische als indogermanische

Sprache. .= 89, 3-30. 1981 *UDPPDWLN�GHV�.ODVVLVFK�$UPHQLVFKHQ�PLW�VSUDFKYHUJOHLFKHQGHQ�

(UOlXWHUXQJHQ. Innsbruck: Institut für vergleichende Sprach-wissenschaft. (,QQVEUXFNHU�%HLWUlJH�]XU�6SUDFKZLVVHQVFKDIW; 32).

1996 Some remarks on Armenian Q U ‘sister-in-law, brother's wife’ . $$UP/ 17, 21-24.

Schrijver, Peter 1991 7KH� UHIOH[HV� RI� WKH� 3URWR�,QGR�(XURSHDQ� ODU\QJHDOV� LQ� /DWLQ.

Amsterdam: Rodopi. Shevelov, George Y. 1964 $� SUHKLVWRU\� RI� 6ODYLF�� 7KH� KLVWRULFDO� SKRQRORJ\� RI� &RPPRQ�

6ODYLF. Heidelberg: Winter. Skautrup, Peter 1928 Klusiler og “yngre” stød i vestjysk. $FWD� 3KLORORJLFD�

6FDQGLQDYLFD 3, 32-51. Solta, Georg R. 1960 'LH� 6WHOOXQJ� GHV� $UPHQLVFKHQ� LP� .UHLVH� GHU� LQGRJHUPDQLVFKHQ�

6SUDFKHQ. Wien: Mechitharisten-Buchdruckerei. 1965 Palatalisierung und Labialisierung. ,) 70, 276-315. Steensland, Lars 1973 'LH�'LVWULEXWLRQ�GHU�XULQGRJHUPDQLVFKHQ�VRJHQDQQWHQ�*XWWXUDOH.

Stockholms: Almqvist & Wiksell. ($FWD�8QLYHUVLWDWLV�8SVDOLHQVLV��6WXGLD�6ODYLFD; 12).

Stempel, Reinhard 1994 Final IE *�V and the Old Armenian plural marker �Nµ. $$UP/ 15,

1-19. Szemerényi, Oswald 1960 6WXGLHV� LQ� WKH� ,QGR�(XURSHDQ� V\VWHP� RI� QXPHUDOV. Heidelberg:

Winter. Thurneysen, Rudolf 1946 $�*UDPPDU� RI�2OG� ,ULVK. Dublin: Dublin Institute for Advanced

Studies.

224 5HIHUHQFHV

Trautmann, Reinhold 1923 %DOWLVFK�6ODYLVFKHV� :|UWHUEXFK. Göttingen: Vandenhoeck &

Ruprecht. Trubetzkoy, Nikolaj S. 1930 Über die Entstehung der gemeinwestslavischen

Eigentümlichkeiten auf dem Gebiete des Konsonantismus. =HLWVFKULIW�I�U�VODYLVFKH�3KLORORJLH 7, 383-406.

Vaillant, André 1966 *UDPPDLUH� FRPSDUpH� GHV� ODQJXHV� VODYHV� ,,,�� /H� YHUEH. Paris:

Klincksieck. Van Windekens, Albert J. 1976 Encore le terme ‘larme’ en indo-européen. .= 90, 12-17. Vasmer, Max 1958 5XVVLVFKHV� HW\PRORJLVFKHV� :|UWHUEXFK. 3. Band. Heidelberg:

Winter. Vennemann, Theo 1986 Syllable-based sound changes in early Armenian. $$UP/ 7,

27-43. Viredaz, Rémy 2000 NµHUE, MHUE, [Gr.] 3J')?. +LVWRULVFKH� 6SUDFKIRUVFKXQJ 113, 290-

307. Vogt, Hans 1938 Arménien et caucasique du sud. 176 9, 321-338. 1958 Les occlusives de l'arménien. 176 18, 143-161. Weitenberg, Joseph J.S. 1975 Armenisch RUWµ ‘Weinstock, Rebe’ , griechisch %-²'K#+ und

hethitisch SDUãGX-. .= 89, 66-75. 1982 Remarks on the pronominal genitive in Classical Armenian.

5($UP 17, 113-121. 1984 The inflexion of PL ‘one’ as a trace of Proto-Armenian nominal

gender. ,QWHUQDWLRQDO� V\PSRVLXP� RQ� $UPHQLDQ� OLQJXLVWLFV��5HSRUWV. Erevan: AN Arm. SSR, 195-218.

1985 Additional -Q in Armenian. $$UP/ 6, 101-106. 1986 Additional K�, initial \�� and Indo-European *\�� in Armenian. /D�

SODFH� GH� ODUPpQLHQ� GDQV� OHV� ODQJXHV� LQGR�HXURSpHQQHV. (= 0pPRLUHV�GH� O$FDGpPLH�5R\DOH�GH�%HOJLTXH��&ODVVH�GHV�/HWWUHV��)RQGV�5HQp�'UDJXHW, Tome III). Leuven: Peeters, 90-101.

1997 The prepositional group L� \�� and the orthography of Gospel manuscript M (Matenadaran 6200). $$UP/ 18, 39-50.

Winter, Werner 1962 Problems of Armenian Phonology III. /DQJXDJH 38, 254-262.

5HIHUHQFHV 225

1965 Armenian Evidence. In: W. Winter (ed.) (YLGHQFH�IRU�ODU\QJHDOV. The Hague: Mouton, 100-115.

1966 Traces of early dialectal diversity in Old Armenian. In: H. Birnbaum & J. Puhvel (eds.)� $QFLHQW� ,QGR�(XURSHDQ� GLDOHFWV. Berkeley: University of California Press, 201-211.

1975 Die Personalendungen des Imperfekts und Aorists im Armenischen. .= 89, 110-122.

1992 $UPHQLDQ�� ,Q�� -�� *YR]GDQRYLü� �HG�� ,QGR�(XURSHDQ� QXPHUDOV. Berlin: Mouton de Gruyter, 347-359.

Zabrocki, Ludwik 1951 8VLOQLHQLH�L� OHQLFMD�Z�M ]\NDFK�LQGRHXURSHMVNLFK�L Z XJURIL VNLP.

3R]QD : Naká�� 3R]QD VNLHgo Tow.� 3U]\MDFLyá� Nauk. (3UDFH�NRPLVML�ILORORJLF]QHM; 13/3).

%LEOLRJUDSKLFDO�DEEUHYLDWLRQV $$UP/�� - $QQXDO�RI�$UPHQLDQ�/LQJXLVWLFV. $3,/.8 - Arbejdspapirer udsendt af Institut for Lingvistik, Københavns

Universitet. %6/ - %XOOHWLQ�GH�OD�6RFLpWp�GH�/LQJXLVWLTXH�GH�3DULV. ,) - ,QGRJHUPDQLVFKH�)RUVFKXQJHQ� -$26 - -RXUQDO�RI�WKH�$PHULFDQ�2ULHQWDO�6RFLHW\. Kl.S. - Kleine Schriften. .= - =HLWVFKULIW�I�U�YHUJOHLFKHQGH�6SUDFKIRUVFKXQJ�DXI�GHP�*HELHWH�

GHU�LQGRJHUPDQLVFKHQ�6SUDFKHQ��.XKQV�=HLWVFKULIW�. 06/ - 0pPRLUHV�GH�OD�6RFLpWp�GH�/LQJXLVWLTXH�GH�3DULV. 066 - 0�QFKHQHU�6WXGLHQ�]XU�6SUDFKZLVVHQVFKDIW. 176 - 1RUVN�7LGVVNULIW�IRU�6SURJYLGHQVNDS. 5($UP - 5HYXH�GHV�eWXGHV�$UPpQLHQQHV (n.s.). 6&DXF - 6WXGLD�&DXFDVLFD.

Documents

Kortlandt Beekes Armeniaca