Levin Slew on Tinsy n These 1980

7/28/2019 Levin Slew on Tinsy n These 1980

1/32

RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI N

D I A LEC T I C S A N D R ED U C TI O N I S M I N EC O LO G Y

The philosophical debates which have accompanied the developmentof science have often been expressed in terms of dichotomouschoices between opposing viewpoints about the structure of nature,the explanation of natural processes, and the appropriate methods forresearch:

Are the different levels of organization such as atom, molecule,cell, organism, species, and community only the epiphenomena ofunderlying physical principles or are the levels separated by realdiscontinuites? Are the objects within a level fundamentally similardespite apparent differences or is each one unique despite seemingsimilarities? Is the natural world more or less at equilbrium or inconstant change? Can things be explained by present circumstancesor is the present simply a reflection of the past? Is the world causal orrandom? Do things happen to a system mostly because of its owninternal dynamic or is causat ion external? Is it legitimate to postulatehypothetica l entities as part of scientific explanation or should sciencestick to observables? Do generalizations reveal deeper levels ofreality or destroy the richness of nature? Are abstractions meaningfulor obfuscatory?

As long as the alternatives are accepted as mutually exclusive, theconflict remains one between mechanistic reductionism championingmaterialism, and idealism representing holistic and sometimes dialec-tical concerns.

It is also possible to opt for compromise in the form of a liberalpluralism in which the questions become quantitative: how differentand how similar are objects? What is the relative importance ofchance and necessity? Of internal and external causes (e.g., heredityand environment)? Such an approach reduces the philosophical issuesto a partitioning of variance and must remain agnostic about strategy.

When we attempt to chose sides restrospectively, we find that it isSynthese 43 (1980) 47-78. 0039-7857/80/0431-0047 $03.20.Copyright (~ 1980by D. Reidel Publishing Co., Dordrecht, Holland, and Boston, U.S.A.


2/32

48 RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI N

not possible to be consistent: we side with the biologists who opposedtheological idealism in insisting upon the continuity between ourspecies and other animals or between living and nonliving matter. Butwe emphasize the discontinuity between human society and animalgroups in opposition to various 'biology is destiny' schools.

As long as we accept the terms of the debate between reductionismand idealism, we must adopt an uncomfortably ad hoc inconsistencyas we see now one side, now the other advancing or holding backscience. Simberloff's essay (1980) seems to us to embody the falsedebate by being based on three fundamental common confusions.These are: the confusion between reductionism and materialism, theconfusion between idealism and abstraction, and the confusion be-tween statistical and stochastic. As a result of these confusions,Simberloff, in his attemmpt to escape from the obscurantist holism ofClements' 'superorganism,' falls into the pit of obscurantist stochas-ticity and indeterminism. For if one commits oneself to a totallyreductionist program, claiming that in fact collections of objects innature do not have properties aside from the properties of theseobjects themselves, then failures of explanation must be attributedultimately to an inherent indeterminism in the behavior of the objectsthemselves. The reductionist program thus simply changes the locusof mystification from mysterious properties of wholes to mysteriousproperties of parts.

We will discuss these three confusions, and some subsidiary ones,in order to outline our disagreements with Simberloff, but also todevelop implicitly a Marxist approach to the questions that have beenraised. Dialectic materialism enters the natural sciences as the simul-taneous negation of both mechanistic materialism and dialecticalidealism, as a rejection of the terms of the debate. Its central thesesare that nature is contradictory, that there is unity and inter-penetration of the seemingly mutually exclusive, and that thereforethe main issue for science is the study of that unity and contradictionrather than their separation either to reject one or to assign relativeimportance.


3/32

D I A L E C T I C S A N D R E D U C T I O N I S M 49R E D U C T I O N I S M A N D M A T E R I A L I S M

The confusion between reductionism and materialism has plaguedbiology since Descartes' invention of the organism as a machine.Despite the repeated demonstrations in philosophy of the errors ofvulgar reductionism, practicing biologists continue to see the ultimateobjective of the study of living organisms to be a description ofphenomena entirely in terms of individual properties of isolatedobjects. A recent avatar is Wilson's (1978) claim that a scientificmaterialist explanation of human society and culture must be in termsof human genetic evolution and the Darwinian fitness of individuals.

In ecology, reductionism takes the form of regarding each speciesas a separate element existing in an environment that consists of thephysical world and of other species. The interaction of the speciesand its environment is unidirectional: the species experiences, reactsto, and evolves in response to its environment. The reciprocalphenomenon, the reaction and evolution of the environment in res-ponse to the species, is put aside. While it is obvious that predatorand prey play both the roles of 'organism' and 'environment,' it isoften forgotten that the seedling is the 'environment' of the soil inthat the soil undergoes lasting evolutionary changes of great mag-nitude as a direct consequence of the activity of the plants growing init, and in turn feeds back on the conditions of existence of theorganisms. But if two species are evolving in mutual response to eachother or if plant and soil are mutually changing the conditions of eachother's existence, then the ensemble of species or of species andphysical environment, is an object with dynamica l laws that can onlybe expressed in a space of appropriate dimensionality. The change ofany one element can be followed as a projection on a single dimen-sion of the changes of the n-tuple, but this projection may showparadoxical features including apparent lack of causality, while theentire ensemble changes in a perfectly regular manner. For example,a prey and predator will approach an equilibrium of numbers by aspiral path in the two dimensional space whose axes are the abun-dances of the two species. This path is completely unambiguous in


4/32

50 R I C H A R D L E V I N S A N D R I C H A R D L E W O N T I Nthe sense that given the location of a point in two dimensional spaceat one instant of time, a unique vector of change can be establishedpredicting its position at the next instant. Each of the two componentspecies, however, is oscillating in abundance so that given only theabundance of the predator, say, its impossible to know whether it willincrease or decrease during the next interval. The description ofchange of the n-dimensional object may then itself be collapsed ontosome new dimension, for example, dis tance from the equilibrium point,which again may behave in a simple monotonic and predictable way.The rule of behavior of the new object is not an obscurantist holsimbut a rule of the evolution of a composite entity that is appropriate tothat level of description and not to others. In the specific case justIgiven, neither the prey not the predator abundances converge mono-Itonically to their final equilibria, and the monotone behavior of thepair object is not predictable from the separate equations of eachspecies. Moreover, the separate behavior of each species is not itselfpredictable from the form of their separate equations of motion, sinceneither of these equations is intrinsically oscillatory and the dampedoscillation of the two species is a consequence of their dynamiccoupling.

The Clementsian superorganism paradigm is indeed idealist. Itscommunity is the expression of some general organizing principle,some balance or harmony of nature. The behavior of the parts iswholely subordinated to this abstract principle, which causes thecommunity to develop toward the maximization of efficiency,productivity, stability, or some other civic virtue. Therefore, a majorpriority would be to find out what does a community maximize.

Having correctly identified the Clementsian superorganism asidealist, Simberloff then lumps with it all form of 'systems modelling.'But the large-scale computer models of systems ecology do not fitunder the heading of 'holism' at all. Rather they are forms oflarge-scale reductionism: the objects of study are the naively given'parts'-abundances or biomasses of populations. No new objects ofstudy arise at the co mmunity level. The research is usually conductedon a single system, a lake, forest, or prairie, and the results are


5/32

D I A L E C T I CS A N D R E D U C T I O N I SM 51measurements and projections for the same lake, forest, or prairiewithout attempting to find properties of lakes, forests, or prairies ingeneral. It requires vast amounts of data for its simulations, and muchof the scientific effort goes into problems of estimation. We agreewith Simberloff that this approach has been generously supported andsingularly unproductive.

Idealism and reductionism in ecology share a common fault: theysee 'true causes' as arising at one level only with other levels havingonly epistemological but not ontological validity. Clementsian ideal-ism saw the community as the only causal reality with the behaviorsof individual species populations as the direct consequence of themysterious organizing forces of the community. One might describethe community for some purpose by a list of species abundances, butthat description was of epiphenomena only. Reductionism, on theother hand, sees the individual species, or ultimately the individuals(or cells, or molecules, for there is no clear stopping place in thereductionist program), as the only 'real' objects while higher levelsare again descriptions of convenience without causal reality. A propermaterialism, however, accepts neither of these doctrinaire positionsbut looks for the actual material relationship among entities at alllevels. The number of barn owls and the number of house miceseparately are important causal factors for the abundance of theirrespective competitors and are material realities relevant to thoseother species, but the particular combination of abundances of owlsand mice is a new object which is a material cause of the volume ofowl pellets and therefore of the abundance of habitat for certainbacteria.

T H E C O M M U N I T Y A S A D I A L E C T I C A L W H O L E

Unlike the idealist holism which sees the whole as the embodiment ofsome ideal organizing principle, dialectical materialism views the wholeas a contingent structure in reciprocal interaction with its own parts andwith the greater whole of which it is a part. Whole and part do notcompletely determine each other.


6/32

52 R I C H A R D L E V I N S A N D R I C H A R D L E W O N T I NThe community in ecological theory is an intermediate entity,between the local species population and biogeographic region, the

locus of species interactions.The region can be visualized as a pat chwork of environments and a

continum of environmental gradients over which populations aredistributed. A local community is linked to the region by thedynamics of local extinction and colonization. Local extinctiondepends on local conditions affecting the populations in question.Colonization depends on the number of propagules (seeds, eggs,young animals) the local population sends out, which depends on thepopulation size achieved locally. It also depends on the behavior ofthese propagules, their ability to cross the gaps between suitablehabitats, their tolerance of conditions along the way, and theircapacity to establish themselves (anchor on the the new substrate, growunder the shade of established trees, defend an incipient ant nest).These properties are biological characteristics of the individual spe-cies which are not directly responsible for abundance and survival inthe local community. Finally, colonization depends on the pattern ofthe environmental mosaic - the distance between patches and whetherthe patches are large or small, the structure of the gradients (whetherdifferent kinds of favorable conditions are positively or negativelyassociated). These biogeographic properties are not implicit in thedynamics of the local set of species.

The whole ensemble of species of a region depends on the origin ofthe biota, the extinction of species in the whole region, and theprocesses of speciation.

Therefore, the biogeographic level gives us a dynamic of extin-ction, colonization, and speciation in which the parameters of migra-tion and extinc ton are givens, partly dependent on local dynamics butnot contained therein.

Below the community are the componen t species populations. Theyenter the community at a rate which depends on their abundance inother communities, in the region as a whole. But once in the localitytheir abundance, persistence, variability and sensitivity to environ-mental variability depends on their interactions with other species andon the parameters of their eco lo gy - birth rate, food and microhabitat


7/32

D I A L E C T I C S A N D R E D U C T I O N I S M 53preferences, mobility, vulnerability to predators, physiologicaltolerances which come from their own genetic makeup. The geneticmakeup in turn is a consequence of the processes of selection,mutation, drift, and gene exchange with other populations of the samespecies, which form the domain of population genetics and reflectpast evolutionary history. The other members of the communityaffect the direction of natural selection within the community andtherefo re influence these parameters, but they are not deducible fromthe general rules of community ecology.

Thus the claim that the ecological communi ty is a meaningful wholerests on its having distinct dynamics-the local demographic inter-actions of species against a background of biogeographic and popu-lation genetic parameters.

From this point of view the question which Simberloff considers tobe of great importance-whether communities exist as discrete enti-ties or are abstracted from a continuum of variation-loses itssignificance. Population genetics has also had to deal with the ques-tion of whether to treat a species as a single interbreeding populationwith non-random mating, a series of discrete "demes" with exchangeof migrants, or a one, two, or three dimensional continuum with adiffusion process, gene flow and local selection producing patterns ofisolation by distance. The solution is usually one of convenience: if therate of migration between habitats is very low, we use the laws oflocal population genetics and correct for migration. As the movementof genes increases, we have the models of patchy environments,multiple niches, etc., with random mating then corected by someinbreeding coefficient.

Similarly, if a patch of habitat is large enough so that interactionsare mostly within the patch and the probability of members ofdifferent species encountering each other closely enough for mutualinfluence is proportional to their abundances, we can treat theensemble as a community with correction for migration. If the pat-ches of habitat are small compared to the range of interaction andpropagation then a within-patch model will not work, and it is bette r toconceive of the community as itself a mosaic of habitats.

On small islands, the terrestrial community is bounded from the


8/32

54 RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI N

aquatic one in a sharp way which allows the models of islandbiogeography to ignore the distinction between island and communityand treat each island as a community. On continental areas or largeislands, the internal structure of the terrestrial habitat is more im-portant but boundaries among communities less clear. Nevertheless,the island biogeography approach to distributions of organisms hasbeen a fruitful one and usually picks out as "i slands" pieces of habitatthat may be regarded as communities.

Simberloff makes three assertions about these distributions: (a)organisms tend to have continuous distributions without abruptboundaries; (b) different species' boundaries do not usually coincide,preventing the identification of discrete communities ; and (c) when (a)and (b) are violated, there is usually some discontinuity in thephysical environment.

The ques tion of the boundaries of communities is really secondary tothe issues of interaction among species. There is nothing inherent in thecommunity concept which excludes physically determined boundaries.However, the insistence on a one-to-one correspondence betweenphysical and biotic distributions makes it more difficult:

(a) To recognize the very rich patchiness of nature, especially forsmaller organisms.

(b) To allow for threshold effects. For instance, a continuousenvironmental gradient can change the relative frequency of a plantspecies, precluding the maintenance of its own herbivores and build-ing an alternative insect community .

(c) To examine the structure of environment. While in some waysplants ameliorate severe environmental conditions and smooth overdifferences, they also create new kinds of environmental hetero-genity. The patchiness of the ant species mosaic described by Leston(1973) and also observed elsewhere reflects the amplification of smallenvironmental differences into more pronounced patchiness.

(d) To cope with alternative communities. As a limiting case, thespecies which is established first in a site may exclude colonists ofother species. This takes place because competition is occurringbetween established, mature adults of one species and the propaguleof the other. This life cycle difference may o ften outweigh differences


9/32

D I A L E C T I C S A N D R E D U C T I O N I SM 55in physiological responses to environment. Then physiologicaldifferences may affect the frequency with which a patch of a giventype is occupied by one species or the other, and a reductionist viewwould lose the competitive exculsions once it found the environ-mental correlation.

This situation obtains in the interaction of the neotropical fire antSolenopsis gerninata and the introduced cosmopolitan Pheidolemegacephala. Both are omnivorous, aggressive, and form largecolonies. Pheiclole is less tolerant of heat than Solenopsis but is moreable to nest and forage in trees. They are almost completely mutuallyexclusive on small islands where the established mature coloniesprevent successful colonization by the other. But on large islands,where patches of mature colonies come in contact, the outcomedepends more on their ecological differences. Each species is alsoassociated with other ants, making the alternative patches more thana single species substitution.

Other diffe rences- the polymorphism of Solenopsis vs. the clearcutdimorphism of Pheidole, the polygynous Pheidole colonies vs. thesingle-queen fire ant colony-are external to the present context, andrepresent effectively random intrusions into the system.

Thus the notion of multiple alternative steady states of com-munities is a natural consequence of the recognition of biologicalcomplexity, not the ad hoc patching of a dying paradigm as Sim-berloff claims.

Our view, a dialectical materialist approach, assigns the followingproperties to the community:

1. The community is a contingent whole in reciprocal interactionwith the lower and higher level wholes, and not completely deter-mined by them.

2. There are properties at the community level which are definablefor that level and which are interesting objects of study regardless ofhow they are eventually explained. Among such properties are diver-sity, equability, biomass, primary production, invasibility, and thepatterning of food webs. What makes these objects interesting is firstthat they appear as striking (tropical as against temperate diversity,the invasion of oceanic islands by cosmopolitan species, the rapid


10/32

56 RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI Novergrowing of abandoned fields) and demand explanation; secondlythat they seem to show some kind of regularity geographically; orfinally that they have been invoked to account for some of thepreviously given properties and are then seen to have their owncurious features (e.g., Joel Cohen's claim that food webs oftencorrespond to interval graphs). This is the weak form of the com-munity paradigm since it makes no claims as to the locus of explana-tion.

3. The properties of communities and the properties of the con-stituent populations are linked by many-to-one and one-to-manytransformations.

Many-to-one-ness means that there are many possible configura-tions of populations which preserve the same qualitative properties atthe level of the whole. This allows communities to be seen as similardespite species substitutions, and allows wholes to persist over timealthough the individual parts are constantly changing. Not all many-to-one relations are obvious: the discovery of those communitymeasures which are many-to-one functions of the component species isone of the major tasks of community ecology. Lane (1975) found thatsome of these measures for zooplankton communities persist over time,differ systematically among lakes, and change with eutrophication.

A second consequence of many-to-one relations is that it is notpossible to go backward from the one to derive the many. Thus lawsat a community level expressed as some persistent properties act asonly weak constraints on the parts. From the perspective of thecommunity, there are many degrees of freedom for the speciespopulations and these have the aspect of randomness with respect tocommunity levels.

The one-to-many relation of parts to wholes reflects the fact thatnot all properties of the parts are specified by rules at the level ofthese parts. For instance, the habitat may specify that all the speciesmust be able to tolerate "or avoid ext reme heat. Whether this isaccomplished by physiological tolerance, behavioral versatility infinding and staying in the cool spots, or dormancy during the hotseason is not deducible from the fact of heat but depends on the pastevolution of the species, yet will be of great importance in determin-


11/32

D I A L E C T I C S A N D R E D U C T I O N I S M 57

ing species interactions. And the mobility of the animal is not directlyrelated to the habitat but will affect its geography.

Therefore, one-to-many-ness is seen as an indeterminacy or ran-domness of the higher level with respect to the lower.

Together, the many-to-one and one-to-many couplings betweenlevels determine both the emergence of persistent features charac-terizing communities and also guarantee that different examples ofthe same kind of community will be different. Looked at over time,they allow us to see the unity of equilbrium (persistence) and change,determination and randomness, similarity with difference.

Things are similar: this makes science possible. Things aredifferent: this makes science necessary. At various times in thehistory of science the important advances have been made either byabstracting away differences to reveal similarity or by emphasizingthe richness of variation within a seeming uniformity. But eitherchoice by itself is ultimately misleading. The general does not com-pletely contain the particular as cases; the empiricist refusal to group,generalize, and abstract reduces science to collecting if not specimensthen examples. We argue for a strategy which sees the unity of thegeneral and the particular through the explanation of patterns ofvariation which are themselves higher order generalities that in turnreveal patterns of variation.

4. The interchange of law and constraint.Scientific explanation within a given level or context is often the

application of some law within constraints of some initial or boundaryconditions. These constraints are external to the domain of the lawand of no intrinsic interest. Thus a physics problem might be posed as"given a string 15 centimeters long, at what frequencies will itvibrate?" Nobody asks why the string is 15 centimeters long. Theinteresting phenomenon is the relation among the frequencies.Similarly, from the point of view of biophysics, the particularconfigurations of molecules and membranes in a cell are the boun-dary conditions within which the laws of thermodynamics happen tobe operating: biophysics is the study of the operation of physical lawsin some rather unusual conditions presented by living things. Butfrom the viewpoint of cell biology, the configurations of molecules


12/32

58 RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI Na nd m e mbra n e s a re p re c i sely t he ob j e c t s o f i nt e re s t. The q ue s t ionsconcern the i r format ion, maintenance , func t ion , and s igni f icance . Thel a w s o f t he rmodyna mic s a nd c onse rva t ion a re now the c ons t ra in t swi th in which ce l l metabol i sm and deve lopment take p lace .

This in te rchange of law and cons t ra in t a l so charac te r izes the popu-la t ion-communi ty re la t ion .

From the perspec t ive of the popula t ion gene t ics of each s ingle spec-ies in the comm uni ty , ' envi ron men t ' cons is t s of the phys ica l condi t ionsand those o ther spec ies which impinge on i t d i rec t ly . The o therme mbe rs o f t he c ommuni ty a re re l e va n t on ly i nso fa r a s t he y a f fe c tthe immediate ly impinging variable , but their influence is indirect anddoes not ente r the equa t ions of na tura l se lec t ion . The d i rec t ly im-pinging var iables ac t as de te rminants of ' f i tness . ' In genera l we exp ec ttha t t hose ge no type s w hic h su rv ive o r r e p roduc e more t ha n o the rgenotypes wi l l inc rease in f requency, thus changing the paramete rs ofthe l i fe table and ecology of the popula t ion .

Bu t f rom the pe rspe c t ive o f t he c ommuni ty , t he ge ne t i c a l l y de t e r -mined paramete rs of reproduc t ion , surviva l , feeding ra tes , habi ta tpre fe rences , and spec ies in te rac t ions a re g ivens , the cons t ra in tswi th in which the dynam ics of popu la t ion change opera te . Thesedyna mic s de pe nd ve ry se ns i t i ve ly on the s t ruc tu re o f t he c ommuni ty .They lead to conc lus ions of the fo l lowing kinds : the more s t rongly thefeeding pre fe rences of spec ies over lap , the less uni form wi l l be the i rre la t ive abundances , and the grea te r the f luc tua t ions over t ime;nut r ient enr ichment in lakes wi l l be p icked up most ly as inc rease inthe inedible spec ies of a lgae ; envi ronmenta l va r ia t ion ente r ing ac ommuni ty a t t he bo t tom of t he food w e b ge ne ra t e s pos i t i ve c o r -re la t ions among spec ies on adjacent leve ls , but va r ia t ion ente r ingf rom a bove ge ne ra t e s ne ga t ive c o r re l a t i ons ; popu la t i ons w hic h a repreyed upon by a spec ia l i s t wi l l be buffe red aga ins t changes a r i s inge lsewhere in the sys tem and wi l l respond through the i r age d is -t r ibut ion more than through to ta l numbers . Note tha t these resul t stake the s t ruc tures as g iven, wi thout inqui r ing as to the or ig ins ofspec ia l i s t s , inedible spec ies , or pe r turba t ions f rom above and be low.

The loose n e ss o f t he c oup l ing o f popu la t ion ge ne ti c a nd c om mun i ty


13/32

D I A L E C T I C S A N D R E D U C T I O N I SM 59phenomena prevents the complete absorption of the one into theother and requires instead the shifting between perspectives. Ittherefore precludes both the mechanistic reductionism and idealistholism.

5. The species of a community interact. This may be direct as inthe predator/prey relation, symbiosis or aggression, or it may be in-direct through the alteration of their common environment; im-mediately through impact on each others' abundance, age distribution,and physiological state or over evolutionary time by determining theconditions of natural selection acting on each one.

This claim would seem to be obvious enough not to require stating.However, Simberloff cites with approval the view that "the spatialdistribution of plants ...(is) ... a consequence of the individual,relatively uncoordinated responses of individual species to gradients inthe physical environment . . . "

If that were true, we would expect to find (a) a species is mostabundant where the physical environment is closest to its phy-siologically optimum conditions; (b) if all species but one are removedfrom a physical gradient, that one would increase but its relativeabundance would remain unchanged; (c) species should succeed eachother in time or space in the same direction as their physicaltolerances.

These expectations have not been tested systematically, but casesare known where it is not true. For example Dayton (1975) studied thedistribution of the alga HedophyUum sessile. The optimum phy-siological conditions for maximum growth occur where there isgreatest exposure to wave action, but in fact it is found only sporadic-ally as a fugitive in such places and is dominant in areas of moderateexposu re. Grassle and Grassle (1974) discussed the recoloniza tion bypolychaete worms of a bottom area depopulated by oil spill. In termsof physiological tolerances Nereis succinea should have come inbefore Capitella capitata, but the reverse was true.

There are many cases of species such as the brine shrimp reachingits greatest abundance where it can escape predators despite in-creased physiological stress, or of plants which are normally restric-


14/32

60 RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI Nted to certain soil types becoming ubiquitous on islands in theconditions of reduced competition, or the species composition of apasture depending on the grazing pattern.

Finally, we note that the asymmetry of the predator/prey relationmakes it impossible for both species to be most abundant where theirinteraction is most favorable: if the predator is most abundant whereits food supply is most favorable, then the food supply (prey) speciesis most common where it suffers greatest predation. Or if the prey isat its highest levels where the predator is absent, then the predatoris most common where the food supply is not optimal.

However, even where the abundance of a population correlateswell with physical conditions, this is not evidence that species aredistributed independently of each other. Here we come to one of themajor harmful consequences of the indivualistic approach to speciesdistribution and abundance: it counterposes the biotic and abioticfactors of a species' ecology, and treats physical factors as statistical'main effects' with relative weights.

In contrast, the community view is not that other species are moreimportant than physical factors but rather that there is a mutualinterpenetration of the physical and biotic aspects, that the ecologicalsignificance of physical conditions depends on relations with otherspecies, that the strong interactions among the components of acommunity make the components of variance approach misleading,and gives spurious support to the original bias.

Consider as an example the distribution of the harveste r ants of thegenus Pogonomyrmex in western North America. Their easternboundary falls between the 18" and 24" rainfall lines, identifyingthem as ants of arid and semi-arid regions. Yet these conditions arequite severe for the ants: the temperature at the surface of the soiloften gets into the 50's celsius, and the ants, which normally ceaseforaging in the upper 40's, have only a few hours a day available forgathering seeds. Experimental shading or watering of their nest areaextends their activity period and food intake. However, it also per-mits increased activity by the agressive fire ants (Solenopsis species)and competitors. The habitat requirement is first of all that there be a


15/32

D I A L E C T I C S A N D R E D U C T I O N I S M 61sufficient time span available for foraging when it is too hot for theother ants but still tolerable to Pogonomyrmex.

Aridity affects the ant distribution in several ways: dry air shows avery steep vertical temperature gradient in the sun, permitting theacceptable temperature range to occur; in arid habitats vegetation issparser so that more surface is exposed; a greater proportion ofplants of arid regions have dormant seeds which can be stored moreeasily, while dry air reduces spoilage in storage; the predators of theants-spiders, l izards, wasps-and the competitors-birds, rodents,other ants-also have their own climatic relations that are equallycomplex. The net result of these interactions is indeed a boundarycorrelated with rainfall, but to assert that therefore the distribution ofthe harvest ant is determined by physical conditions is to eliminatethe richness of ecology in favor of a statistical correlation.

Simberloff does not insist on physical determination exclusively.He will allow the importance of two or three other species. But hereagain the same issues arise: first, a strong correlation with anotherspecies is not sufficient grounds fo r assigning it causal predominance;second, if it is indeed the major cause of a species' abundance thismust itself be explained by its position in the community.

6. The way in which a change in some physical parameter orgenetic characteristic of a population affects the populations incommunity depends both on their individual properties and the waythe community is structured.

This is perhaps the critical claim of community ecology. It is notthe assertion that all components are equally important or that whathappens is the result of some superorganismic imperatives.

It is a necessary consequence of species interactions, relativelyindependent of how those interactions are described. Despite Sim-berloff's misunderstanding of it, it certainly does not depend on theassumptions of the logistic model. If species do interact, then com-munity structure determines the cons equences of the interaction; andif th~ outcome were to turn out to be deducible only from the unitinteractions themselves, this would not constitute a refutation of therole of community structure but rather would reveal a remarkable


16/32

62 RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI Nbehavior of that structure which would have to be accounted for.One way of representing community structure is by a graph inwhich the vertices are variables in the system, the lines connectingthem interactions identified only by sign: >for positive effectand O for negative effect. The mathematical procedures are givenin Levins (1975). The technical problems associated with identifyingthe appropriate graph are not relevant to its use here: the demon-stration that community structure determines what happens in com-munities, that these qualitative results do not depend on the finedetails of population level interactions but only on a few many-to-onequalitative properties.

This particular approach deals with systems in a moving equil-brium. More recent work shows that many but not all the results canbe extended to more general situations, and that even where theparticular results are different the relevant result-that the responsedepends on community structure - still holds.

Experimental verification of some of the predictions of this analysiswas provided in the recent experiments of Briand and McCauley(1978).In the graphs of Figure 1, we show some hypothetical communitiesof a few species. The corresponding table shows the direction ofchange in each variable when some parameter change enters thesystem in such a way as to increase the variable shown in the firstcolumn.

Figure la is a simple nutrient/consumer system. Any increase in theinput of nutrient to the system is completely taken up by the con-sumer, but a change in conditions affecting the survival of A affects Aand N in opposite directions, generating a negative correlation be-tween them.

In Figure lb, A is density-dependent in some way other than byconsumption of N. Now changes in N are absorbed both by N and Ain the same direction. The correlation between N and A depend onthe relative magnitudes of variation entering from above and frombelow.

In Figure lc, A is consumed by H. Now A no longer responds tochanges in N, which are passed on to H. (Although the population


17/32

(a)

D I A L E C T I C S A N D R E D U C T I O N I S M(b)

63

(e) ) (d)

(e) ~~ (f)Fig. 1. Graph representationof communitystructure.

level of A is unaffected, its turn-over rate and age distribution arealtered.) Once again we observe that change from above generatespositive correlation.

Figure ld introduces a second, inedible consumer. A2 picks up allthe effect of changing the input of N, leaving A1 and H relativelyinsensitive.

In Figure le, the second consumer, A2, also inhibits the growth ofA1 (perhaps by secreting a toxin, as in the case of blue-green algae).


18/32

64 RICH A RD L E V IN S A N D RICH A RD L E W O N T IN

The ef fec t o f thi s change in g raph s t ruc tu re i s seen on ly in the impact son H of var i a t ion en ter ing the sys tem v ia N1 or H.

Fina l ly , F igure I f in t rodu ces a second nu t r i en t cons ume d on ly byA2. Th i s a l te rs the re sponse s o f N1 to paramet ers en ter ing the sys te mat N1 or H and in t roduces ambigui t ies in to the responses of H.

An exa mina t ion o f the g raph model s o f F igure 1 and the con-s eq u en ces o f p a rame t e r ch an g e d e r i v ed f ro m t h em i n T ab l e I s h o w t h efo l lowing :

(a ) The response o f a spec ies to the d i rec t impact o f the ex terna lenv i ronment depends on the way tha t spec ies f i t s in to the communi ty .Species A1 responds di fferent ly to di rect inputs or changes in N1 in

TABLE I The direction of response of communityvariables to parameter changes entering the system atdifferent nodes. The responses are those of a slowlymoving equilibrium after transient effects are damped.Model Changeentering effect on:

through NI N2 AI A2 Ha NI 0 +A1 - +b Nl + +A - +c Nl + 0 +A 0 0 +H + - +d N~ 0 0 + 0A~ 0 0 0 +

A2 - 0 + -H 0 - + 0e N~ 0 0 + -A~ 0 0 0 +A2 - 0 + -

H 0 - + -f Nt + - 0 + ?N2 - + O + -A~ 0 0 0 0 +

A2 - - 0 + -H + - - + ?


19/32

DIALECTICS AND REDUCTIONISM 65Figure la and lb from its response in all other graphs, and Hresponds in opposite ways to the same physical impact in lc and le.

(b) Some species respond to changes arising almost anywhere in thesystem (A2, H) while A1 is insensitive to most inputs, responding onlyto the changes arising in H wherever H is present. This might bemisinterpreted as insensitivity to the environment or resourcechanges, or be taken as evidence of lack of competition with A2, butis really due to H playing the role of a sink that absorbs impactsreaching A1 from elsewhere.

(c) Some species (A2, H) affect most other variables in the com-munity, whereas changes entering through A1 are observed only inchanges in H. Thus the graph analysis supports the observation thatone or two species may dominate the community, but gives a com-pletely different explanation.

(d) A change in the structure of a community may be detectable notat the point of change but elsewhere. The difference between models(d) and (e) is only in the A2 OAI link, but the effects are seen onlyin the response of H to changes entering at H or NI.

(e) In table If we see that changes in parameters produce correlatedresponses in the variables of the system, and that the same pair ofvariables may have positive or negative correlations depending onwhere the variation enters the system (see the rela tion of A2 to N1 andN2, of N1 with N2, and H with N2).

(f) Parameter changes may be the result of natural selection. Ingeneral, the response to selection is to increase a parameter having apositive input to a variable. But this positive input may have positive,zero, or negative effects on the population size: population geneticsalone does not determine the demographic response to selection. Butsince population size does affect the numbers of migrants sent out tocolonize new sites, there is a discontinuity between populationgenetic and biogeographic processes that can be bridged only byspecifying community structure.

(g) The notion of a species being of critical importance or dominanthas at least three different meanings: H may be the major cause ofdeath of Az or only minor; N2 may be the main food for A2 or only asupplement. That in itself does not determine whether A1 responds to


20/32

66 R I C H A R D L E V I N S A N D R I C H A R D L E W O N T I Nchanges in H, A2 to changes in N2. Nor does does it answer thequestion of whether a species is critical to the structure in the sensethat for example the addition of N2 in model If changes the responseof N1 to its own parameters and to A2.

(h) The graph analysis opens up new possibilities for researchstrategy: in If it indicates where measurements are needed; con-sequences of parameter change which are concordant across modelsare robust results insensitive to details of the models; where differentmodels give different results we are directed to the critical observa tionfor deciding among them.

This dialectical approach to the ecological community allows for agreater richness than the reductionist view. It permits us to work withthe relative autonomy and reciprocal interaction of systems ondifferent levels, shows the inseparability of physical environment andbiotic factors, and the origins of correlations among variables, makesuse of and interprets both the many-to-one relations that allow forgeneralization and the one-to-many that impose randomness andvariation.

Where particular techniques are unsatisfactor y, the remedy is likelyto be not a retreat from complexity to reductionist strategies but afurther enrichment of the theory of complex systems.

A B S T R A C T I O N A N D I D E A L I S M

It is Simberloff's view that abstractions are a form of idealism andthat the materialism in science necessarily overthrows abstractionand replaces them with sort of 'real' entities which are then eachunique, because of the immense complexity of interacting forces oneach and because of the underlying stochasticity of nature. Howev er,he cannot really mean that all abstractions are to be eliminated or elsenothing would remain but chronicles of events. If any causalexplanations are to be given, except in the trivial sense that anhistorically antecedent state will be said to be the cause of later ones,then some degree of abstraction is indispensible. There is no predic-tability or manipulation of the world possible except that events canbe grouped into classes, and this grouping in turn means that unique


21/32

D I A L E C T I C S A N D R E D U C T I O N I S M 67properties of events are ignored and the events are abstracted. Thus,we can hardly have a serious discussion of a science without ab-straction. What makes materialist science is that the process ofabstraction is explicit and recognized as historically contingent withinthe science. Abstraction becomes destructive when the abstractbecomes reified and when the historical process of abstraction hasbeen fo rgotten so that the abstract descriptions are taken for descrip-tions of the actual objects.

The level of abstraction appropriate in a given science at a giventime is an historical question. No ball rolling on an inclined planebehaves as an ideal Newtonian body, but that in no way diminishesthe degree of understanding and control of the physical world that wehave acquired from Newtonian physics. Newton was perfectly con-scious of the process of abstraction and idealization that he hadundertaken, and he says in the De Motu Corporum:Every body per sever es in its state of rest, or of uniform motion in a right line, unless isit compelled to change that state by forces impressed thereon.Yet Newto n points out immediately that even "the great bodies of theplanets and comets" have such perturbing forces impressed uponthem and that no body perseveres indefinitely in its motion.

On the other hand, the properties of falling bodies that have beenabstracted away are replaced when necessary; Newton himself, inlater sections of the Principia, considered friction and other suchforces. Landing a space capsule on the moon requires the physics ofNewtonian ideal bodies moving in vacuums for only part of itsoperation. For other parts, an understanding of friction, hydro-dynamics, and aerodynamics in real fluid media must be involved.Finally, there are correction rockets, computers and human minds tocope with the idiosyncracies of actual events. No space capsule couldland on the moon without Newton ian abstractions nor solely with them.The problem for science is to understand the proper domain ofexplanation of each abstraction rather than becoming its prisoner.

The argument given by Lewontin (1974) that Darwin and Mendelrepresented a materialist revolution in biology was not based on theassertion that they overthrew abstract ions but that they overthrew


22/32

68 RI CH A RD ' I ~EV I N S A N D RI CH A RD LEW O N TI NPlatonic ideals. Darwin's and Mendel's works are filled with ab-stractions (species, hereditary factors, natural selection, varieties,etc.). The error of idealism is the belief that the ideals are unchangingand unchangeable essences that enter into actual relationships witheach other in the real world. Ideals are abstractions that have beentransformed by fetishism and reification into realities with an in-dependent ontological status. Moreover, idealism sees the relation-ships entered into by the ordinary objects of observation as directcausal consequences, albeit disturbed by other forces, of the actualrelations between the essences. Marx, in discussing the fetishism ofcommodities in Chapter I of Capital, draws a parallel with "themist-enveloped regions of the religious world. In that world theproductions of the human brain appear as independent beingsendowed with life, and entering into relations both with one anotherand the human race." In a similar way idealistic, pre-Darwinian,biology saw the actual organisms and their ontogenetic histories ascausal consequences of real relations among ideal, essential types, asopposed to the materialistic view that sees the actual physical rela-tions as occuring between actual physical objects with any 'types' asmental constructs, as abstractions from actuality. The precisedifficulty of pre-Darwinian evolutionary theory was that it could notreconcile the actual histories of living organisms, especially theirsecular change, with the idea that these histories were the causalconsequences of relationships among unchanging essences. Theequivalent in Newtonian physics would have been to suppose (asNewton never did) that if a body departed from perfectly rectilinear,unaccelerated motion, there nevertheless remained an entity, the'ideal body,' that continued in its ideal path and to which the actualbody was tied in some causal way. What appears to be the patentabsurdity of this view of moving bodies should make clear to us thecontradictory position in which pre-Darwinian evolutionists foundthemselves.

In ecology the isolated community is an abstraction in that no realcollection of species exists which interacts solely with its own mem-bers and which receives no propagules from outside. B/ut the totalisolation of a group of species from all interactions with other species


23/32

D I A L E C T I C S A N D R E D U C T I O N I SM 69is not a requirement of the usefulness of the community as ananalytical tool. Some of Simberloff's confusion on this point arisesfrom a failure to appreciate that general principles of interaction arenot the same as quantitiative expressions of these interactions. It isundoubtedly true that every body in the universe creates a gravita-tional field that, in theory, interacts with every other one. Yet whenwe get up in the morning, our muscles and nerves do not have tocompensate for the motion of every body in the universe, or even ofevery other person in the same house with us. The intensity ofgravitational interaction is so weak that, except for extraordinarilymassive objects like planets or extraordinarily close objects likenucleons, it is irrelevant, and we can treat our own persons asgravitationally independent of each other. In like manner, all speciesin the biosphere interact, but the actual matrix of interactioncoefficients is essentially decomposable into a large number of sub-matrices separated by zeroes. The problem for the ecologist is not toreplace these zeroes by the infinitestimally small actual numbers, butto find the boundaries of the submatrices and to try to understand therather large interaction coefficients that exist within them. Thus, it isnot an argument against the population or the community as entities,that boundaries are not absolute between them, any more than thatthe exis tence of some intersexes destroys the usefulness in biology ofdistinguishing males and females.

To put the matter succinctly, what distinguishes abstractions fromideals is that abstractions are epistemological consequences of theattempt to order and predict real phenomena, while ideals are regar-ded as ontologically prior to their manifestation in objects.

ST O C H A ST I C I T Y A N D ST A T I ST I C S

Simberloff correctly observes that a major trend in ecology andevolutionary biology has been the replacement of deterministicmodels by stochastic ones. He draws from that observation twoconclusions, however, with which we disagree. First, he concludesthat stochastic models are in essential contradiction to predictivemodels, and that stochasticity is the negation of cause-and-effect.


24/32

70 RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI NThus, he writes "The neo-Darwin syntheses sounded the death-knellfor Newtonian cause-and-effect determinism in biology." As a gen-eral statement about biology, this certainly is not and cannot betrue. As an historical fact, the entire development of molecularbiology shows the continuing power of simple deterministic models ofthe 'b~te-machine' nor is there the slightest reason to introducestochasticity into models of, say, how an increase in adrenalinsecretion will affect the concentration of sugar in the blood. Indeed,stochasticity may be an obfuscation rather than a clarification in suchcases. The neurosecretory system is a complex network of non-lineardynamic relations that are incompletely understood. If two in-dividuals (or the same individual at different times) are given identicaltreatments of a hormone, there may be qualitatively different and evenopposite consequences. That is because in such a non-linear system, theconsequences of a perturbation in one variable are strongly dependen t onthe levels of the othe r constituents. The lack of repeatability of responsecould be passed off as the con sequenc e of 'stochast icity' or the inherentfailure of 'Newtonian cause-and-effect,' but to do so would be to preventprogress in understanding and control of the system. The example ofPark's experiments in ecology, cited by Simberloff as a triumph ofstochas tic modelling, is right to the point. Mixed popula tions of Tribol iumc o n f u s u m and T . c as tane um sometimes resulted in replacement of onespecies, sometimes of the other. Conditions of food, moisture, etc weremade as near ly identical as possible and the initial population mixtureswere also controlled. A stochastic model of this competive experimentwas con structed by Neyman, Pa rk and Scott (1956) which was consistentwith the variable outcome of the experiments. But in constructing such astochas tic model, which seems untestable, an alternative that would leaddirectly to experiment and measurement has been rejected. Thisalternative is that there are two stable states of dynamic sys tems, one atpure Tribol ium cas taneu m, one at pure T . c on fusum , and that the domainsof attraction of these stable states are demarked by a separa trix along anaxis that has not been controlled in the initial population mix, so that theexper iments begin sometimes on one side of this separatrix, sometimeson the other. Park did not examine, for example, the effect of small


25/32

D I A L E C T I C S A N D R E D U C T I O N I SM 71per turbat ions in the initial age distribution within species, or in the initialac tual fecundities of the samples of beetles in each vial.

It may indeed be t rue that notions of cause-and-effect are inapplic-able at the level of the spontaneous disintegration of a radioactivenucleus, but there is no reason to make uncertainty an ontologicalproperty of all phenomena. It is a curious inversion of the idea ofmaterialistic explanation to suggest that "the next great breakthr oughmay come when we have the courage to junk cause-and-effectentirely, and strike off at some other angle." Astrology, perhaps, orE.S.P.?

The question of whether non-predictability of outcome is to besubsumed under a general stochasticity, or whether previously un-controlled variables are to be controlled in an attempt to producepredictable outcomes, is an ad hoc issue to be decided in each case.

If we wish to understand the changes in gene frequency in apopulation, it may be quite sufficient to envoke the 'chance' nature ofMendelian segregation and the Poisson distribution of the number ofoffspring from families in a finite population of size N. That isbecause such a stochastic explanation is a sufficient alternative to atheor y of perf ect adaptation by natural selection. It is an explanationat the same level of phenomena as the adaptive story. On the otherhand, if we are interested in the consequences of human demographicchange, the probability distribution of family sizes is not a sufficientlevel of analysis, and we must look into the differentiation of familysize by region, class, etc. The demand that all phenomena must beexplicable by deterministic cause and effect at an arbitrary level ofexplanat ion is clearly doomed to failure, as for example the attemptto explain all evolutionary change as the result of determinativenatural selection. But the asse rtion that cause-and-ef fect at a lowerlevel cannot replace the stochasticity at h~gher levels, if it becomesuseful to do so, is obfuscatory.

Moreover, the shift from stochastic to deterministic statementsabout the world can occur in changing from one level of explanationto another in either direction. Not only can the apparently random beexplained as a result of deterministic forces in higher dimensionality


26/32

72 RI CH A RD LEV I N S A N D RI CH A RD LEW O N TI Nwith more specification, but a reduction in dimensionality byaveraging also converts stochasticity into determination. The sto-chasticity of molecular movements in a gas lies at the basis of thecompletely deterministic gas laws that relate temperature, pressure andvolume. Even if the disintegration of a radioactive nucleus is an'uncaused' event and thus perfectly gtochastic, clocks accurate tomillionths of a second are built precisely on the basis of the randomness ofthose disintegrations. Thus stochastic processes may be the basis ofdeterministic process and deterministic the basis of stochastic. They donot exclude each other.

Stochastic and deterministic processes interact also at the samelevel of organization of phenomena, and this interaction is of especialimportance in population biology and evolution. The notion ofdeterminism may carry with it the false implication that only singleend state is possible for processes if all of the parameters of thedynamic system are fixed. But this is not true. Because of thenon-linear dynamics of evolutionary processes, there exist multiplepossible outcomes for a process even with fixed parameters. Inmathematical terms, the vector field has multiple points of attraction,each surrounded by a domain of attraction. Which end point theprocess ac tually achieves depends upon in which domain of attractionthe system begins, its initial condition. Thus, the same force ofnatural selection may cause a population to evolve in different direc-tions dependent upon the initial genetic composition. If, in addition tothe deterministic force of natural selection, there are random varia-tions in genetic composition from genera tion to generation because offinite population size and random migration, a population in one domainof attraction, may be pushed into another domain and thus may achieve afinal state different from that predicted on the basis of its previoustrajectory. Indeed, a good deal of evolution by natural selection is onlymade possible by stochastic events, because these events allow apopulation that has been restricted to a domain of attraction in thegenotypic space to evolve into other compositions. The synthetic theoryof evolu tion developed by Wright (1931) is based upon this 'exploration'of the field of possible evolutionary outcomes by the interaction ofstochastic and deterministic forces, both operating at the level of


27/32

D I A L E C T I C S A N D R E D U C T I O N I SM 73genotypic composition. Again we see that the apparent contradictionbetween stochastic and deterministic is resolved in their inter-action.

Second, Simberloff's discussion of stochasticity also turns on itshead the relationship of material explanation and statistical explana-tion. He writes that "Statistics is inherently materialistic and anti-typological since it takes 'noise' as it object of study ." Nothing couldbe further from the truth. Some of the greatest problems of scientificexplanation come from concepts and practices that lie at the heart ofmodern statistics which, in many ways, is the embodiment of ideal-ism, at least as practiced by natural and social scientists.

In the first place, statistics does not take 'noise' as its object ofstudy, but on the contrary consists largely of techniques for reducing,discounting or separating 'noise' so that 'real' effects can be seen. Thetheory of hypothesis testing and most of the theory of estimation had astheir primary purpose the detection of true difference between objectsor the assignment of intervals in which parameters of universes arethought to lie, in spite of variation between individuals. While state-ments about differences or parameters must of necessity be phrased interms of probabilities, that is not regarded as a virtue by statisticaltheory, but as a limitation. The reason for searching for el~cientestimators and uniformly most powerful tests is precisely in order tominimize the effect of variation between individuals on the desiredinferences about ideal universes. The distinction between first andsecond moments is absolutely fundamental to statistical theory 1and thepurpose of statistical procedures is to distinguish that fraction of thedifference between first moments which is ontologically the same as thesecond moment, from that fraction which arises from different causes,the 'real' differences between the populations. Most of the theory ofexperimental design such as randomization, orthogonal plots andstratification are a substitute for complete knowledge and control of allrelevant variables. The purpose is not to study the 'error' variance, butto tame it and minimize it and finally to remove, if possible, the veil ofobscurity that it interposes between the observer and those idealuniverses whose parameters are the object of study.


28/32

74 R I C H A R D L E V I N S A N D R I C H A R D L E W O N T I NThe branches of statistics that seem at first glance to be concerned

most directly with variance as an object of study, the analysis ofvariance and multivariate correlation and regression theory, are aspracticed by natural and social scientists, if not by sophisticatedstatisticians, the most mystified by idealism. The analysis of variance is atautological partitioning of total variance among observations into maineffects and interactions of various orders. Yet, as every professionalstatistician knows, the partitioning does not separate causes exceptwhere there is no interaction. (See Lewont in, 1974, for a discussion of thispoint in the context of population genetics.) Yet natural and socialscientists persist in reifying the main effect and interaction variance thatare calculated, converting them into measures of separate causes andstatic interactions of causes. Moreover, they act as if 'main effects' werereally 'main' causes in the every day English meaning of the word and thatinteractions are really of a second order of importance. Interaction in thisview is what is left over after main effects are accounted for. Thisattitude toward main effects and interactions is a form of the ceterisparibus assumption that plays such a central role in all Cartes ian science,but that has become an unconscious part of the ideology of the analysisof variance.

The most egregious examples of reification are in the use ofmultiple correlation and regression and various forms of factor andprincipal components analysis by social scientists. Economists,sociologists, and especially psychologists believe that correlationsbetween transformed orthogonal variables are a revelation of the'real' structure of the world.

Biologists are apparently unaware that in the construction of thecorrelation analysis itself they impose a model on the world. Theirassumption is that in performing a correlation analysis they areapproaching the data in a theory-free manner and tha t data will 'speak tothem' through the analysis. If, however, we examine the actual relation-ship between dynamic systems and correlations, it becomes clear thatcorrelation can create relationships that do no t exist. For example, thesimplest prey-predator relations predict that as prey increase there willbe a consequent increase in predators, so the correlat ion between preyand predator should be positive, but as predators increase, all other


29/32

D I A L E C T I C S A N D R E D U C T I O N I S M 75t h i ngs be i ng e qua l , p r e y s hou l d de c r e a s e s o t he r e w i l l be a ne ga t i vec o r r e l a t i o n i n a b u n d a n c e . T h e s p ir a l n a t u r e o f t h e d y n a m i c s i n t h e t w od i m e n s i o n a l p r e y - p r e d a t o r s p a c e s h o w s u s i m m e d i a t e l y t h a t p r e y a n dp r e d a t o r a b u n d a n c e s m a y b e e i th e r p o s i t i v e l y o r n e g a t iv e l y c o r r e l a t e dd e p e n d i n g u p o n t h e p a r t i c u l a r p a r t o f t h e s p i ra l th e p o p u l a t i o n s a r e i nh i s t o r i c a l l y . I f a n a t he o re t i c c o r r e l a t i on a na l ys i s i s c a r r i e d ou t , ac o r r e l a t i o n w il l b e o b s e r v e d a n d , i n th e a b s e n c e o f a n y a prior i t h e o r y , t h ec o r r e l a t i on w i l l l e a d t o a t he o re t i c a l s t o ry t ha t r e f l e c t s the pa r t i c u l a r s i gnt h a t t h e c o r r e l a t i o n h a s i n th a t s e t o f d a ta . T h u s , c o r r e l a t i o n s m a y b e t h ec o n s e q u e n c e o f c a u s a l p r o c e s s e s , b u t t h e y c a n n o t b e u s e d r e l i a b l y t o i n f e rt h o s e p r o c e s s e s .

B e c a u s e t h e m e t h o d o l o g y o f c o r r e l a t i o n i s i n t r i n s i c a l l y w i t h o u tt h e o r e t i c a l c o n t e n t a b o u t t h e r e a l w o r l d ( t h a t is t h o u g h t t o b e i t s g r e a t e s tv i r t u e ) a n y s t a t e m e n t s a b o u t t h e re a l w o r l d m u s t c o m e f r o m t h e c o n t e n ti m p o r t e d i n t o t h e a n a l ys i s. S o , i f w e w i s h t o u n d e r s t a n d t h e c a u s e s o fs ome va r i a b l e , s a y s pe c i e s a bunda nc e , by us i ng a c o r r e l a t i ona l a p -p r o a c h , i t b e c o m e s n e c e s s a r y t o d e c i d e w h i c h a s p e c t s o f t h e w o r l d a r e tob e m e a s u r e d t o c o r r e l a t e w i t h s p e c i e s a b u n d a n c e . H a v i n g c h o s e n t h ei n d e p e n d e n t v a r i a t e s , t h e c o r r e l a t i o n s t h a t a r e c a l c u l a t e d c o m e t o b ei n t e r p r e t e d a s r e a l c a u sa l c o n n e c t i o n s . S o i f t e m p e r a t u r e t u r n s o u t t o b eh i g h ly c o r r e l a t e d w i t h a b u n d a n c e , i t w il l b e a s s e r t e d t h a t t e m p e r a t u r ei t se l f is a n i m p o r t a n t c a u s a l f a c t o r ; a s if th e d a t a t h e m s e l v e s r a t h e r t h a nt h e o b s e r v e r h a d c h o s e n t h is v a r i a bl e . O f c o u r s e e v e r y i n v e s t i g a to r w il lr e p e a t e n d l e s s l y t h a t c o r r e l a t i o n s h o u ld n o t b e c o n f u s e d w i t h c a u s a t i o na n d t h a t , i n t h e e x a m p l e g i v e n , t e m p e r a t u r e m a y b e o n l y a p r o x y f o rs om e o t he r va r i a b l e o r va r i a b l e s w i t h w h i c h i t i s i n t u rn c o r r e l a t e d . B u ts u c h a d i s c l ai m e r is d i s in g e n u o u s . N o o n e w o u l d b o t h e r t o c a r r y o u t ac o r r e l a t i o n a n a l y s is i f t h e y t o o k s e r i o u s ly t h e caveat t h a t c o r r e l a t i o n sa r e no t c a us a t i ons . A f t e r a ll , w h a t i s t he u s e o f t he a n a l ys i s un l e s s i t i s t ob e u s e d f o r i n f e r e n c e s a b o u t c a u s a t i o n ?

U n fo r t un a t e l y , i t i s ne a r l y a l w a ys t he c a s e t ha t i n a c o l l e c t i o n o fm u l t i v a r i a te d a t a i n w h i c h t h e s e t o f i n d e p e n d e n t v a r i a b l e s a c c o u n t s f o ra r e a s o n a b l e p r o p o r t i o n o f t h e v a r i a n c e , a r a t h e r l ar g e p r o p o r t i o n o f t h a tva r i a nc e w i l l be a s s o c i a t e d w i t h a sma l l p ro po r t i o n o f t he va r i a b l e s . T h i sl oa d i ng o f the va r i a nc e on t o a s ma ll s e t o f va r i a b l e s i s a pu re l y num e r i c a lc o n s e q u e n c e o f a s s e m b l i n g a h e t e r o g e n e o u s g r o u p o f i n d e p e n d e n t


30/32

7 6 R I C H A R D L E V I N S A N D R I C H A R D L E W O N T I N

variates in a multiple regression analysis . A consequence of this loadingis that one or two variates will always appear to be the 'main' dependentvariable. Yet, if the analysis is repeated with a different set of variables,some other may now appear as the 'main' causes. In this way thepractice of multivaria te analysi s is self-reinforcing since it appears fromthe analysis that a few real 'main' causes have been discovered and sofaith in the methodo logy is built.

When extrinsic variables are not introduced specifically as explanatoryfactors, a complex set of data may be examined internally for pattern orstructure whose discovery is thought to be a revelation of real world,while, in fact it is only a tautological relationship among a set ofnumbers. The mos t famous example is the g-factor that is created in thefactor-analytic treatment of I.Q. tests and which is widely believed bypsychologists to be a real thing, general intelligence. Statisticalmethodo logy in the hands of natural and social scientists thus becomesthe mos t powerful form of reinforcing praxis of which idealism is thetheory.2

C O N C L U S I O N S

Biology above the level of the individual organism - population ecologyand genetics, community ecology, biogeography and evolution-requires the study of intrinsically complex systems. But the dominantphilosophies of western science have proven to be inadequate for thestudy of complexity:

(1) The reductionist myth of simplicity leads its advocates to isolateparts as completely as possible and study these parts. It underestimatesthe importance of interactions in theory, and its recommendations forpractice (in agricultural programs or conservation and environmentalprotection) are typically thwarted by the power of indirect andunanticipated causes rather than by error in the detailed description oftheir own objects of study.

(2) Reductionism ignores properties of complex wholes; the effects ofthese properties are therefore seen only as noise; this randomness iselevated into an ontological principle which leads to the blocking of


31/32

DIALECTICS AND REDUCTIONISM 77investigation and the reification of statistics, so that data reduction andstatistical prediction often pass for explanation.

(3) The faith in the atomistic nature of the world mak es the allocationof relative weights to separate causes the main object of science, andmakes it more difficult to study the nature of interconnectedness.

Where simple behaviors emerge out of complex interactions, it takesthat simplicity to deny the complexit y; where the b ehavio r is bewilder-ingly complex it reifies its own confusion into a denial of regularity~

Both the internal theoretical needs of eco logy and the social demandsthat it inform our planned interactions with nature require an ecologythat makes the understanding of complexity the central problem: it mustcope with interdependence and relative autonomy, with similarity anddifference, with the general and the particular, with chance andnecessity, with equilbrium and change, with continuity and dis-continuity, with contradictory processes. It must become increasinglyself-conscious of its own philosophy, and that philosophy will beeffective to the extent th at it bec ome s not only materialist, but dialectical.School of Public Health andMuseum of Comparative Zoology,Harvard University

NOTESI We owe this realization to a remark by William Kruskal.2 Thus, we subscribe completely to the point of view expressed by Isadore Nabi in hisessay 'On the Tendencies of Motion.' Dr. Simberloff is off the mark, however, in hisattribution of its authorship.

BIBLIOGRAPHYCohen, J.: 1978,Food webs and niche space, Monographs in Population Biology 11,Princeton University Press, Princeton,Lane, P.: 1975, The dynamics of aquatic systems: a comparative study of the structure offour zooplankton communities),Ecol. Monog.95(4), 307-336, Belknap Press of HarvardUniversity Press.Leston, Dennis: 1973, The ant mosaic - tropical tree crops and the limiting of pests anddiseases', PANS 19, 311-341.


32/32

78 R I C H A R D L E V I N S A N D R I C H A R D L E W O N T I N

Levin s, R.: 1975, 'Evo lutio n in communities near equilibrium', in Martin L. C ody andJared Diamond, eds . , Ecology and Evolution of Communities, Harvard Univers i tyPress, Cambridge, Mass.

Lewo ntin, R. C.: 1974, 'Th e analysis of variance and the analysis of caus es', Amer. I. ofHum. Gen. 26, 400-411.Lewo ntin, R. C.: 1974, 'Darw in and Mendel - the materialist revo lutio n',in J. Neym an, ed.,The Heritage of Copernicus: Theories 'More Pleasing to the Mind,' M.I.T. Press , N ew

York.Ney man , J., T. Park, and E. L. Scott: 1956, 'Struggle for existence. The Tribolium model:biological and statistical aspects ' . Third Berkeley Symposium on Math. Stat. & Prob.4

41-79.Simberloff, B.: 1980, 'A succession o f paradigms in ecology: Essen tialism to materialismand probabilism', Synthese, this issue, 3-39.Wilson, E. O.: 1978, On Human Nature, Harvard Univers i ty Press , Cambridge , Mass.

Documents

Levin Slew on Tinsy n These 1980