Codification, patents and the geography of knowledge transfer in …aj/archives/articles/Reiffen... · 2015-02-06 · Codification, patents and the geography of knowledge transfer

Codification, patents and the geography ofknowledge transfer in the electronic musicalinstrument industry

TIM REIFFENSTEINDepartment of Geography, Mount Allison University, Sackville, NB, Canada, E4L 1A7 (e-mail: [email protected])

Recent research in economic geography hasemphasized tacit knowledge as the basis of industriallearning. In contrast, codification and the practices ofindustrial writing have received little attention for theroles they play in mobilizing knowledge across space.This paper offers insight into the geographies ofcodification through an examination of technologytransfer in the electronic musical instrument industrybetween 1965 and 1995. The research draws on avariety of primary and secondary data that includeinterviews with inventors, biographical accounts andpatent analysis. These sources offer perspective onthe career trajectories of three U.S. inventors whotransferred knowledge from various contexts inCalifornia’s high-tech industry to the Japanese firm,Yamaha. Conceptually, the paper draws on theactor–network theory and Latour’s idea of translationto highlight the detours inventors must take toregister novelty. The analysis reveals the problematicnature of codified knowledge and its transfer; in thiscase codified knowledge was mobile internationallybut not locally, at least until it reached Japan. Thepaper argues for the need to understand how textssuch as patents are produced—the context of theirauthorship, the geographies of their circulation andtheir efficacy for shaping further innovative practice.

Les recherches actuelles en geographie mettentl’accent sur les connaissances tacites commefondement de l’apprentissage industriel. Cependant,la codification et les pratiques relatives a lacomposition industrielle ont ete peu etudiees du pointde vue de leurs roles dans la mobilisation desconnaissances dans l’espace. Cet article donne unapercu des geographies de la codification suite a uneanalyse du transfert technologique dans l’industriedes instruments de musique electronique entre 1965et 1995. Fondee sur un ensemble de donneesprimaires et secondaires, cette etude presente uneserie d’entrevues realisees aupres d’inventeurs, descomptes-rendus biographiques et des analyses debrevets. Ces donnees permettent de considerer avecrecul le cheminement professionnel de troisinventeurs americains responsables du transfert desconnaissances depuis differents secteurs del’industrie de haute technologie californienne vers lasociete japonaise Yamaha. Sur un plan conceptuel,l’article reprend la theorie acteur-reseau et aborde lanotion de traduction developpee par Latour afin demettre en relief les principaux detours qu’empruntentles inventeurs pour obtenir un brevet d’innovation.L’analyse fait ressortir le caractere problematiquedes connaissances codifiees et de leur transfert;

The Canadian Geographer / Le Geographe canadien 50, no 3 (2006) 298–318C© / Canadian Association of Geographers / L’Association canadienne des geographes

Codification, patents and the geography of knowledge transfer 299

dans ce cas, les connaissances codifiees etaientmobiles a l’echelle internationale et non a l’echellelocale avant qu’elles n’arrivent au Japon. Cet articleplaide en faveur de la necessite de comprendrecomment les textes tels que les brevets sont elabores:le contexte entourant la redaction du document, lesgeographies de leur diffusion et les repercutions surles pratiques novatrices.

Introduction

Economic geographers have long recognized in-dustrial learning as a situated process that isterritorially and socially embedded. The vast lit-erature on industrial districts, clusters and learn-ing regions that accumulated through the 1990semphasized place-based learning (Markusen 1996;Braczyk et al. 1998; Malmberg and Maskell 2002).More recently, a second current of inquiry hassought to unravel how learning takes place acrossspace and between regions (Bunnel and Coe2001; Oinas and Malecki 2002; Bathelt et al.2004). Both groups of studies have frequently re-hearsed Michael Polanyi’s (1967) conceptual dis-tinction between tacit and codified knowledge.In a world where codified knowledge is thoughtto be increasingly ubiquitous (Maskell and Malm-berg 1999) many of these accounts view the lead-ing edge, the quality that makes places special,as the province of tacit knowledge. In contrastthere remains a lacuna in our understanding ofhow industrial geographies shape and are shapedby the practice of codification. This imbalanceis regrettable, especially considering Nonaka andTakeuchi’s (1995) widely cited observation thatthe cycling between codified and tacit knowledgeholds the key to industrial problem solving. Toaddress this research gap, it is necessary to morefully consider the codified dimension and the artof industrial writing.

Codified knowledge serves as the recipe for in-dustrial learning. Codification is the process ofinscribing information, the translation from prac-tice to page or what Nonaka and Takeuchi re-fer to as ‘externalization’. Maskell and Malmberg(1999, 15) capture the significance of this pro-cess when they state that ‘technological progressis to a large extent the result of an interlinkedprocess of knowledge creation and subsequentcodification. Codification is thus at the heart of

the whole philosophy of industrialization’. In theindustrial arena, knowledge externalization pro-duces several outcomes including blueprints, for-mulas, sketches and patents. Though each ofthese inscriptions fixes knowledge in a potentiallymobile form, they vary according to the situatedconditions governing their authorship and use. Apatent serves as a more authoritative claim tonovelty than an idea hastily sketched on a cock-tail napkin.

Intellectual properties (IP)1 are the currencyof a knowledge-driven economy. Taplin (2004,1) estimates that these sorts of intangible as-sets comprise 70 percent of the assets of majorcorporations. As Macdonald (2004, 135) argues,patents have acquired ‘a strategic value increas-ingly independent of innovation’. IBM’s patentportfolio, for example, annually generates U.S.$1 billion in profit through licensing. To pro-duce a comparable figure, IBM would have tosell an additional $20 billion in products (Rivetteand Kline 2000, 56). IP management has thus be-come a crucial firm competency to be integratedthroughout the product cycle. According to itsformer vice president of R&D, when razor man-ufacturer Gillette sought to develop the Sensorbrand, ‘the first challenge . . . was mapping out thepatent landscape surrounding the shaver’s keyperformance attribute’ (Rivette and Kline 2000,58; emphasis added). Accompanying these spa-tial metaphors that hint at the tacit componentof working with these inscriptions, is an increas-ing recognition that in the world of IP, geographymatters. National patent systems across majorindustrialized countries differ significantly andfirms face great uncertainty in both registering

1 Intellectual properties comprise patents, trademarks, copy-rights and trade secrets.

The Canadian Geographer / Le Geographe canadien 50, no 3 (2006)

300 Tim Reiffenstein

and defending their patents across borders. TheUnited States judges patents according to the‘first to invent’ principle, whereas virtually everyother country follows the ‘first to file’ system.However, moves towards international patent har-monization, such as the WTO’s Trade RelatedAspects of Intellectual Property (TRIPS) protocol,have been criticized by countries in the globalsouth for undermining indigenous knowledge sys-tems (Shiva 2000). As these examples illustrate,codified knowledge matters, yet there is a poorunderstanding of the processes through whichpatents and other industrial inscriptions are pro-duced, mobilized and officiated.

This paper offers insight into the geographiesof codification through an examination of theelectronic musical instrument (hereafter EMI) in-dustry between 1965 and 1995. This period ismarked by two profound shifts in the EMI in-dustry. First, the technological basis of the EMIindustry shifted from analog to digital sound syn-thesis (Theberge 1997; Pinch and Trocco 2002).Second, the industrial advantage shifted deci-sively from the United States to Japan, specif-ically to Hamamatsu, an industrial city locatedon the Pacific coast midway between Tokyo andOsaka (Reiffenstein 2004, 2005). Yet, the keypatents—a vital form of codified knowledge—thatprovided the catalyst for paradigmatic technolog-ical change were made by American engineersbased in California, and American industry hadthe opportunity to access their breakthroughs.Why was this vital codified knowledge not prop-erly appreciated in the United States? How andwhy was this codified knowledge transferred, un-derstood, translated and elaborated by Japanesefirms? These questions underline the problematicnature of codified knowledge in industrial loca-tion dynamics and regional development and pro-vide the empirical goals of this paper.

Conceptually, I connect the themes of knowl-edge conduits with actor–network theory and La-tour’s science studies, especially in relation to theidea of ‘translation’. Actor–network theory andLatour’s idea of translation offer complementaryinsights into how knowledge networks are mo-bilized, including through the interdependenciesbetween people (e.g., engineers) and things (e.g.,patents). This discussion privileges patenting asa problematical process that connects scientificpractice, engineers and innovation. Empirically,

the paper sketches the career trajectories of threeU.S. inventors who transferred knowledge fromvarious contexts in California’s high-tech industryto Yamaha, a Japanese firm based in Hamamatsu.In this transfer of knowledge, special attentionis paid to the role of patents, most notably key‘breakthrough’ patents that became the focus ofconsiderable controversy. In terms of organiza-tion, the paper first elaborates on spatial trans-fers of knowledge through conduits and networkscomprising people, texts and artifacts, primarilywith reference to science studies and the themeof translation. Empirically, the paper then ana-lyzes the role of engineers, firms and patents inthe transformation and relocation of the EMI in-dustry.

Knowledge Transfers, Codificationand Translation over Space

The tacit–codified knowledge dichotomy is fre-quently the starting point for economic ge-ography’s examination of knowledge creationand transfer (Asheim 1999; Cowan et al. 2000;Johnson et al. 2002). Codified knowledge, alsoknown as formal, explicit and articulated knowl-edge, concerns those forms of scientific or engi-neering knowledge that exist in textual form andare thereby potentially tradable. Tacit or informalknowledge, on the other hand, in Polanyi’s (1967,4) words is the highly personalized and con-textual realm of knowledge in which, ‘we knowmore than we can tell’. In practice the two areinterdependent. As Nonaka and Takeuchi (1995;Nonaka et al. 2000) argue, knowledge creationproceeds from the cycling between tacit and cod-ified knowledge. Geographers have seized on thismutually constitutive quality to caution againstbinary spatial logics that read tacit knowledge aslocal and sticky and codified knowledge as globaland slippery. Asheim (1999) proposes an inter-mediate form of contextual knowledge, ‘disem-bodied knowledge’ that comprises locally stickyforms of codified knowledge. Bathelt et al. (2004,31) similarly question ‘the view that tacit knowl-edge transfer is confined to local milieus whereascodified knowledge may roam the globe almostfrictionlessly’. While there has been increasing in-timations about the importance of tacit knowl-edge in economic geography (Gertler 2004), the



unpacking of codified knowledge has been left farbehind.

Knowledge conduits

A useful, though limited, approach to under-standing spatial knowledge transfer is to considerthe conduits (texts, people or artifacts) throughwhich it is exchanged. The so-called knowledgespillover literature focuses on the transfer oftexts (Jaffe et al. 1993; Audretsch and Feldman1996; Breschi and Lissoni 2001). These studiesexamine patent data, by looking for patterns ofknowledge exchange. Howells (2002, 876) takesissue with the conceptual inferences derived fromthis type of data analysis, arguing that patent ci-tations are a metric that, ‘impl[ies] the impartingof knowledge, but do[es] not actually measure it’.Absent in the discussion of knowledge spilloversare discussions of how patents are produced—the context of their authorship, the geographiesof their circulation and their efficacy for shapingfurther innovative practice.

For the second conduit, people, Pinch andHenry’s (1999; Henry and Pinch, 2000) work onthe British Motor Sport Industry (BMSI) clusterhas tracked the movements of talent amongst en-terprises. Their analysis highlights the turnoverof engineers among competing teams. It furtherillustrates how knowledge was translated acrosssectoral boundaries to produce design innova-tions in the cars themselves—an element that re-curred in the present study. Nevertheless, theiremphasis on the exchange of tacit knowledge em-bodied in individuals hinders insight into the fullspectrum of the knowledge cycle. A concept thatmight lend traction here is the notion of ‘knowl-edge enablers’, that is, agents usually operatingwithin corporations, who facilitate the diffusionof tacit knowledge (von Krogh et al. 2000, cited inGertler 2004). The challenge for these individuals,as Gertler (2004, 146–147) notes, is that it is ‘dev-ilishly difficult to disseminate’ this knowledgewithout ‘at least partial codification in the pro-cess of transmission’. In other words, the move-ments of people and texts need to be viewed asbeing complementary to each other.

Finally, technology transfer occurs through themobilization of artifacts, such as consumer orcapital goods. Especially in the case of complexmachinery, Gertler has repeatedly (1995, 2001)

called into question the limits to the transfer ofthese technological artifacts, a point commonlymade about technology transfer by developmentscholars. On the other hand, reverse engineering,a central plank in the strategy of Japanese post-war reindustrialization (Freeman 1987; Partner1999; Takahashi 2000), illustrates how artifacts,in concert with texts and expert advice, mobilizeknowledge between places. This example affirmsboth the inadequacy of theorizations that treatpeople and things as operating in mutually exclu-sive domains and the need to look beyond NorthAmerica and Europe. These issues are addressedin the communities of practice literature and byactor–network theory.

Communities of practice and actor–networks

Communities of practice are intra- and inter-organizational problem-solving networks (Wenger1998). Debate in geography questions whetherthese communities effectively mobilize knowl-edge between locations. Amin and Cohendet(1999) assert that the organizational and rela-tional closeness that define communities of prac-tice substitute and even supersede physicallyproximate relations. Gertler (2001), however, ex-presses skepticism with the fact whether so-cial or institutional proximities can really over-come distance. Coe and Bunnel (2003, 446) stakeout an intermediate position in reasoning that,‘while such communities will originally almostcertainly be local configurations, over time sus-tained and repeated interaction facilitated by“boundary crossers” may create new spatially ex-tensive constellations’. Boundary crossers sounda lot like the knowledge enablers discussed ear-lier, suggesting, in part, that we are workingthrough a conceptually cluttered space. In its em-phasis of shared repertoires and stories of com-munity participants, this literature is still primar-ily working through the idea of knowledge in thetacit dimension. This emphasis stands in contrastto actor–network theory.

Actor–network theory and the ‘science stud-ies’ literature of Bruno Latour (1987, 1999) de-velop a conceptual framework and vocabularyto interpret the circulation of scientists and en-gineers through society. Actor–networks are socalled since their architecture derives from theability of humans and non-humans (machines,



books, etc.) to enroll new actors into the net-work. Importantly, this perspective solves manyof the difficulties inherent in the compartmental-ized knowledge conduits approach since it viewstexts, people and artifacts as mutually constitut-ing entities. It further opens up new possibilitiesfor refining our understanding of the geographiesof tacit and codified knowledge. Central to thisperspective is the idea of ‘translation’, or the no-tion that within networks, actors are forced to‘take detours through the goals of others’ (Latour1999, 89). These ‘translations’ are the displace-ments in meaning and objective that result fromthe inevitable compromise among disparate in-terests. Within geography, various authors haveillustrated the utility of actor–network and sci-ence studies’ perspectives (Murdoch 1994; deLaet2000; Barnes 2001; Winder 2001; Evenden 2004).Actor–network theory and translation also haveparticular reference to the practice of patentingand the detours that engineers are forced to taketo authorize their inventions.

Patents and the inscription of actor–networks

In contrast to approaches that interpret patentsas a proxy for innovation (Pred 1966; Schmookler1966; Mensch 1979; Freeman et al. 1982; Ceh2001), actor–network theorists begin from the as-sumption that patents are problematic. Their con-cern is with patents as texts, and the practiceof patenting as a situated social process. Patentsare instruments that actors use to ‘mobilize theworld’ (Latour 1987, 223) in order to translatethe interests of inventors and their assigningfirms into the public realm. At sophisticated lev-els of strategic practice, patenting as a formof codification aims to snooker the competitionby making it difficult for other parties to closethe circuit between codified and tacit domains(Grandstrand 1999). Patents map new vistas oftechnological space at the same time as theyenact a property claim that says, in effect: donot enter . Before reaching this plateau, however,patents must be carefully drafted through suc-cessive rewritings and demonstrations of origi-nality before various audiences (peers, investors,examiners)—in other words, through a cycling be-tween tacit and codified domains of knowledge.Myers (1995, 101) provides an excellent illustra-tion of this process as he traces the procedures

through which two scientists translate their jour-nal articles into patent documents. In doing sothey must negotiate the boundaries between ‘theacademic world of making knowledge and the le-gal world of owning things’. Patents further oper-ate within a spatial topology distinguishable fromother inscriptions. On the one hand, as deLaet(2000) suggests, patents conform to what Latour(1987) terms ‘immutable mobiles’, or a particu-lar type of instrument that maintains its formin various contexts. This quality enables firms toact from afar, albeit indirectly through the legalsystem, to discourage infringement. On the otherhand, this conceptualization is problematic, be-cause the property rights associated with patentsare less mobile than the inscriptions themselves.Consequently within and among various contexts,the ‘tenacity’ of patents may have as much to dowith the intrinsic quality and originality of thetechnical claims as it does with the strength ofthe network that upholds these claims (deLaet2000). Japanese industry, particularly since the1970s has demonstrated an almost singular ca-pacity for underwriting its economic and techno-logical development with a sophisticated nationalapproach to the mobilization of IP (Odigiri andGoto 1993).

Several studies have argued that the Japaneseapproach to patenting is predicated on a logicwhereby the rewards for invention are measuredin terms of social rather than individual bene-fits (Rosen and Usui 1994; Takenaka 1994; Kotlerand Hamilton 1995; Granstrand 1999). The sys-tem within Japan has been designed to allowrival firms access to each other’s patents soonafter application with the purpose of improvingthe overall level of technological learning. In con-cert with a national innovation system (Freeman1987) that encouraged technology imports, li-censed or otherwise (Prestowitz 1989; Partner1999), the value placed on knowledge codificationat home gave Japanese industry a boost in inter-national competition. At present five of the topten firms awarded patents by the U.S. Patent andTrademark Office are Japanese (McKenna 2006).Granstrand (1999) discusses the strategies thatproduce such outcomes. The first involves gain-ing access to competitors’ technological break-throughs by ‘surrounding’ them with one’s ownpatents, thereby engaging the competitor to seekcross-licensing agreements. The second strategy



is to obscure an enterprise’s own technologicalpriorities by ‘blanketing’ the technological spacearound key patents with minor derivative and an-cillary patents (Granstrand 1999, 220). Throughthe recognition of these practices it is possible tomake sense of Japan’s high patent propensity invarious fields including EMIs. However, to solelycredit the role corporate strategy plays in Japan’spost-war industrial ascendancy, overlooks the fewbut significant episodes through which key ideasproduced elsewhere were translated into the IPportfolios of Japanese firms. Indeed, by workingbackwards from the patents awarded to any firm,Japanese or otherwise, it is possible to produce arather different narrative in which the credit forproprietary knowledge is less stable and more de-pendent on a broader network of actors than theassigning or licensing corporation.

Research Design

Methodologically, understanding how patentingand other processes of codification unfold inspace requires tracing the actor–networks of in-ventors, firms, the patents they register and ar-tifacts that follow. Those who work with patentsare immersed in the tacit practicalities of writingand rewriting novelty. Yet, these actions producea signature that enables comparison of the recordof invention, the patent document, with other bio-graphical and less objective accounts. With patentdatabases available in digital form, the record ofcodification is relatively straightforward to ob-tain. To evaluate the geography of patent registra-tion, as a proxy measure for regional innovativeoutput, I conducted an ‘all field search’ onthe U.S. Patent and Trademark Office’s Database(www.uspto.gov) using the term ‘electronic musi-cal instrument’. The data set consisted of all U.S.patents (n = 2,375) published between 1965 and1995. Patents published after 1975 electronicallylist all the patents that reference that document.For those patents published before 1975, I had toundertake the time-consuming process of work-ing backwards from existing patents to see whichpatents were cited as prior art. These tallies wereused to construct lists of influential, widely citedpatents. It was this process that produced theinventors and firms who appeared to exert thegreatest influence on the technological trajectory

of the industry. At this stage the methodologicalapproach switched to that of biography.

Various authors have sought to problematizethe role of biography in economic geography(Barnes 2001; Schoenberger 2001). Biographies, asBarnes (2001, 415) notes, ‘are a melange of factand fiction—of events that happened, but alsoof rhetorical strategies, vested interests, and actsof interpretation, selective memory, and wish-ful thinking’. They connect career trajectorieswith corporate or other institutional interests andcontextualize the harmony and dissonance thatarise between the two. Cases of patent infringe-ment bring into relief the uneasy tension betweenpeople, texts and artifacts noted by Howells(2002). Barnes (2001) offers further advice in sift-ing through the multiple worlds that constitutepeoples lives. The biographer must first nego-tiate between the individual and their context(agency/structure). Moreover, there needs to berecognition that biography is the re-invention oflives in a manner that Livingston (1999; quotedin Barnes 2001, 415) calls ‘controlled fiction’.Third, the position and intention of the authormust also be clear. Barnes’ analysis of the actor–networks that produced geography’s quantitativerevolution is instructive because it situates theproduction of texts in the context of the livesof their authors. This paper similarly embeds theinscriptions of inventors within their broader ca-reer biographies in order to illustrate what fac-tors served to unmoor their contributions fromtheir place of origin. The author is a studentof industry who translated a practical interest inmusical instruments into his doctoral thesis ingeography. This article details a portion of thatresearch project.

Codification and the Transfer of EMITechnology from the United Statesto Japan

Japanese firms headquartered in the city ofHamamatsu dominate the global musical instru-ment industry. In terms of sales, Yamaha, Kawaiand Roland2 sit one, two, and three atop theleague tables (The Music Trades 2000). It is rare

2 Roland is nominally headquartered in Osaka, although thefirm’s principal administrative units and factories are inHamamatsu.



to find examples, perhaps with the exceptionof Detroit during high-Fordism, where an indus-try’s power is so spatially concentrated in oneplace. In the field of synthesizers and other EMIthis locus of dominance extends to the Tokyo-based manufacturer Korg, which since 1988 hasbeen 51 percent owned by Yamaha. It is curi-ous that in the burgeoning list of English lan-guage accounts detailing technological evolutionin the EMI industry (Chadabe 1997; Theberge1997; Pinch and Trocco 2002), there is scant men-tion of the agency of Japanese firms. Instead, theemphasis is on the role of key centres like BellLabs and Stanford University, institutions such asthe Audio Engineering Society (Reiffenstein 2005)and inventors such as Bob Moog who rank asthe pioneers of the analog era. Indeed, after dif-fuse early developments in places like Russia andGermany, America had emerged by the 1950s asthe crucible of electronic music, a position it heldthrough the 1970s and then promptly lost to theJapanese. Apart from elegies to the demise ofthe once great U.S. EMI industry (Milano 1993)there is a critical gap in our understanding ofhow the technological transition from analog todigital sound is linked to the geographical shiftin the industry’s center of gravity from the UnitedStates to Japan.

The Japanese literature, on the other hand, of-fers several perspectives on the industry’s evolu-tion in that country. These include stories toldby the firms themselves as corporate histories(Yamaha 1987; Kawai 1997; Roland 1999), auto-biographies of firm founders (Kakehashi 2002),books written by leading engineers (Mochidaand Aoki 1994) and other interpretive accounts(Nakagawa 1984), including research reports pub-lished by geographers (Ohtsuka 1980; Takeo1988). Notwithstanding these contributions, withthe exception of Johnstone (1994, 1999), there isa dearth of research that explains the role thattechnology transfer from the United States playedin the ascendancy of Hamamatsu’s musical in-strument industry, and it is this lacuna which thispaper addresses. The investigation begins with anassessment of the patent record.

The lessons and limits of patent data

From the early 1970s, Japanese firms, especiallythose located in Hamamatsu, had eclipsed their

American counterparts in the registration of U.S.patents (Table 1). Indeed by the 1990s, the for-mer industrial core of the Midwest, home to suchformerly venerable instrument manufacturers asWurlitzer, Hammond,3 Lowery and Kimball, hadceased to be a factor. In contrast, these descrip-tive statistics confirm that Japanese instrumentmanufacturers placed an increasing strategic pre-mium on the registration of intellectual proper-ties with the U.S. Patent and Trademark Office.What they do not indicate, and herein lies thelimitations of patent counts, is the range of prac-tices that lies behind the figures. They tell uslittle about how corporate and individual inven-tor biographies converge in the patent domain—how texts are rendered and how these texts begetother texts, producing what Latour (1987) terms‘chains of translations’. The following empiricalmaterial is presented to show the relative rolesof codification and tacit knowledge in the trans-fer of innovative power from California to Hama-matsu, Japan. This will be accomplished throughcase studies of three key EMI inventor/engineersin the 1970s and 1980s.

Career Biographies and the Processof Codification

The remainder of the paper presents the careerbiographies of three Californian engineers whoseinscriptions enabled the translation of knowledgeto the Japanese firm Yamaha. The engineers areRalph Deutsch, Dave Smith and John Chown-ing. These engineers connect the demise of theUnited States as an industrial core to the ascen-dancy of Japanese firms. Indeed in all three cases,potential U.S. allies balked at the opportunityto enroll the engineers in a domestic network.That these individuals and their milieu failed toconnect is especially surprising since the inven-tors were working in institutional settings—theSouthern California aerospace industry (Deutsch),Silicon Valley (Smith) and Stanford University(Chowning)—that are normally associated with

3 Interestingly, the Wurlitzer and Hammond names live on asdigital sound ‘patches’ on synthesizers. My Korg synthesizeroffers, respectively, ‘Amp Driven Wurly’ and ‘Rotor Organ’files that evoke the analog tones of these vintage instruments.The Hammond Organ brand was acquired by Hamamatsu-based Hammond-Suzuki in 1988.



Table 1U.S. patents for electronic musical instruments by region of assigning company 1965–1994

U.S. U.S. U.S. Japan Japan Japan SouthEurope East Midwest West Kanto Hamamatsu Kansai Korea Total

1965–69 18 46 138 24 4 9 3 0 2421970–74 11 62 149 40 13 177 12 0 4641975–79 9 34 91 20 9 150 23 0 3361980–84 23 47 110 17 38 239 18 0 4921985–89 15 19 12 16 73 177 8 0 3201990–94 2 10 3 23 109 353 14 7 521

Total 78 218 503 140 246 1,105 78 7 2,375

SOURCE: USPTO (http://www.uspto.gov); last accessed 21 April 2006.U.S. East are those states bordering the Atlantic, U.S. West include the Pacific States and Arizona, while the U.S. Midwest are all the rest, butprimarily Illinois, Ohio and Indiana. Japan Kanto includes Tokyo, Saitama and Kanagawa prefectures, Japan Kansai includes Osaka, Kyoto andHyogo prefectures, while Japan Hamamatsu records those patents registered in a concentrated area of western Shizuoka prefecture surroundingthe city of Hamamatsu.

California’s post-war ascendancy as a center oftechnoscience. In contrast, Yamaha proved eagerto wed its interests with these U.S. engineers, tomobilize the latter’s inventions by licensing, as-signing or citing their inventions. Against thesesimilarities, the three inventors differ in their ap-proaches to codifying the knowledge they pro-duced. Ralph Deutsch, according to his tally ofpublished patents, was the most prolific inven-tor in the EMI field, John Chowning recorded onehighly significant patent while Dave Smith neverregistered a single patent.

In constructing these narratives I relied on myown interviews with Ralph Deutsch and DaveSmith in addition to secondary sources that in-clude firm histories (Yamaha 1987; Markowitz1989), interviews and oral histories containedin trade publications (Vail 1993) and other in-terpretive accounts (Nakagawa 1984; Johnstone1999).4 These multiple biographical sources serveas a counterpoint to reflect on the publicaccomplishments of these individuals (patents,instrument designs, etc.) while illustrating thesocial (as opposed to technical) dimension ofinnovation.

4 Comprehensive accounts of Chowning’s contribution to EMIand Yamaha are detailed in Johnstone (1994, 1999). Conse-quently, he was not interviewed.

Vignette I: Ralph Deutsch—knowledgemobilization through the reinscriptionof one’s own idea

The outline of Ralph Deutsch’s contribution tothe evolution of EMI is as follows. Throughthe 1970s and 1980s Deutsch was one of themost prolific inventors in the field of EMI, reg-istering more than 140 patents. During its life-time, Deutsch’s ‘digital organ’ patent (U.S. Patent3515792), the first patented rendering of a musi-cal tone in binary code, was cited more than 114times in subsequent patents, more than any othersimilar document (Table 2). He authored four ofthe top ten most cited patents in the field. Theseaccomplishments, however, have been overshad-owed by a restless career the shifts of whichhave proven highly controversial. Several U.S. andJapanese firms, among them, North AmericanRockwell, Yamaha and Kawai have sponsored hisinventions (Table 3).

The story that animates these facts is compli-cated by conflicting vantage points. Details thatDeutsch narrated to me sit uneasily alongsidethe highly critical account of Deutsch found inthe autobiography of Jerome Markowitz (1989),President of Allen Organ, the Pennsylvania firmthat licensed the digital organ technology fromDeutsch’s employer, North American Rockwell.In reference to Deutsch, the dust jacket ofMarkowitz’s (1989) Triumphs and Trials of an



Table 2The top ten most cited U.S. patents for EMI, 1965–1994

Inventor nameYear Cites Assignee name Assignee address Assignee sector Inventor address U.S. patent number

1970 114 North American Rockwell Pasadena, CA Aerospace Sherman Oaks, CA Deutsch, R.3515792

1971 111 North American Rockwell Pasadena, CA Aerospace Tustin, CA Watson, G.3610799

1974 91 Deutsch Research Labs Sherman Oaks, CA Musical Instruments Sherman Oaks, CA Deutsch, R.3809786

1991 80 Stanford University Stanford, CA Education Menlo Park, CA Smith, J.4984276

1993 66 Pioneer Tokyo, JP Electronics Tokyo, JP Okamura, M., et al.5247126

1975 62 Yamaha Hamamatsu, JP Musical Instruments Hamamatsu, JP Tomisawa, N., et al.3882751

1977 58 Stanford University Stanford, CA Education Palo Alto, CA Chowning, J.4018121



1967 50 Seeburg Chicago, IL Jukebox Gilford, NH Campbell, R.3358068

1983 49 Yamaha Hamamatsu, JP Musical Instruments Hamamatsu, JP Nagai and Okamoto4383462

SOURCE: USPTO (www.uspto.gov); last accessed 21 April 2006.

Organ Builder highlights, ‘How one of the keyfigures in the development of digital musicaltechnology sold out to a leading Japanese com-pany’ (emphasis added). One of Markowitz’schapter titles is: ‘Ralph Deutsch and the DarkSide’. Not surprisingly, Japanese authors (Nak-agawa 1984) paint a different picture. In piec-ing together this legacy, the challenge is one ofachieving balance between the hard facts and

Table 3A synopsis of Ralph Deutsch’s patenting record

Number of Number of patents Legal representationAssigning firm Dates patents with co-inventor (Patent agent/attorney)

N.A. Rockwell 1970–1974 9 1a Haman & HumphriesDeutsch Research Labs 1974–1979 17 5b 1)Howard Silber (Flam & Flam);

2) Ralph DeutschYamaha 1974–1977 20 3c Howard Silber (Flam & Flam)Kawai 1977–1989 94 23b Ralph Deutsch4 firms 19 years 140 patents 32 co-invented 3 different legal representatives

a George Watson, employee of Autonetics Division Rockwell.b Leslie Deutsch, son of Ralph Deutsch.c Glen Griffith, Tustin, CA.SOURCE: USPTO (http://www.uspto.gov); last accessed 21 April 2006.

claims of Deutsch’s prodigious patent record,Deutsch’s actions and the multiple interpreta-tions of those actions.

On 8 August 1967, Pasadena, CA-based NorthAmerican Rockwell (hereafter Rockwell) and theirengineer Deutsch filed a patent application forthe digital organ. Soon after, Deutsch demon-strated this invention to the large U.S. organmanufacturers, almost all of whom failed to



appreciate its significance, or balked at the po-tential development costs. Markowitz’s firm AllenOrgan proved to be the solitary interested party.Rockwell licensed the technology to Allen andDeutsch headed the team of engineers respon-sible for developing and delivering the digitalcircuit boards to Allen. Unfortunately, Rockwellan aerospace firm, and Allen an organ manufac-turer, perceived the technological horizon forgedthrough their partnership in fundamentally dif-ferent ways. In drafting the contract Markowitz(1989, 77), ‘expressed [Allen’s] requirements inthe technical language familiar to us and Deutsch‘translated’ these requirements into the ‘for-eign’ language of the new technology’. Markowitz(1989, 70) dubbed this language ‘Anaheimtech-speak’.

This linguistic detour produced a compromise.Markowitz was impressed with the digital cir-cuitry of the prototype but dissatisfied with itstonal expression. Beyond these technical con-cerns, personal differences between Deutsch andMarkowitz almost scuttled the relationship. Nev-ertheless, in June 1971 Rockwell delivered thedigital organ prototype to Allen and its techni-cal attributes proved immediately apparent. Forexample, Allen’s analog organs took one week totest whereas the circuitry of the digital organscould be tested in under an hour (Ralph Deutsch,personal communication, 12 October 2002, con-versation). More importantly, Allen’s license offirst refusal to exploit the digital organ technol-ogy proved vital to the firm’s long-run fortunes.Competitors who sought to enter the digital arenahad little choice but to sign licenses with Allenor infringe on the patent. Largely due to the roy-alties derived from this patent, Allen survivesas the only major organ company that is stillAmerican-owned.

At the North American Music Merchants(NAMM) trade convention in June 1971, Deutscharranged a private demonstration of Allen’s digi-tal organ to Yamaha’s president and an entourageof 25 delegates. In January 1972, Deutsch leftRockwell and a few months later was hired byYamaha for a 3-year contract to produce tech-nologies for their Electronic Organs (Electones).Deutsch headed the New Electone System (NES)division, one of two internal teams created atYamaha to develop instruments for the digitalage. While the NES initiative was geared towards a

digital solution to the problem of tone synthesis,a parallel program operating out of the newlyconstructed Toyooka factory focused its effortson an analog/digital hybrid system known as PAS(passive analog system). These two teams devel-oped a healthy rivalry whereby

Deutsch’s appearance stimulated a competitivespirit at Toyooka’s PAS development centre andalso raised the aggregate power for developing thelogic circuit which had thus far proved to be amajor hurdle . . . The PAS team had the spirit anddetermination that it was not acceptable to loseagainst Deutsch (Nakagawa 1984, 86–87; transla-tion by the author).

After a year and a half of this bifurcated devel-opment, an internal trial was staged in front ofYamaha’s President Kawakami Gen’ichi5 who se-lected the PAS system. Deutsch (personal com-munication, 12 October 2002, conversation) sum-marized this setback by noting that Yamaha onlypartially incorporated the technologies he devel-oped and patented on their behalf. He was alsocritical that his remuneration package did not in-clude royalties accruing from any of his patents.

During the Yamaha contract, Kawai approachedDeutsch to undertake a similar function with re-gard to that firm’s digital organ program. Thisturn of events, not surprisingly, proved contro-versial, and Yamaha threatened to sue Deutschfor breach of contract. Deutsch defended him-self by citing the non-exclusive nature of theircontractual obligations. Moreover, as he claimed,‘there was nothing worthwhile [in our relation-ship] that was not covered by the patents’ (RalphDeutsch, personal communication, 12 October2002, conversation). The knowledge Deutsch cod-ified for Yamaha, in this sense, was absolutelycrucial both in terms of his remuneration andin terms of the firm’s ability to proprietizehis contribution. However, to suggest that thepatents could operate independent of Deutsch’stacit interpretation of their contents is indeed astretch—it is the cycling between the two that ul-timately mattered. Put another way, it is difficultto suggest any action by Deutsch that was not insome way tethered to the patents he authored.The following episodes illustrate this point.

5 Japanese names are rendered according to the conventionwhereby the family name precedes the given name.



How did codification facilitate Deutsch’s breakfrom the Rockwell–Allen partnership and thetransfer of knowledge to Yamaha? Or, as theauthor of one of the most fundamental tech-nologies in the field, could he separate his owntacit knowledge that formed the basis for thedigital organ patent from the codified knowl-edge inscribed in that patent? In early 1972, af-ter leaving Rockwell but before joining Yamaha,an independent Deutsch Research Labs appliedfor a ‘Computor Organ’ (sic) patent (U.S. Patent3809786), the third most cited patent in the field(Table 2). This patent claimed to ‘exhibit . . . all ofthe listed advantages of digital wave shape gen-eration’, a nod to his own Rockwell–Allen dig-ital organ patent (U.S. Patent 3515792), ‘yet ina manner totally different from that known inprior art’. The latter invention’s tonal quality was‘determined by the stored harmonic coefficientvalues’ as opposed to the ‘actual storage of adigital representation of a waveshape’. Notably,it was preferable that ‘the harmonic coefficientsare stored in digital form and the computationsare carried out digitally’. The extent to whichU.S. Patent 3809786 had its basis in U.S. Patent3515792 and in Deutsch’s work for Rockwell–Allen is far from straightforward as evidenced bywhat followed.

Deutsch, R&D staff in the Electone division atYamaha and eventually Deutsch and engineersin the R&D division at Kawai soon published abattery of ancillary and dependent patents thatlaid claim to a large portion of intellectual spaceand served to discourage rivals form enteringthe increasingly digital realm of EMI. Many ofthese documents referenced either the digital or-gan patent or its cousin the ‘computor’ organ.Complicating matters, Yamaha instruments thatseemed dependent on these technologies startedappearing in the marketplace. Allen’s objectionsto these actions resulted in initial and retalia-tory lawsuits not only between Allen and Yamaha,and Allen and Kawai, but also between Allen andother American organ makers who also appearedto be infringing. By Markowitz’s (1989, 124) reck-oning, the computer organ patent, and muchof Deutsch’s subsequent work for Yamaha, con-travened the Allen–Rockwell contract. Eventually,Yamaha and other Japanese firms, in recognitionof Allen’s position, settled. These lawsuits under-line the significance of the disputed knowledge

and the inseparability of technological and in-scriptive dimensions of industrial novelty.

The lawsuit between Indiana-based Kimball In-ternational, Inc. and the Allen Organ Companylasted between 1981 and 1988—the latter accus-ing the former of infringement6 and the for-mer contending the latter’s patent was invalid.7

Much of Kimball’s claims of invalidity rested onthe fact that Deutsch and Rockwell had ‘demon-strated’ its digital organ technology to Rodgersand Conn (two other U.S. manufacturers) morethan a year before it filed for the digital or-gan patent. The act of demonstration, Kimball ar-gued, implied that this knowledge was alreadyextant, or in the terms of patent law in ‘publicuse’, a condition which if proven would invali-date the patent. Eventually, the jury granted Kim-ball a pyrrhic victory. Adrift in a sector wheresuccess was increasingly contingent upon owningintellectual properties for digital, as opposed toanalog, technologies, Kimball like most other U.S.EMI manufacturers soon went out of business. In-scriptions such as key patents both tie togetherand work to modify the behaviour of various ac-tors in innovation networks. Indeed, it is fair todescribe this episode as illustrative of a commu-nity of practice in which the members, to useWenger’s (1998) phrasing, are ‘passionate’ abouttheir vocation. In this case, though, the actionsof the U.S. EMI manufacturers proved creativelydestructive for one another.

The story continues, for U.S. Patent 3515792(p. 22) met with critical reappraisal when ananonymous competitor petitioned the U.S. Patentand Trademark Office in 1985 to re-examineits claims. The result of this scrutiny was thatin 1987, one year prior to the expiry of thepatent, every one of its 46 claims were cancelled.Someone had signalled to the examiner a farmore extensive body of prior art than was con-tained in the original patent document. When itwas granted in 1970, Deutsch acknowledged fivepatents as prior art; upon re-examination, 104

6 Allen Organ Co. v. Kimball International Inc., Nos. 86-767,86-789, UNITED STATES COURT OF APPEALS FOR THE FED-ERAL CIRCUIT, 839 F.2d 1556; 1988 U.S. App. LEXIS 1763; 5U.S.P.Q.2D (BNA) 1769, February 12, 1988, Decided.

7 Kimball International Inc. v. Allen Organ Co., No. EV 78-73-C,UNITED STATES DISTRICT COURT FOR THE SOUTHERN DIS-TRICT OF INDIANA, EVANSVILLE DIVISION, 1981 U.S. Dist.LEXIS 17329; 212 U.S.P.Q. (BNA) 584, May 1, 1981.



patents and 32 other documents were cited asprior art. Articles published in the Journal of theAudio Engineering Society (AES), technical reportspublished by Bell Labs and other documents werebrought forward to suggest that Deutsch’s digitalstep was not as radical as claimed. Deutsch ad-mits in hindsight that there are conceptual sim-ilarities between his early work and that of MaxMatthews of Bell Labs (Ralph Deutsch, personalcommunication, 12 October 2002, conversation).Nevertheless, for close to 15 years, the musicalinstrument industry operated under the assump-tion that the Rockwell–Allen patent defined thestate of things to come. In this sense the patent’s‘tenacity’ (deLaet 2000) proved radical even if thetechnology amounted to the most officially (if in-correctly) sanctioned articulation of the state ofthe art at the dawn of the digital frontier. Thatsuch a significant patent was awarded with whatin retrospect appears to be only a cursory exami-nation is by no means a unique occurrence (WallStreet Journal 2006).

Deutsch’s career path traces a geographythrough a number of key firms in both the UnitedStates and Japan. As framed by Markowitz, thismigration was a matter of ‘selling out’. Fromthe Japanese standpoint, Deutsch served as acatalyst to an instrument development programthat was far more sophisticated than any firmin the United States—witness Yamaha’s decisionto make its own microchips (Nakagawa 1984).8

Deutsch’s contribution goes far beyond these mi-grations. Patents bearing his name as the inventor(140 in total), and notably U.S. Patent 3515792,cast a long shadow over the trajectories of bothhis assigning firms and those companies forcedto navigate the technological space demarcated

8 Yamaha had originally sought to buy generic chips from NECand Hitachi, but these proved inadequate for their needsso the firm set about to produce application-specific chips.They dispatched their best and brightest engineers to train atthe laboratory of the leading Japanese scientist in the field,Nishizawa Junichi of Tohoku University. Armed with this prac-tical knowledge, the cohort returned in 1969 to oversee theset-up of Yamaha’s Toyooka laboratory and factory outsideHamamatsu. By 1976 Yamaha was mass-producing chips atToyooka and at a second factory in Kagoshima (Nakagawa1984). By the 1990s, Yamaha’s specialization paid off. In 1994,Yamaha’s sound chips accounted for 95 percent of sales ofthe $1billion market for sound boards (Johnstone 1994). How-ever, in 2003, Yamaha sold its integrated circuit division tothe Kyoto-based Rohm Corporation.

by these documents. In particular, the tenacity ofthe digital organ patent shaped the actions of ri-val firms, and the actions of Deutsch when hewent to work for those rivals. Moreover, his storycaptures the process that Nonaka and Takeuchi(1995) refer to as the ‘cycling’ between tacit andcodified forms of knowledge. For instance, perti-nent court cases turned on the scrutiny of cod-ified knowledge (specifically the claims in thepatents) relative to Deutsch’s actions (demon-stration). Deutsch’s decision to become his ownpatent agent (Table 3), to advocate on behalf ofhis inscriptions reinforces the idea that there is atacit dimension to the process of codification.

Vignette II: John Chowning—knowledge cyclingand the genesis of the Stanford–Yamaha license

The case of John Chowning and the FM syn-thesis patent is another story of spatial knowl-edge transfer that links a key patent authoredby a Californian inventor with the Japanese firmYamaha (Johnstone 1994, 1999). In the early1960s, Chowning was a graduate student of mu-sic composition at Stanford University. Inspiredby a paper published by Bell Labs’ Max Mathews(1963) in the journal Science, Chowning shiftedhis course of study to computer music. He vis-ited Bell Lab’s in New Jersey, met his guru andreturned to Stanford with Mathews’ computermusic program. He enrolled in computer sciencecourses and spent 5 years at Stanford’s ArtificialIntelligence Laboratory connecting the theory ofcomputer music to its practical application.

One night in late 1967 while tinkering with apair of oscillators, using one to control the pitchof the other, he made an ‘ear discovery’.

At a frequency of around 20 Hz, he noticed that in-stead of an instantaneous change in pitch from onepure tone to another, a recognizable tone color,one that was rich in harmonics, emerged fromthe machine. It was a discovery that an engineerwould have been unlikely to make. What Chowninghad stumbled upon, it later turned out, was fre-quency modulation—the same technique that radioand television broadcasters use to transmit noise-free signals. Of this, the composer was blissfullyignorant: All he wanted to do was make colorfulsounds. Chowning began tweaking his algorithmand pretty soon, as he recalls, ‘using only two



oscillators, I was making bell tones and clarinet-like tones and bassoon-like tones, and I thought,you know, this is interesting’ (Johnstone 1994, 3).

Chowning by this stage was a junior facultymember in the music department. Unfortunately,there were few colleagues capable of confirmingthe radical nature of his discovery.9 Part of theproblem is that Chowning had yet to ‘connecthis ear to the theory’ (Johnstone 1999, 217). AsJohnstone (1999) explains:

This conjunction would not occur until 1971,when Chowning remembered some synthetic trum-pet tones that a Bell Labs researcher had playedfor him and wondered whether he could indeedachieve a similar effect using FM synthesis. Itturned out that he could indeed produce somequite realistic brass tones. It was at this point thatChowning realized that his technique was a lotmore than he had first thought.

In order to realize the potential of his inven-tion, it was necessary for Chowning to enlist oth-ers with a similar tacit knowledge base into theensemble. When Chowning presented his discov-ery to the staff at Bell Labs, its significance wasindeed apparent, especially to Max Mathews’ bossJohn Pierce who instructed Chowning to ‘patentit!’ In other words, Pierce commanded Chowningto both protect his invention and complete theknowledge cycle from its tacit state back to aform that could be translated to others. SinceStanford was not in the business of manufactur-ing, this idea had to be pitched to those whocould put Chowning’s recipe to use.

With echoes of the cold shoulder given toRockwell and Deutsch by all of the U.S. industrysave Allen Organ, in 1971 Chowning encountereda similar inability of U.S. EMI manufacturers (in-cluding the firms Allen, Hammond and Lowery)to comprehend what was shown to them. De-spite Chowning’s entreaties, he failed to align hisinvention with the technical aspirations of anydomestic firm. Hammond Organ of Chicago, forinstance, responded to Stanford’s invitation bysending four engineers and a professional organ-ist. While the latter was impressed with what he

9 Phil Lesh, avant garde composer, and Grateful Dead bassplayer, was one contemporary who recognized and supportedChowning’s achievement (Johnstone 1994).

heard, the engineers failed to recognize that thistechnology heralded the dawn of a new fron-tier. Here are Chowning’s thoughts about thatmeeting:

They kept asking me about how many pins it wouldneed. Well, I didn’t know anything about pins,chips or analog circuitry at all so I couldn’t answerthem. I said, ‘Look, it’s an algorithm and here’s thecode [but] it was simply not part of their world’(Johnstone 1999, 221; emphasis added).

This is a good illustration of what Asheim(1999) calls ‘disembodied knowledge’ for Chown-ing’s algorithm was a text that proved opaqueto those outside a community of practice withan insufficient stock of tacit knowledge. Yamaha,on the other hand, sent an engineer, IshimuraKazukiyo, with a background in communicationstechnology. Ishimura, in Chowning’s words, ‘tookall of ten minutes’ (Johnstone 1994), to grasp thetechnology and immediately initiated negotiationsto take out an exclusive license on the patent.From the standpoint of Ishimura’s boss, MochidaYasunori, FM synthesis dovetailed perfectly withYamaha’s objectives.

As an engineer, you are very lucky if you encountera simple and elegant solution to a complex prob-lem, FM was such a solution and it captured myimagination. The problems of implementing it wereimmense, but it was such a wonderful idea that Iknew in my heart that it would eventually work(Johnstone 1994, 5).

If it was tacit knowledge that enabled Ishimura tograsp the import of Chowning’s demonstration,Yamaha’s costly appropriation of the technol-ogy had to be defendable through an exclusivelicense.

In 1983, 10 years after they started devel-oping applications based on Chowning’s inven-tion, Yamaha released the DX7, the world’s first,all digital programmable, polyphonic synthesizer.Pinch and Trocco (2002, 317) regard the DX7as the ‘breakthrough digital instrument, the firstone to achieve commercial success’. Whereasthe signature synthesizer of the early 1970s, theMinimoog, achieved lifetime sales of 12,000, theDX7 sold over 200,000 units in 3 years (Colbeck1996).

For Stanford, Silicon Valley’s nearest and mostimportant institutional neighbor and a university



Table 4Technological relations binding Stanford University and Yamaha

Date of invention(of agreement) Technology Details of Stanford–Yamaha relationship

1968 (1975) FM synthesis—invented by JohnChowning

Yamaha takes out an exclusive license on the FM technology.

1985 (1989) Waveguide synthesissampling—invented by Julius O.Smith III

Waveguide synthesis is licensed in a non-exclusive fashion to Yamaha and Californiafirms such as Chromatic Research, Crystal Semiconductor and Seer Systems

(1997) Sondius-XG Joint licensing agreement to cooperatively promote the development and use oftheir respective intellectual property portfolios (over 400 patents andapplications) in the computer tone generation and sound synthesis areas. TheSondius program is Stanford’s first effort to not only patent discoveries made oncampus but to trademark them as well.

SOURCE: Johnstone (1999); http://www.sondiusxg.com/presrelease.html; last accessed 21 April 2006.

synonymous with academic–industry collabora-tion, Chowning’s invention has become thesecond-highest generator of royalties for its Of-fice of Technological Licensing (Johnstone 1999).Stanford’s relationship with Yamaha has per-sisted to the mutual enhancement of both parties(Table 4). Most importantly, the royalties fromChowning’s FM patent funded the establishmentof Stanford’s Center for Computer Research inMusic and Acoustics (a.k.a. ‘Karma’). CCRMA cur-rently employs 14 research and teaching facultyincluding John Chowning and Max Mathews. Suc-cessive generations of key basic technologies thatinclude Julius Orion Smith’s wave guide samplingpatent (Table 2, #4) have served to keep Stanfordat the leading edge in forging university–industryliaisons. Primarily these relations have been chan-nelled through Yamaha. Importantly, the rela-tionship has recently evolved to a joint licens-ing arrangement that weds the basic research ofStanford with Yamaha’s applications. For the re-lationship to work, though, a tacit arrangement isnot sufficient, the partnership must be codified.

Vignette III: Dave Smith and the perilsof not patenting

Dave Smith is the inventor of the Prophet 5synthesizer, the world’s first mass market pro-grammable instrument. He is also recognized forthe leadership role he played in launching MIDI(Musical Instrument Digital Interface), the proto-col that would enable the everyday connectionof computers and musical instruments (Chadabe

1997). Smith entered the world of electronic mu-sic armed with a degree from UC Berkeley incomputer engineering and work experience in Sil-icon Valley’s aerospace and computer industries.Smith further had a background as an amateurbass player in San Francisco’s music scene. Ofinterest to this paper is that in the patentingarena, Smith does not register. It is this differ-ence in the way that he approached the problemof codification, especially as it concerns the trans-fer of knowledge to Japanese firms that I seek tohighlight.

In 1977, as President of Sequential Circuits,Dave Smith licensed a polyphonic keyboardpatent from a firm based near Santa Cruz,CA called E-Mu. Since his fledging firm hadconsumed most of its financial resources in get-ting the Prophet-5 keyboard to market an agree-ment was made to postpone royalty payments.By 1979 the success of Prophet-5, the world’sfirst instrument to incorporate a microprocessor,had vaulted Sequential to the head of the in-dustry and the royalty checks which Sequentialcould now afford to write to E-Mu representeda large transfer of funds. This cash flow in turnhelped finance the R&D efforts of E-Mu. However,in May 1980, the staff at E-Mu received a letterfrom Smith informing them that he had decidedto stop paying royalties on the Prophet-5 design.A lawsuit between the two ensued. According toE-Mu’s founder Dave Rossum, ‘E-Mu went fromhaving a positive cash flow to having a fairlysubstantial negative cash-flow to pay the lawyersto get Sequential to pay us the money that we



thought they owed us’ (quoted in Vail 1993, 199).The two parties eventually settled out of court inan amicable fashion.

Over the course of the Sequential–E-Mu rela-tionship, the latter’s patent worked in a numberof ways. At the outset, the patent sanctioned thetransfer of codified knowledge in a manner thatwas only partially consistent with the conventionsof licensing. Such a soft agreement between thetwo firms was likely the outcome of a shared en-trepreneurial culture: one that was perhaps evenembedded in a larger set of loosely defined re-gional codes of practice that characterized theformative days of Silicon Valley (Saxenian 1994).A little later, after the patent’s worth had provenitself, Sequential, for a time, recognized this valueby conceding royalties to E-Mu. The terminationof royalty payments is more difficult to inter-pret except to say that Sequential viewed theworth of the E-Mu patent in limited terms. Fi-nally, the lawsuit and resolution prove the im-mense power of this inscription to discipline thebehaviour of the disputing parties. Nonetheless,this experience shaped Smith’s views on the valueof patents thereafter.

We never patented anything at Sequential. It waspartially the cost, partially laziness and partiallymy own philosophy which was that in most casesif you are trying to patent something then it isonly as good as your ability to defend it . . . . If youhave a lot of patents then you are pretty well cov-ered and nobody is going to come after you (DaveSmith, personal communication, 11 October 2002,conversation).

In contrast, with no patents to bolster their po-sition, Sequential Circuits, had little to fall backupon when sales slumped in the mid-1980s andthe firm became a takeover target for Japanesecompetitors. These firms were already familiarwith Smith, since he was the lone American key-board manufacturer with the prescience to col-laborate on the non-proprietary MIDI protocol.Though he had discussed the idea of initiating acommon serial interface with several U.S. firms,the detour proved too daunting.

No one in the States seemed to be interested . . . andwe lost interest trying to round everyone up, sowe worked with the Japanese companies (quotedin Chadabe 1997, 195).

In 1988, Sequential was bought by Yamaha,closed 6 months later and reopened by Tokyo-based Korg, who that same year had become51 percent owned by Yamaha. Without a patentportfolio, what role did other forms of industrialwriting play in the transfer of knowledge fromSequential to Yamaha-Korg?

One of the things I did when I [first] came overto [Yamaha] was I brought a two-page list of prod-ucts that could be built—brief descriptions, a para-graph on each one. Some of them were follow-upsof products that we had already done and oth-ers were new ideas. I was describing one of ourideas for a new product and so I was drawing somestuff on the board . . . Basically, what I drew on their[Yamaha’s] chalkboard was the original concept, de-scribing this stuff, saying how it’s going to work,it’s going to have these things, etc. And the netresult of the meeting was their engineers saying,‘Well we don’t think that’s going to work’. And Isaid, ‘Well, sure it’s going to work’. . . We ended updoing the product, not for Yamaha, but for Korg(Dave Smith, personal communication, 11 October2002, conversation).

Yamaha’s engineers had not been workingthrough the same sort of ideas as Smith and thuscould not initially appreciate his inscriptions. Butonce presented with this recipe, they were in aposition to act. Interestingly, their first act ap-pears to be one of re-inscription. A few yearslater, it was to Smith’s surprise that the followingepisode transpired.

I was looking through a trade magazine—the AESJournal always prints a summary of patents in theback, a paragraph on each one—and I was brows-ing through it just to see what was going on. Iwas reading one from Yamaha, because, of course,they always have a bunch of them in there. AndI thought, ‘wait a minute’, that sounds familiar.So I went back and re-read it and it was basi-cally a patent of what we were doing for thatsynthesizer and it was based on those drawingsthat I did when I was describing it to them (DaveSmith, personal communication, 11 October 2002,conversation).

That these ideas first echoed to Smith while hebrowsed the AES Journal also points out thatinscriptions, even scribbles, once translated intopatent form, perform in a variety of contexts.



Table 5A selection of U.S. engineers employed by Japanese EMI manufacturers

Name Employment prior to Japan Employment for Japanese firms

Ralph Deutscha North American Rockwell 1965–1973 Contract invention and consulting for Yamaha 1973–1975,Kawai 1975–1986

John Chowningb Stanford University Department of Music 1966–present Licensed FM synthesis patent to Yamaha, consulting 1975Dave Smithc Founder and President of Sequential Circuits 1974–1988 Headed Yamaha DSD 1988–1989, Korg R&D US 1989–1994John Bowend Moog Music 1973–1976, Sequential Circuits 1982–1987 Yamaha DSD 1988-89, Korg R&D US 1989–1998Chris Meyerd Sequential Circuits 1984–1987, Digidesign 1988–1989,

Marion Systems 1990Roland R&D 1990–1997 (tasks included analysis of

competitors intellectual property)Roger Linne Founder and President of Linn Electronics 1978–1986 Contract invention for Akai 1987–1994Tom Oberheimf Founder and President Oberheim Electronics 1970–1985,

Marion Systems 1987–presentContract invention for Roland 1988–1990

Alberto Kniepkampg Head of R&D for Lowery Organs 1965–1969, Norlin1969–1978

Patent and product design consulting for Roland1989–present

Herb Deutschh Hofstra University and Moog Music 1964–1977, Norlin Contract consulting for RolandCarlo Lucarellig Technical manager Farfisa (Itl), Founder president of SIEL (Itl.) President of Roland Europe 1988Dennis Houlihang Lowery Organs, author of ‘Owner’s Manual’, VP Marketing President of Roland USAnthony Billiasd Professional Musician, Music Retail Shop Floor Demonstrator Product Specialist at Korg, Director of Marketing for

Technology Products YamahaJohn Lemkuhld Music retail Synthesizer voicing Korg 1988Nick Howesd Degree in Astrophysics, Professional Musician, Music Retail

(UK)Yamaha R&D (UK) 1992–

Mark Moffati Studio Engineer EMI (Aus) 1972–1980, Record Producer,Nashville 1982

Contract consulting for Roland 1978–

SOURCES:a Personal communication, 12 October 2002, conversation.b Johnstone 1999.c Personal communication, 11 October 2002, conversation.d Sonik Matter website (http://sonikmatter.com), accessed 01 April 2004.e Roger Linn Design website (http://www.rlinndesign.com), accessed 01 April 2004.f Sound on Sound website (http://www.soundonsound.com/sos/1994 articles/mar94/tomoberheim.html?session=95455209cfc1c8b0c61ed0dd6b190e0a), accessed 01 April 2004.g Kakehashi 2002, 149–151.h Herb Deutsch website (http://www.hofstra.edu/academics/hclas/music/music deutsch.cfm), accessed 01 April 2004.i Roland User’s Group on-line newsletter (www.rolandsus.com/community/rug/Fall 01/currents.asp), accessed 02 December 2002.

In the case of the AES Journal they commu-nicated to a particular community of practiceboth what Yamaha was doing and, in theory atleast, what Yamaha could proscribe others fromdoing.

Smith’s story should be further highlighted be-cause it sets the mould for a subsequent pat-tern of U.S. engineers who became vectors ofknowledge transfer to Japan, once their firmsfailed (Table 5). Several of Smith’s team at Se-quential Circuits found employment with variousJapanese EMI manufacturers. Chris Meyer, for ex-ample, turned his tacit knowledge to IP analy-sis for Roland. The eponymous firms of RogerLinn and Tom Oberheim failed, as Sequentialdid. Thereafter, Linn did contract work for Akai,

while Oberheim turned his talents to problemsolving for Roland. Alberto Kniepkamp, onceLowery’s head of engineering who subsequentlyworked for Norlin before it too was dissolved,gained employment with Roland as, among otherthings, a patent consultant. Denis Houlihan, au-thor of Lowery Organ’s ‘Owner’s Manual’, trans-lated his knowledge and, importantly, technicalwriting skills to Roland. Indeed, Japanese EMImakers have selectively hired specialists from allfacets of the production chain, from R&D to retailin their efforts to absorb knowledge from theirformer American competitors. In several of thesecases, individuals were hired so that their tacitknowledge could be applied in some capacity tothe analysis and codification of texts.



Codification in Practice and Theory

A comparison of the Deutsch, Chowning andSmith career trajectories is instructive for severalreasons. To summarize the vignettes, Deutschtravelled to Japan to work with Yamaha; Chown-ing patented at Stanford but his landmark FMsynthesis patent was licensed to Yamaha; andSmith gave tacit knowledge to Yamaha who laterpatented his idea. In their own way, each of thesebiographies illustrates that knowledge formationwas the result of a cycling process between tacitand codified domains. The case studies affirmthat tacit knowledge matters, but, not surpris-ingly in highly idiosyncratic ways. For example,rational calculation did not precipitate Chown-ing’s ‘Eureka!’ moment, his ‘ear discovery’ of FMsynthesis. As Johnstone (1999, 216) infers, it wasa discovery that ‘an engineer would have beenunlikely to make’. Nevertheless, Chowning’s mu-sical training, his later interest in learning thepracticalities of computer science and his connec-tion to Bell Labs contributed to his stock of tacitknowledge. Yet what ultimately put this knowl-edge in play was the moment of externalizationwhen Chowning and Stanford fixed this knowl-edge in a proprietary manner by patenting andlicensing it to Yamaha.

In addition, a combination of institutionalcontexts and individual practices shaped theapproaches to and outcomes of knowledge cod-ification. For Ralph Deutsch it was prudent topatent every single technology he developed sinceit was this practice that would safeguard hisinterests as he moved among several corpora-tions. It was his way of making sure that heand his inscriptions became indispensable withinan evolving network of (often competing) inter-ests. Crucially, the act of codification made hiscontributions translatable across space, but itwas still necessary for him to travel to Japanto demonstrate to Yamaha (and subsequentlyKawai) the practical application of his inventions.In the case of FM synthesis, Chowning’s lonepatent cemented a relationship between Stanfordand Yamaha, the legacy of which is CCRMA,an institution that has fostered the developmentof a number of fundamental technologies thathave directly benefited both the university andYamaha. Finally, for Smith, the entrepreneur, theinscription of knowledge in patent form was a

practice more suitable for those with the re-sources and energy to defend that knowledge.Smith’s approach to codified knowledge, whetherin respect to E-Mu’s patent, or to the designs hejotted down on Yamaha’s blackboard, appears farmore relaxed and perhaps naive than Deutsch’s.This might be a personal or generational matter;it also might indicate a difference in the indus-trial cultures of the Los Angeles aerospace indus-try versus that of the milieu that blossomed atthe intersection of the Bay Area music scene andthe Silicon Valley community.

The way that these individuals related toYamaha is additionally instructive in what it saysabout the geographies of embedding, disembed-ding and reembedding of knowledge from onesetting to another. At the outset, these engineersacted as ‘boundary crossers’ who cultivated abody of knowledge positioned at the cusp be-tween two communities of practice: electronicsand music. In the case of Deutsch and Smith thissetting was the high-tech sector in California ata time when military technologies were being ap-plied to meet consumer ends. For Chowning, thenetwork he forged with Bell Labs and the train-ing he undertook in Stanford’s computer sciencedepartment would appear to have provided a suf-ficient basis for further translations. However, inattempting to translate these quite radical sets ofknowledge to the American EMI, each of these in-dividuals faced resistance from prevailing habitsof thought held by their geographically proxi-mate peers: the digital organ idea was rejectedby most U.S. firms; Chowning’s FM idea was notpart of Hammond’s ‘world’. Even Smith, who atthe time was the bright young star of the U.S.synthesizer industry, could not persuade his U.S.colleagues of the importance of MIDI. It is alsosurprising, at least in the Chowning and Smithcases, that Silicon Valley, a region that standsas the bench mark for industrial learning, failedto adopt these radical ideas. Smith, for instance,was unable to secure venture capital from outsidesources to finance Sequential. Indeed, the utterlack of local linkages stands out and calls intoquestion Markowitz’s claim that Deutsch ‘soldout’. In fact, like Smith and Chowning, the onlydetour available to Deutsch entailed working withthe Japanese.

In each case, Yamaha was prescient enough tooffer these individuals the best deal to facilitate



the translation of their interests. Even at thisstage, the transfer of knowledge was far fromstraightforward and translations had to be per-formed in both the literal and figurative sense.Deutsch took it upon himself to learn Japanese.The only way for Smith to communicate was via ablackboard. In their relations with Yamaha, bothDeutsch and Smith found that their ideas werenot automatically celebrated or their roles privi-leged. Deutsch’s NES lost out to the PAS in theYamaha trials of 1974, while Smith left his meet-ing at Yamaha with the belief that they were notgoing to use his ideas—they were just scribbles,after all. In the final analysis, each of these con-tributions played a role in transferring knowl-edge to Yamaha. Deutsch acted as a pioneer inthe implementation of digital sound, Chowningtransferred the method by which digital synthe-sis could be realized, while Smith’s collaborationon the MIDI protocol enabled the entire horizon-tal integration of the industry that swung in theJapanese favour in the latter half of the 1980s.

Conclusion

This case study sheds light on how the prac-tice of industrial writing shapes industrial ge-ographies. By emphasizing codification it lendsbalance to a dialogue that has perhaps over-stated the importance of tacit knowledge. Thehistory of industrialization is one of geograph-ical concentration and dispersal (Hayter 1997).While it is too much of a simplification to saythat the accumulation of tacit knowledge explainsthe concentrations and codification the disper-sals, the ability to translate ideas to new contextsis greatly facilitated once knowledge has beeninscribed. The problem, however, is that those in-scriptions rarely operate independent of the net-work of inventors, firms and, importantly, otherinscription in which they are embedded. This iswhy the practice of asserting novelty throughwriting and enacting property claims based onthese writings, in short patenting, is so fraughtwith potential hazards. The problem is especiallyacute when new, radical knowledge is produced.In these cases the scripts that might lend con-text to these inventive leaps are absent and theability to enroll other actors into the network

hinges on whether they possess a common frameof tacit knowledge. This study demonstrated thatproximity is not necessarily a good predictor ofwhether or not these connections between ac-tors will be forged. When a connection is madeand novelty recognized, the new linkage mustbe externalized. Contracts must be signed, keypatents licensed and new derivative patents cod-ified to anchor the translation. The motives andactions of individuals and firms may shift andstrain against these anchors, with the result thatwhole projects may come unmoored and dis-placed. Through a juxtaposition of the roles of in-ventors and their writings in the spatial transferof knowledge, I have sought to highlight the com-plexities that arise when Nonaka and Takeuchi’s(1995) knowledge cycle is spatialized.

The paper makes the further methodologicalargument that economic geography has madeeffective, but ultimately partial use of patentstatistics. It critically reflects on the knowledgespillovers school and its method of countingpatent citations. Indeed, there appear to be sev-eral logics of citing ‘prior art’ not all of whichare straightforward indicators of the significanceof a given invention. Rather, particular inscrip-tive practices appear to wed important inven-tions with insignificant derivative patents, thelatter serving as noise, to complicate the land-scape for imitators. Patents are not merely anoutcome of research, a proxy metric for innova-tion. These records are the tip of the iceberg—meaningless when divorced from the mass ofindividual tactics, corporate strategies and in-stitutional settings that characterize the autho-rization of novelty. When closely examined, theypresent just one facet of the lives of inventionsor the inventors who create them. In highlightingthis juxtaposition, the study further proposes abiographically informed actor–network methodol-ogy that tracks the connections forged betweenpeople, inscriptions and artifacts.

This perspective additionally enabled empiricalinsight into the debate in geography over whether‘communities of practice’ can be made to op-erate at a distance (Amin and Cohendet 1999;Gertler 2001; Coe and Bunnel 2003). In particular,this case study suggests that it is in the absenceof community—inventors who could not conformto the practices of corporations and inter-firm



alliances, engineers with genuinely novel ideasthat could not win the support of their (local)peers, etc.—when actors are forced to mobilizetheir knowledge to new contexts. Moreover, byfocusing attention on the practice of patenting,this paper provided illustration of how inscrip-tions work to include or exclude parties from acommunity. With this evidence in mind, the socialarchitecture of communities of practice would ap-pear more suited to function both locally and ata distance only when the problem to be solved iswell defined. In contrast, during episodes whennew technologies are in their infancy and thepath forward is uncertain, the process of knowl-edge transfer across boundaries is complex andcannot easily be determined in advance.

Further research is required to understand thecomparative institutional contexts that embedpractices such as patenting; for instance, by ex-amining the work of ancillary actors that in-clude patent attorneys, examiners and contractreverse engineers. Historical analyses inform usof discreet national examples (Kumagai 1999;Lamoreaux and Sokoloff 2002), yet there havebeen few systematic studies that update the pi-oneering work of Penrose (1951) or, more criti-cally, address the manner in which mobilized IPbring these systems into collision. Especially inan era of purported harmonization between theconventions of European, American and Japanesepatent regimes, it is imperative to assess the de-gree to which each of these regional contextsserves to modify the traffic of intellectual prop-erties that move across their frontiers. For ex-ample, Japanese patent law allows the submis-sion of applications in English, under the provi-sion that they be translated into Japanese priorto examination. What sort of technical or lin-guistic detours does such a process entail? Howdoes geography matter to these processes ofcodification?

Acknowledgements

This research was supported by a SSHRC Doctoral Fellowshipand Monbukagakusho Research Scholarship. The author wouldlike to thank Roger Hayter for his constructive input through-out this research project, Professors Ishikawa Yoshitaka andNishihara Jun for their advice and assistance in Japan, thethree anonymous reviewers and finally the participants fortheir time and insight. Any remaining errors and omissionsare mine.

References

AMIN, A., and COHENDET, P. 1999 ‘Learning and adaptation in decen-tralized business networks’ Environment and Planning D:Society and Space 17, 87–104

ASHEIM, B. 1999 ‘Interactive learning and localized knowledge inglobalizing learning economies’ Geojournal 49, 345–352

AUDRETSCH, D., and FELDMAN, M. 1996 ‘R&D spillovers and the ge-ography of innovation and production’ American EconomicReview 86, 630–640

BARNES, T. 2001 ‘Lives lived and lives told: biographies of geog-raphy’s quantitative revolution’ Environment and PlanningD: Society and Space 19(4), 409–429

BATHELT, H., MALMBERG, A., and MASKELL, P. 2004 ‘Clusters and knowl-edge: local buzz, global pipelines and the process of knowl-edge creation’ Progress in Human Geography 28, 31–56

BRACZYK, H., COOKE, P., and HEIDENREICH, M. ed 1998 Regional Innova-tion Systems: The Role of Governances in a Globalized World(London: UCL Press)

BRESCHI, S., and LISSONI, F. 2001 ‘Localized knowledge spillovers vs.innovative milieux: knowledge “tacitness” reconsidered’ Pa-pers in Regional Science 80(3), 255–274

BUNNEL, T., and COE, N. 2001 ‘Spaces and scales of innovation’Progress in Human Geography 25(4), 569–589

CEH, B. 2001 ‘Regional innovation in the United States: Evidenceof spatial transformation’ Papers in Regional Science 80(3),297–317

CHADABE, J. 1997 Electric Sound: The Past and Promise of Elec-tronic Music (Upper Saddle River, NJ: Prentice Hall)

COE, N., and BUNNELL, T. 2003 ‘“Spatializing” knowledge communi-ties: towards a conceptualization of transnational innova-tion networks’ Global Networks 3(4), 437–456

COLBECK, J. 1996 Keyfax Omnibus (Emeryville, CA: Mix Books)COWAN, R., DAVID, P., and FORAY, D. 2000 ‘The explicit economies of

knowledge codification and tacitness’ Industrial and Corpo-rate Change 9, 211–253

DE LAET, M. 2000 ‘Patents, travel, space: Ethnographic encounterswith objects in transit’ Environment and Planning C: Societyand Space 18, 149–168

EVENDEN, M. 2004 ‘Locating science, locating salmon: institutions,linkages and spatial practices in early British Columbiafisheries science’ Environment and Planning D: Society andSpace 22, 355–372

FREEMAN, C. 1987 Technology Policy and Economic Performance:Lessons from Japan (London: Pinter Publishers)

FREEMAN, C., CLARK, J., and SOETE, L. 1982 Unemployment and Techni-cal Innovation: A Study of Long Waves and Economic Devel-opment (London: Pinter Publishers)

GERTLER, M. 1995 ‘“Being there”: proximity, organization and cul-ture in the development and adoption of advanced manu-facturing technologies’ Economic Geography 71, 1–26

—. 2001 ‘Best practice? Geography, learning and the institu-tional limits to strong convergence’ Journal of Economic Ge-ography 1, 5–26

—. 2004 Manufacturing Culture: The Institutional Geographyof Industrial Practice (New York: Oxford University Press)

GRANSTRAND, O. 1999 The Economics and Management of Intellec-tual Property (Cheltenham: Edward Elgar)

HAYTER, R. 1997 The Dynamics of Industrialization: The Factory,the Firm and the Production System (Chichester: Wiley)

HENRY, N., and PINCH, S. 2000. ‘The industrial agglomeration of mo-tor sport valley: a knowledge, space, economy approach’ in



Knowledge, Space, Economy, eds J. Bryson, P. Daniels, N.Henry, and J. Pollard (London: Routledge) 120–141

HOWELLS, J. 2002 ‘Tacit knowledge, innovation and economic ge-ography’ Urban Studies 39(5-6), 871–884

JAFFE, A., TRAJTENBERG, M., and HENDERSON, R. 1993 ‘Geographical local-ization of knowledge spillovers as evidenced by patent ci-tations’ The Quarterly Journal of Economics 108, 577–586

JOHNSON, B., LORENZ, E., and LUNDVALL, B.-A. 2002 ‘Why all the fussabout codified and tacit knowledge?’ Industrial and Corpo-rate Change 11, 245–262

JOHNSTONE, B. 1994 ‘Wave of the Future’ Wired 2(3), 1–7 (http://www.wired.com/wired/archive/2.03/waveguides.html?pg=2&topic; last accessed 21 April 2006)

—. 1999 We Were Burning: Japanese Entrepreneurs and theForging of the Electronic Age (New York: Basic Books)

KAKEHASHI, I. 2002 I Believe in Music: Life Experiences andThoughts on the Future of Electronic Music by the Founderof the Roland Corporation (Milwaukee: Hal Leonard)

KAWAI. 1997 Sekai Ichi no Piano Tsukuri Mezashite [Aimingto Make the World’s Best Piano] (Hamamatsu: The KawaiCorporation)

KOTLER, M., and HAMILTON, G. 1995 A Guide to Japan’s Patent System(Washington: Japan Information Access Project)

KUMAGAI, K. 1999 History of Japanese Industrial Property Sys-tem (Tokyo: Japanese Patent Office – Asia Pacific IndustrialProperty Centre JIII)

LAMOREAUX, N., and SOKOLOFF, K. 2002 Intermediaries in the U.S. Mar-ket for Technology 1870–1920 (Cambridge, MA: NBER) Na-tional Bureau of Economic Research—Working Paper Series

LATOUR, B. 1987 Science in Action: How to Follow Scientistsand Engineers through Society (Cambridge, MA: HarvardUniversity Press)

—. 1999 Pandora’s Hope: Essays on the Reality of Science Stud-ies (Cambridge, MA: Harvard University Press)

LIVINGSTON, D. 1999 ‘No more disembodied minds’ Books and Cul-ture 5(3), 21–23

MACDONALD, S. 2004 ‘When means become ends: considering theimpact of patent strategy on innovation’ Information Eco-nomics and Policy 16, 135–158

MALMBERG, A., and MASKELL, P. 2002 ‘The elusive concept of local-ization economies: Towards a knowledge-based theory ofspatial clustering’ Environment and Planning A 34, 429–449

MARKOWITZ, J. 1989 Triumphs & Trials of an Organ Builder(Macungie, PA: Allen Organ Company)

MARKUSEN, A. 1996 ‘Sticky places in slippery space: a typology ofindustrial districts’ Economic Geography 72, 293–313

MASKELL, P., and MALMBERG, A. 1999 ‘The competitiveness of firmsand regions. ‘Ubiquification’ and the importance of local-ized learning’ European Urban and Regional Studies 6(1),9–25

MATHEWS, M. 1963 ‘The digital computer as a musical instrument’Science 142, 553–557

MCKENNA, B. 2006 ‘An idea worth patenting: dedicated patentcourts’ The Globe and Mail January 31: B8

MENSCH, G. 1979 Stalemate in Technology: Innovation Overcamethe Depression (Cambridge, MA: Ballinger)

MILANO, D. 1993 ‘American synthesizer builders: triumph andcrises for an industry in transition’ in Keyboard PresentsVintage Synthesizers: Groundbreaking Intsruments and Pi-oneering Designers of Electronic Music Synthesizers, ed M.Vail (San Francisco: Miller Freeman Books) 17–28

MOCHIDA, Y., and AOKI, G. 1994 Gakki to Konpyu-ta [Musical Instru-ments and Computers] (Tokyo: Kyoritsu Shuppan)

MURDOCH, J. 1994 ‘Sustainable knowledge’ Geoforum 25(2), 115–132

MYERS, G. 1995 ‘From discovery to invention: the writing and re-writing of two patents’ Social Studies of Science 25, 57–105

THE MUSIC TRADES. 2000 ‘The Global 225: The World’s Largest Mu-sic Products Suppliers’ The Music Trades December: 80–84

NAKAGAWA, Y. 1984 Nippon Gakki no LSI Kaihatsu Senryaku[Yamaha’s LSI Development Strategy] (Tokyo: Diamond)

NONAKA, I., and TAKEUCHI, H. 1995 The Knowledge Creating Com-pany: How Japanese Companies Create the Dynamics of In-novation (New York: Oxford University Press)

NONAKA, I., TOYAMA, R., and NAGATA, A. 2000 ‘A firm as a knowledge-creating entity: a new perspective on the theory of the firm’Industrial and Corporate Change 9, 1–20

ODAGIRI, A., and GOTO, H. 1993 Innovation in Japan (Oxford: Claren-don Press)

OHTSUKA, M. 1980 ‘Hamamatsu Chiiki ni Okeru Gakki Kougyouno Shuuseki’ [Musical Instrument Industry in Hamamatsu]Chirigaku Hyouron 53(3), 157–170

OINAS, P., and MALECKI, E. 2002 ‘The evolution of technologies intime and space: from national and regional to spatial inno-vation systems’ International Regional Science Review 25(1),102–131

PARTNER, S. 1999 Assembled in Japan: Electrical Goods and theMaking of the Japanese Consumer (Berkeley: University ofCalifornia Press)

PENROSE, E. 1951 The Economics of the International Patent Sys-tem (Baltimore: Johns Hopkins University Press)

PINCH, S., and HENRY, N. 1999 ‘Discursive aspects of technologicalinnovation: the case of the British motor-sport industry’ En-vironment and Planning A 31(4), 665–682

PINCH, T., and TROCCO, F. 2002 Analog Days: The Invention andImpact of the Moog Synthesizer (Cambridge, MA: HarvardUniversity Press)

POLANYI, M. 1967 The Tacit Dimension (New York: Anchor Books)PRED, A. 1966 The Spatial Implications of US Industrial Growth

(Cambridge, MA: MIT Press)PRESTOWITZ, C. 1989 Trading Places: How We Are Giving Our Fu-

ture to Japan and How to Reclaim It (New York: BasicBooks)

REIFFENSTEIN, T. 2004 Transectorial Innovation, Location Dynamicsand Knowledge Formation in the Japanese Electronic Musi-cal Instrument Industry (Unpublished PhD Dissertation, De-partment of Geography, Simon Fraser University, Burnaby,BC)

—. 2005 ‘Radical innovation and the emergence of new eco-nomic spaces: the formation of knowledge in the Ameri-can and Japanese electronic musical instrument industry’in New Economic Spaces, New Economic Geographies, edsR. LeHeron and J. W. Harrington (Aldershot: Ashgate) 29–43

RIVETTE, K., and KLINE, D. 2000 ‘Discovering new value in intellectualproperty’ Harvard Business Review 78(1), 54–66

ROLAND CORPORATION. 1999 Roland Corporate Guide (Hamamatsu:Roland Corporation)

ROSEN, D., and USUI, C. 1994 ‘The social structure of Japanese intel-lectual property law’ UCLA Pacific Basin Law Journal 13(1),32–69



SAXENIAN, A-L. 1994 Regional Advantage: Culture and Competitionin Silicon Valley and Route 128 (Cambridge, MA: HarvardUniversity Press)

SCHMOOKLER, J. 1966 Invention and Economic Growth (Cambridge,MA: Harvard University Press)

SCHOENBERGER, E. 2001 ‘Corporate autobiographies: the narrativestrategies of corporate strategists’ Journal of Economic Ge-ography 1(3), 277–298

SHIVA, V. 2000 ‘North-south conflicts in intellectual propertytights’ Peace Review 12, 501–508

TAKAHASHI, Y. 2000 ‘A network of tinkerers: the advent of theradio and television receiver industry in Japan’ Technologyand Culture 41, 460–484

TAKENAKA, T. 1994 ‘The role of the Japanese patent system inJapanese industry’ UCLA Pacific Basin Law Journal 13(1),25–31

TAKEO, H. 1988 ‘Hamamatsu ni Okeru Gakki Kougyou no Henyou’[Metamorphosis of Musical Instrument Industry in Hama-matsu] Chirigaku Houkoku 67, 55–66

TAPLIN, R. 2004 Valuing Intellectual Property in Japan, Britain andthe United States (Oxford: RoutledgeCurzon)

THEBERGE, P. 1997 Any Sound You Can Imagine: Making Music/Consuming Technology (Hanover, NH: Wesleyan UniversityPress)

VAIL, M. 1993 Vintage Synthesizers: Groundbreaking Instrumentsand Pioneering Designers of Electronic Music synthesizers(San Francisco: GPI Books)

VON KROGH, G., ICHIJO, K., and NONAKA, I. 2000 Enabling Knowledge Cre-ation: How to Unlock the Mystery of Tacit Knowledge andRelease the Power of Innovation (Oxford: Oxford UniversityPress)

WALL STREET JOURNAL. 2006 ‘Patently Absurd’ March 11, A14WENGER, E. 1998 Communities of Practice: Learning, Meaning and

Identity (Cambridge: Cambridge University Press)WINDER, G. 2001 ‘Building trust and managing business over a

distance: a geography of reaper manufacturer D.S. Morgan’scorrespondence, 1867’ Economic Geography 77(2), 95–121

YAMAHA. 1987 Yamaha 100 Nenshi (Hamamatsu: Yamaha)


Documents

Codification, patents and the geography of knowledge transfer in …aj/archives/articles/Reiffen... · 2015-02-06 · Codification, patents and the geography of knowledge transfer