The Body and the Archive Author(S): Allan Sekula Reviewed Work(S): Source: October, Vol

The Body and the Archive Author(s): Allan Sekula Reviewed work(s): Source: October, Vol. 39 (Winter, 1986), pp. 3-64 Published by: The MIT Press Stable URL: http://www.jstor.org/stable/778312 . Accessed: 27/08/2012 13:20

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

The MIT Press is collaborating with JSTOR to digitize, preserve and extend access to October.

http://www.jstor.org The Body and the Archive*

ALLAN SEKULA

. theremust be arrangeda comprehensive systemof exchanges, so thatthere might grow up somethinglike a universalcurrency ofthese banknotes,or promisesto pay in solid sub- stance,which the sun has engravedfor the greatBank ofNature. - Oliver Wendell Holmes, 1859

On theone side we approachmore closely to whatis goodand beautiful;on theother, vice and sufferingare shut up withinnarrower limits;and we have to dreadless themon- strosities,physical and moral,which have the power to throwperturbation into the social framework. --Adolphe Quetelet, 1842

The sheer range and volume of photographic practice offersample evidence of the paradoxical status of photographywithin bourgeois culture. The simultaneous threatand promise of the new medium was recognized at a very early date, even beforethe daguerreotypeprocess had proliferated.For exam-

* Earlier versions of this essay were presented at the National Gallery of Canada, Ottawa, October 2, 1982, and at the College Art Association Annual Meeting, New York, February 13, 1986. This version was completed with the assistance of a Visiting Senior Fellowship at the Center forAdvanced Studies in the Visual Arts,The National Gallery of Art, Washington, D.C., summer 1986. 4 OCTOBER

ple, followingthe French governmentannouncement of the daguerreotypein August 1839, a song circulated in London which began with the following verse: O Mister Daguerre! Sure you're not aware Of half the impressionsyou're making, By the sun's potent rays you'll set Thames in a blaze, While the National Gallery's breaking. Initially, photography threatens to overwhelm the citadels of high culture. The somewhat mockinghumor of thisverse is more pronounced ifwe consider that the National Gallery had only moved to its new, classical building on Trafalgar Square in 1838, the collection having grown rapidly since the gallery's founding in 1824. I stress this point because this song does not pit photographyagainst a static traditionalculture, but ratherplays on the possibilityof a technologicaloutpacing of alreadyexpanding cultural institutions.In this context, photographyis not the harbinger of modernity,for the world is already modernizing. Rather, photographyis modernityrun riot. But danger resides not only in the numerical proliferationof images. This is also a prema- ture fantasyof the triumphof a massculture, a fantasywhich reverberateswith political foreboding. Photography promises an enhanced mastery of nature, but photographyalso threatensconflagration and anarchy, an incendiarylevel- ing of the existingcultural order. By the thirdverse of this song, however, a new social order is predicted: The new Police Act will takedown each fact That occurs in its wide jurisdiction And each beggar and thiefin the boldest relief Will be givinga colorto fiction.' Again, the last line of the verse yields a surplus wit, playing on the figurative ambiguityof "givinga color," which could suggestboth the elaboration and un- masking of an untruth,playing furtheron the obvious monochromaticlimita- tionsof the new medium, and on the approximatehomophony of colorand collar. But thisvelvet wit plays about an iron cage which was then in the process ofbe- ing constructed.Although no "Police Act" had yet embraced photography,the 1820s and '30s had engendered a spate of governmentalinquiries and legisla- tion designed to professionalizeand standardize police and penal procedures in Britain, the most importantof which were the Gaols Act of 1823 and the Met- ropolitan Police Acts of 1829 and 1839. (The prime instigatorof these modernization efforts,Sir Robert Peel, happened to be a major collectorof seven-

1. Quoted in Helmut and Alison Gernshiem, L. J M. Daguerre,New York, Dover, 1968, p. 105 (italics in original). .?,Al:'pe.?...... 4!x!

nggrIm ...... Rol

Vt ...... mm...... E. N."",

1.2 JILPR;-K W. pijq4fI*t:. H. k0 ...... WilliamHenry Fox Talbot.Articles of China, plate III fromThe Pencilof Nature,1844. teenth-centuryDutch paintings,and a trusteeof the National Gallery.) Directly to the point of the song, however,was a provisionin the 1839 act fortaking into custodyvagrants, the homeless, and otheroffenders "whose name and residence [could] not be ascertained."2 Although photographicdocumentation of prisoners was not at all common until the 1860s, the potentialfor a new juridical photographicrealism was widelyrecognized in the 1840s, in the general contextof these systematicefforts to regulate the growingurban presence of the "dangerous classes," of a chroni- cally unemployed sub-proletariat.The anonymous lyricistvoiced sentiments thatwere also heard in the higherchambers of the new cultureof photography. Consider that incunabulum in the historyof photography, Henry Fox Talbot's The Pencilof Nature. Talbot, the English gentleman-amateurscientist who paralleled Daguerre's metallic inventionwith his own paper process, produced a lavish book that was not only the first to be illustrated with photographic prints, but also a compendium of wide-ranging and prescient meditationson the promise of photography.These meditationstook the formof briefcommentaries on each of the book's calotypeprints. Talbot's aestheticambition was clear: forone austere image of a broom leaning beside an (allegori-

2. The MetropolitanPolice Act, 1839, in Halsbuiy'sStatutes of England,vol. 25, London, Butterworth,1970, p. 250. For a usefulsummary of parliamentary debates on crimeand punish- mentin thenineteenth century, see CatalogueofBritish Parliamentary Papers, Dublin, Irish University Press,1977, pp. 58-73. On thehistory of the National Gallery, see MichaelWilson, The National Gallery:London, London, Philip Wilson Publishers. 6 OCTOBER

cally) open door, he claimed the "authorityof the Dutch school of art, fortaking as subjects of representationscenes of daily and familiaroccurrence."3 But an entirelydifferent order of naturalism emerges in his notes on another quite beautiful calotype depicting several shelves bearing "articles of china." Here Talbot speculates that "should a thiefafterwards purloin the treasures- if the mute testimonyof the picture were to be produced against him in court- it would certainly be evidence of a novel kind."4 Talbot lays claim to a new legalistictruth, the truthof an indexical ratherthan textualinventory. Although this frontalarrangement of objects had its precedentsin scientificand technical illustration,a claim is being made here that would not have been made fora drawing or a descriptivelist. Only the photograph could begin to claim the legal status of a visualdocument of ownership. Although the calotype was too insensitiveto light to record any but the most willing and patient sitters,its evidentiarypromise could be explored in this-property-consciousvariant of the still life. Both Talbot and the author of the comic homage to Daguerre recognized a new instrumentalpotential in photography:a silence that silences. The protean oral "texts"of the criminal and pauper yield to a "mute testimony"that "takes down" (that diminishes in credibility,that transcribes) and unmasks the disguises, the alibis, the excuses and multiple biographies of those who find or place themselves on the wrong side of the law. This battle between the presumed denotative univocality of the legal image and the multiplicityand presumed duplicityof the criminal voice is played out during the remainderof the nineteenthcentury. In the course of thisbattle a new object is defined- the criminal body- and, as a result, a more extensive "social body" is invented. We are confronting,then, a double system: a system of representation capable of functioningboth honorificallyand repressively.This double operation is most evidentin the workingsof photographic portraiture. On the one hand, the photographicportrait extends, accelerates, popularizes, and degrades a tradi- tional function.This function,which can be said to have taken its early modern formin the seventeenthcentury, is that of providingfor the ceremonial presentation of the bourgeois self. Photography subverted the privileges inherentin portraiture,but without any more extensive leveling of social relationships, these privileges could be reconstructedon a new basis. That is, photography could be assigned a proper role withina new hierarchyof taste. Honorificcon- ventions were thus able to proliferate downward.5 At the same time,

3. WilliamHenry Fox Talbot, ThePencil of Nature, 1844, facsimileedition, New York, Da Capo, 1968,pl. 6, n.p. 4. Ibid., pl. 3. 5. The clearestof the early,optimistic understandings of photography'srole withina new hierarchyof taste, necessitating a restructuring ofthe portrait labor market along industrial lines, can be foundin an unsignedreview by Elizabeth Eastlake, "Photography," Quarterly Review, vol. 101, no. 202 (April1857), pp. 442-468. The Bodyand theArchive 7

photographic portraiturebegan to performa role no painted portraitcould have performedin the same thoroughand rigorousfashion. This role derived, not fromany honorificportrait tradition, but fromthe imperativesof medical and anatomical illustration.Thus photographycame to establish and delimit the terrainof the other,to defineboth thegeneralized the typology- and the look-- contingentinstance of deviance and social pathology. Michel Foucault has argued, quite crucially, that it is a mistake to describe the new regulatorysciences directed at the body in the early nineteenth century as exercises in a wholly negative, repressive power. Rather, social power operates by virtue of a positive therapeuticor reformativechan- neling of the body.6 Still, we need to understand those modes of instrumental realism that do in factoperate according to a very explicitdeterrent or repressive logic. These modes constitutethe lower limitor "zero degree"of socially in- strumentalrealism. Criminal identificationphotographs are a case in point, since they are designed quite literallyto facilitatethe arrestof theirreferent.7 I will argue in the second part of this essay that the semantic refinementand rationalization of preciselythis sort of realism was central to the process of defin- ing and regulatingthe criminal. But first,what general connections can be charted between the honorific and repressivepoles of portraitpractice? To the extentthat bourgeois order de- pends upon the systematicdefense of social relationsbased on privateproperty, to the extentthat the legal basis of the selflies in the model of propertyrights, in what has been termed"possessive individualism,"every proper portraithas its lurking, objectifyinginverse in the files of the police. In other words, a covert Hobbesian logic links the terrainof the "National Gallery" with that of the "Police Act."8

6. See Michel Foucault, Disciplineand Punish: The Birthof thePrison, trans. Alan Sheridan, New York, Pantheon, 1977, and, The Historyof Sexuality,Volume I: An Introduction,trans. Robert Hurley,New York,Pantheon, 1978. 7. Any photographsthat seek to identifya target,such as militaryreconnaissance photographs,operate according to thesame generallogic. See my 1975essay "The InstrumentalImage: Steichen at War," in Photographyagainst the Grain: Essays and PhotoWorks, 1973-1983, Halifax, The Pressof theNova ScotiaCollege of Art and Design, 1984. 8. The theoreticalground for the constructionof a specificallybourgeois subject can be found in Hobbes'sLeviathan (1651). C. B. Macphersonhas arguedthat Hobbes's axiomatic positing of an essentiallycompetitive individual human "nature" was in factquite specific to a developingmarket society,moreover, to a marketsociety in whichhuman labor power increasingly took the form of an alienablecommodity. As Hobbes put it,"The Valueor WORTH ofa man, is as of all things,his Price;that is to say, so muchas wouldbe givenfor the use ofhis Power:and thereforeis notab- solute;but a thingdependent on theneed andjudgement of another" (Thomas Hobbes, Leviathan, Harmondsworth,Penguin, 1968, Chap. 10, pp. 151-152. See Macpherson'sintroduction to this edition and his PoliticalTheory of Possessive Individualism: Hobbes to Locke, London, Oxford University Press,1962). Whileit would be farfetchedto presentHobbes as a theoristof the "bourgeois portrait," it is interestingto notehow he definedindividual autonomy and itsrelinquishment through contractual obligationin termsof dramaturgical metaphors, thus distinguishing between two categories of the 8 OCTOBER

In the mid-nineteenthcentury, the terms of this linkage between the sphere of culture and that of social regulation were specificallyutilitarian.9 Many of the early promotersof photographystruck up a Benthamite chorus, stressingthe medium's promise fora social calculus of pleasure and discipline. Here was a machine forproviding small doses of happiness on a mass scale, for contributingto JeremyBentham's famous goal: "the greatesthappiness of the greatestnumber."'0 Thus the photographicportrait in particularwas welcomed as a socially ameliorative as well as a socially repressive instrument.Jane Welsh Carlyle voiced characteristichopes in 1859, when she described inex- pensive portraitphotography as a social palliative: Blessed be the inventor of photography. I set him even above the inventorof chloroform!It has given more positive pleasure to poor sufferinghumanity than anythingthat has been "cast up" in my time .- this art, by which even the poor can possess themselves of tolerable likenesses of their absent dear ones."

In the United States, similar but more extensive utilitarianclaims were made by the portraitphotographer Marcus Aurelius Root, who was able to ar- ticulate the connection between pleasure and discipline, to argue explicitlyfor a moral economy of the image. Like Carlyle, he stressedthe salutoryeffects of photographyon working-classfamily life. Not only was photographyto serve as a means of cultural enlightenmentfor the workingclasses, but familyphoto- graphs sustained sentimentalties in a nation of migrants. This "primal house- hold affection"served a socially cohesive function,Root argued - articulatinga nineteenth-centuryfamilialism that would surviveand become an essentialideo-

person,the "Author"and the "Actor"(Leviathan, Chap. 16, pp. 217-218). The analogy be- tweensymbolic representation and political-legalrepresentation is central to his thought. (An amus- inghistory of portraitphotography could be writtenon thevicissitudes of the Hobbesian struggle betweenphotographer and sitter,both in theactual portrait encounter and in thesubsequent recep- tionof portrait photographs.) Furthermore,the frontispieceto Leviathantook the formof an allegoricalportrait. The commonwealth,or state,is literallyembodied in thefigure of a sovereign,an "artificialman," whose bodyis itselfcomposed of a multitudeof bodies, all ofwhom have ceded a portionof their individual powerto thecommonwealth in orderto preventthe civil war that would inevitably result from their uncheckedpursuit of "natural"appetites. Thus the"body" of the Leviathanis a kindof pressure vessel,containing explosive natural forces. This imageis perhapsthe first attempt to diagramthe socialfield visually. As such,it has a definite,if usually indirect, resonance in nineteenth-century attemptsto constructvisual metaphors for the conceptual models of thenew socialsciences. 9. "The utilitariandoctrine . . . is at bottomonly a restatementof theindividualist principles whichwere worked out in theseventeenth century: Bentham built on Hobbes"(C. B. Macpherson, Political Theoryof PossessiveIndividualism, p. 2). 10. JeremyBentham, "A Fragmenton Government"(1776), in Mary P. Mack, ed., A Bentham Reader,New York, Pegasus, 1969, p. 45. 11. Quoted in Helmut Gernsheim, The Historyof Photography.- From the Camera Obscura to theBe- ginningof theModern Era, New York, McGraw-Hill, 1969, p. 239. The Bodyand theArchive 9

logical featureof American mass culture. Furthermore,widely distributedpor- traits of the great would subject everyday experience to a regular parade of moral exemplars. Root's concern for respectabilityand order led him to ap- plaud the adoption of photographyby the police, arguing that convicted of- fenderswould "not find it easy to resume their criminal careers, while their faces and general aspects are familiarto so many, especially to the keen-sighted detectivepolice."''12 The "so many" is significanthere, since it implicitlyenlists a wider citizenryin the vigilant work of detection. Thus Root's utilitarianism comes full circle. Beginning with cheaply affordableaesthetic pleasures and moral lessons, he ends up with the photographicextension of that exemplary utilitariansocial machine, the Panopticon.'3

12. Marcus Aurelius Root, The Cameraand thePencil, 1864, reprint,Pawlett, Vermont, Helios, 1971, pp. 420-421. 13. The Panopticon, or Inspection House, was Jeremy Bentham's proposal, writtenin 1787, foran architecturalsystem of social discipline, applicable to prison, factory,workhouse, asylum, and school. The operative principles of the Panopticon were isolation and perpetual surveillance. Inmates were to be held in a ring of individual cells. Unable to see into a central observation tower, theywould be forcedto assume that theywere watched continually. (As Hobbes remarked over a centuryearlier, "the reputation of Power is Power.") The beneficial effectsof this program were trumpetedby Bentham in the famous opening remarksof his proposal: "Morals reformed- health preserved- industry invigorated- instruction diffused- public burdens lightened-- Economy seated, as it were, upon a rock- all by a simple idea of architecture"(John Bowring, ed., The WorksofJeremy Bentham, vol. 4, London, Simpkin, Marshall, 1843, p. 49). With Ben- tham the principle of supervision takes on an explicit industrial capitalist character: his prisons were to functionas profit-makingestablishments, based on the private contracting-outof convict labor. Bentham was a prototypicalefficiency expert. (On these last two points see, respectively, Gertrude Himmelfarb, "The Haunted House ofJeremyBentham," in VictorianMinds, New York, Knopf, 1968, pp. 32-81; and Daniel Bell, "Work and Its Discontents," in The End ofIdeology: On theExhaustion of PoliticalIdeas in theFifties, Glencoe, Illinois, Free Press, 1960, pp. 227-274.) For Foucault, "Panopticism" provides the central metaphor formodern disciplinarypower based on isolation, individuation, and supervision (Disciplineand Punish,pp. 195-228). Foucault traces the "birthof the prison" only to the 1840s, just when photographyappears with all of its instrumental promise. Given the central optical metaphor in Foucault's work, a reading of the subsequent development of disciplinary systems would need logically to take photography into account. John Tagg has writtena Foucauldian account of the "panoptic" character of early police and psychiatricphotography in Britain. While I am in frequentagreement with his argument, I disagree with his claim that the "cumbersome architecture"of the Panopticon became redundant with the development of photography ("Power and Photography: Part 1, A Means of Surveil- lance: The Photograph as Evidence in Law," ScreenEducation, no. 36 [Winter 1980], p. 45). This seems to accord too much power to photography,and to imply that domination operates entirely by the forceof visual representation.To suggest that cameras replaced prisons is more than a little hyperbolic. The fact that Bentham's plan was never realized in the form he proposed has perhaps contributed to the confusion; models are more easily transformedinto metaphors than are realized projects. Once discourse turns on metaphor, it becomes a simple matterto substitute a photographic metaphor foran architecturalone. My main point here is that any historyof disciplinary institutionsmust recognize the multiplicityof material devices involved--some literally concrete- in tracing not only the importance of surveillance, but also the continued importance of confinement.After all, Bentham's proposal was partially realized in the cellular and separate systems of confinementthat emerged in the nineteenth century. At least one "genuine" panopticon prison was constructed:the Stateville Penitentiaryin Illinois, built between 1916 and 1924. (For works on early prison history,see D. Melossi and M. Pavarini, The Prisonand theFactory. 10 OCTOBER

Notwithstandingthe standard liberal accounts of the historyof photography, the new medium did not simplyinherit and "democratize"the honorific functionsof bourgeois portraiture.Nor did police photographysimply function repressively,although it is foolish to argue that the immediate function of police photographs was somehow more ideological or positivelyinstrumental than negatively instrumental. But in a more general, dispersed fashion, in servingto introducethe panoptic principle into daily life, photographywelded the honorificand repressive functionstogether. Every portraitimplicitly took its place withina social and moral hierarchy.The privatemoment of sentimen- tal individuation, the look at the frozen gaze-of-the-loved-one,was shadowed by two othermore public looks: a look up, at one's "betters,"and a look down, at one's "inferiors."Especially in the United States, photographycould sustain an imaginary mobilityon this vertical scale, thus provoking both ambition and fear, and interpellating,in class terms, a characteristically"petit-bourgeois" subject. We can speak then of a generalized, inclusive archive,a shadowarchive that encompasses an entire social terrainwhile positioning individuals withinthat terrain.4" This archive contains subordinate, territorializedarchives: archives whose semantic interdependenceis normallyobscured by the "coherence"and "mutual exclusivity"of the social groups registeredwithin each. The general, all-inclusivearchive necessarilycontains both the traces of the visible bodies of heroes, leaders, moral exemplars,celebrities, and thoseof the poor, the diseased, the insane, the criminal, the nonwhite,the female, and all otherembodiments of the unworthy.The clearest indication of the essential unityof this archive of images of the body lies in the fact that by the mid-nineteenthcentury a single hermeneutic paradigm had gained widespread prestige. This paradigm had

Origins of the PenitentiarySystem, trans. Glynis Cousin, London, Macmillan, 1981; David Rothman,The Discovery of the Asylum: Social Order and Disorder in theNew Republic, Boston, Little, Brown,1971; and Michael Ignatieff,A JustMeasure of Pain: ThePenitentiary inthe Industrial Revolu- tion,1750-1850, London, Macmillan, 1978.) Certainly prison architectureand the spatial positioning of prisons in the larger environ- ment remain matters of crucial importance. Especially in the United States, where economic crisis and Reaganite judicial tough-mindednesshave lead to record prison populations, these are paramount issues of what is euphemistically called "public policy." In fact, the currentwave of ambitious prison building has led to at least one instance of (postmodern?) returnto the model of the Panopticon. The new Montgomery County Detention Center in Virginia was designed by prison architectJames Kessler according to a "new" principle of "podular/directsupervision." In this scaled-down, rumpus-room version of the Panopticon, inmates can see into the central control room fromwhich they are continually observed (see Benjamin Forgey, "Answering the Jail Question," The WashingtonPost, August 2, 1986, pp. G1-G2). 14. For earlier arguments on the archival paradigm in photography, see Rosalind Krauss, "Photography's Discursive Spaces: Landscape/View," The ArtJournal, vol. 42, no. 4 (Winter 1982), pp. 311-319; and Allan Sekula, "Photographybetween Labour and Capital," in B. Buchloh and R. Wilkie,eds., MiningPhotographs and OtherPictures: Photographs byLeslie Shedden, Halifax, The Press of the Nova Scotia College of Art and Design, 1983, pp. 193-268. The Bodyand theArchive 11

two tightlyentwined branches, physiognomyand phrenology.Both shared the belief that the surface of the body, and especially the face and head, bore the outward signs of inner character. Accordingly,in revivingand to some extent systematizingphysiognomy in the late 1770s, Johann Caspar Lavater argued that the "originallanguage of Nature, writtenon the face of Man" could be deciphered by a rigorousphysiog- nomic science.15Physiognomy analytically isolated the profileof the head and the various anatomic featuresof the head and face, assigninga characterological significanceto each element: forehead, eyes, ears, nose, chin, etc. Individual character was judged through the loose concatenation of these readings. In both its analytic and syntheticstages, this interpretiveprocess required that distinctiveindividual featuresbe read in conformityto type. Phrenology,which emerged in the firstdecade of the nineteenthcentury in the researches of the Viennese physician Franz Josef Gall, sought to discern correspondences between the topographyof the skull and what were thoughtto be specificlocalized mental facultiesseated withinthe brain. This was a crude forerunnerof more modern neurological attemptsto map out localized cerebral functions. In general, physiognomy,and more specificallyphrenology, linked an everyday nonspecialist empiricismwith increasinglyauthoritative attempts to medicalize the studyof the mind. The ambitious effortto constructa materialist science of the selfled to the dissection of brains, including those of prominent phrenologists,and to the accumulation of vast collectionsof skulls. Eventually this effortwould lead to a volumetricsof the skull, termed craniometry.But presumably any observant reader of one of the numerous handbooks and manuals of phrenologycould master the interpretivecodes. The humble origins of phrenological research were described by Gall in these terms: I assembled a large number of persons at my house, drawn fromthe lowest classes and engaged in various occupations, such as fiacre driver, street porter and so on. I gained their confidence and in- duced them to speak franklyby givingthem money and having wine and beer distributedto them. When I saw that they were favorably disposed, I urged them to tell me everythingthey knew about one another,both theirgood and bad qualities, and I carefullyexamined their heads. This was the origin of the craniological chart that was seized upon so avidly by the public; even artiststook it over and dis- tributeda large number among the public in the formof masks of all kinds.16

15. John [ sic] Caspar Lavater, Preface to Essays on PhysiognomyDesigned to Promote the Knowledge and theLove ofMankind, vol. 1, trans. Henry Hunter, London, J. Murray, 1792, n.p. 16. Quoted in Louis Chevalier, LabouringClasses and DangerousClasses in Paris duringthe First Half of theNineteenth Century, trans. Frank Jellineck, London, Routledge, 1973, p. 411. 12 OCTOBER

The broad appeal and influenceof thesepractices on literaryand artisticrealism, and on the general culture of the mid-nineteenth-centurycity is well known."7 And we understand the culture of the photographicportrait only dimly if we failto recognizethe enormousprestige and popularityof a generalphysiognomic paradigm in the 1840s and 1850s. Especially in the United States, the prolifera- tion of photographyand that of phrenologywere quite coincident. Since physiognomy and -phrenologywere comparative, taxonomic disciplines, they sought to encompass an entirerange of human diversity.In this respect, these disciplines were instrumentalin constructingthe very archive theyclaimed to interpret.Virtually every manual deployed an array of individual cases and types along a loose set of "moral, intellectual,and animal" con- tinua. 18Thus zones of genius, virtue, and strengthwere charted only in relation to zones of idiocy, vice, and weakness. The boundaries between these zones were vaguely demarcated; thus it was possible to speak, forexample, of "moral idiocy." Generally, in this pre-evolutionarysystem of difference,the lower zones shaded offinto varieties of animality and pathology. In the almost exclusive emphasis on the head and face we can discoverthe idealist secretlurking at the heart of these putativelymaterialist sciences. These were discoursesof the headforthe head. Whateverthe tendencyof physiognomic or phrenologic thought- whetherfatalistic or therapeuticin relation to the in- exorable logic of the body's signs, whether uncompromisinglymaterialist in tone or vaguely spiritualistin relation to certain zones of the organic, whether republican or elitist in pedagogical stance--these disciplines would serve to legitimateon organic grounds the dominion of intellectualover manual labor. Thus physiognomyand phrenologycontributed to the ideological hegemonyof a capitalism that increasinglyrelied upon a hierarchical division of labor, a capitalism that applauded its own progress as the outcome of individual cleverness and cunning. In claiming to provide a means for distinguishingthe stigmata of vice fromthe shiningmarks of virtue, physiognomyand phrenologyoffered an essential hermeneuticservice to a world of fleetingand oftenanonymous market transactions. Here was a method for quickly assessing the character of strangersin the dangerous and congested spaces of the nineteenth-centurycity. Here was a gauge of the intentionsand capabilities of the other. In the United States in the 1840s, newspaper advertisementsfor jobs frequentlyrequested

17. In addition to Chevalier's book just cited, see Walter Benjamin's 1938 essay, "The Paris of the Second Empire in Baudelaire," in CharlesBaudelaire: A LyricPoet in theEra ofHigh Capitalism, trans. Harry Zohn, London, New Left Books, 1973, pp. 35-66. See also Judith Wechsler, A Human Comedy:Physiognomy and Caricaturein NineteenthCentury Paris, Chicago, University of Chicago Press, 1982. For specific histories of phrenology, see David de Guistino, Conquestof Mind: Phrenologyand VictorianSocial Thought,London, Croom Helm, 1975; and John Davies, Phrenology:Fad and Science,New Haven, Yale University Press, 1955. 18. Lavater, vol. 1, p. 13. The Bodyand theArchive 13

that applicants submit a phrenological analysis.19Thus phrenologydelivered the moral and intellectual"facts" that are today delivered in more "refined"and abstract formby psychometriciansand polygraph experts. Perhaps it is no surprise,then, that photographyand phrenologyshould have met formallyin 1846 in a book on "criminaljurisprudence." Here was an opportunityto lend a new organic facticityto the already established medical and psychiatricgenre of the case study.20A phrenologicallyinclined Amer- ican penal reformerand matron of the women's prison at Sing Sing, Eliza Farnham commissioned Mathew Brady to make a series of portraitsof inmates at two New York prisons. Engravings based on these photographs were ap- pended to Farnham's new edition, entitled Rationaleof Crime,of a previously unillustrated English work by Marmaduke Sampson. Sampson regarded criminal behavior as a formof "moral insanity." Both he and Farnham sub- scribed to a variant of phrenologythat argued forthe possibilityof therapeutic modification or enhancement of organically predetermined characteristics. Presumably, good organs could be made to triumphover bad. Farnham's con- tributionis distinctivefor its unabashed nonspecialist appeal. She sought to speak to "the popular mind of Republican America," in presentingan argument for the abolition of the death penalty and the establishmentof a therapeutic system of treatment.21Her contribution to the book consisted of a polemical introduction,extensive notes, and several appendices, including the illustratedcase studies. Farnham was assisted in her selection of case-study subjects by the prominent New York publisher-entrepreneurof phrenology, Lorenzo Fowler, who clearly lent furtherauthority to the sample. Ten adult prisoners are pictured, evenly divided between men and women. Three are identifiedas Negro, one as Irish, one as German; one woman is identifiedas a "Jewessof German birth,"another as a "half-breedIn- dian and negro." The remaining three inmates are presumably Anglo-Saxon, but are not identifiedas such. A series of eight pictures of child inmates is not annotated in racial or ethnicterms, although one child is presumablyblack. Al- thoughFarnham professeda variant of phrenologythat was not overtlyracist - unlike other pre-Darwinian head analysts who sought conclusive proof of the "separate creation" of the non-Caucasian races --this differentialmarking of race and ethnicityaccording to age is significantin otherways. Afterall, Farn- ham's work appeared in an American context- characterized by slavery and the massive immigration of Irish peasants--that was profoundly stratified

19. Davies, p. 38. 20. On the historyof the illustratedpsychiatric case study, see Sander Gilman, Seeingthe Insane, New York, J. Wiley, 1982. 21. Eliza Farnham, "IntroductoryPreface" to Marmaduke Sampson, Rationaleof Crimeand its AppropriateTreatment, Being a Treatiseon CriminalJurisprudence Considered in Relation to Cerebral Organization,New York, Appleton, 1846, p. xiii. 14 OCTOBER

along these lines. By markingchildren less in racial and ethnicterms, Farnham avoided stigmatizingthem. Thus children in general were presented as more malleable figuresthan adults. Children were also presented as less weighted down by criminal biographies or by the habitual exercise of their worst faculties. Despite the fact that some of these boys were explicitlydescribed as incorrigibles, children provided Farnham with a general figure of moral renewal. Because theirpotential for"respectability" was greaterthan thatof the adult offenders,they were presented as miniature versions of their potential adult-male-respectable-Anglo-Saxon-proletarianselves. Farnham, Fowler, and Brady can be seen as significantinventors of that privilegedfigure of social reformdiscourse: the figureof the child rescued by a paternalisticmedicosocial science.22 Farnham's concerns touch on two of the central issues of nineteenth- centurypenal discourse: the practical drawing of distinctionsbetween incor- rigible and pliant criminals, and the disciplined conversion of the reformable into "useful"proletarians (or at least into useful informers).Thus even though she credited several inmates with "well developed" intellects,and despite the fact that her detractorsaccused her of Fourierism, her reformistvision had a definiteceiling. This limit was definedquite explicitlyby the conclusion of her study. There she underscored the baseness shared by all her criminal subjects by illustratingthree "heads of persons possessing superior intellect"(two of which, both male, were treated as classical busts). Her readers were asked to note the "strikingcontrast."23 I emphasize thispoint because it is emblematicof the manner in which the criminalarchive came into existence. That is, it was only on the basis of mutual comparison, on the basis of the tentativeconstruction of a larger, "universal" archive, that zones of deviance and respectabilitycould be clearlydemarcated. In this instance of the firstsustained application of photographyto the task of phrenological analysis, it seems clear that the comparative descriptionof the criminalbody came first.The book ends with a self-congratulatorymirror held up to the middle-classreader. It is strikingthat the pictoriallabor behind Farn- ham's criminalsample was thatof Brady, who devoted virtuallyhis entireante- bellum career to the constructionof a massive honorificarchive of photographs of "illustrious,"celebrated, and would-be celebrated American figures.24

22. For a readingof the emergence of this system in France,see Jacques Donzelot, The Policing ofFamilies, trans. Robert Hurley, New York, Pantheon,1979. Donzelot seems to place inor- dinate blame on women forthe emergenceof a "tutelary"mode of social regulation.For a Marxist-feministcritique of Donzelot,see MichelleBarrett and Mary McIntosh,The Anti-Social Family,London, New LeftBooks, 1982. 23. Sampson,p. 175. 24. See Madeline Stern, "Mathew B. Brady and the Rationaleof Crime,"The QuarterlyJournal ofthe Library of Congress, vol. 31, no. 3 (July1974), pp. 128-135;and Alan Trachtenberg,"Brady's Portraits,"The YaleReview, vol. 73, no. 2 (Winter1984), pp. 230-253. The Bodyand theArchive 15

Thus far I have described a number of early attempts,by turns comic, speculative, and practical, to bring the camera to bear upon the body of the criminal. I have also argued, followingthe general line of investigationcharted in the later works of Foucault, that the position assigned the criminalbody was a relativeone, that the inventionof the modern criminal cannot be dissociated fromthe constructionof a law-abiding body- a body that was eitherbourgeois or subject to the dominion of the bourgeoisie. The law-abiding body recognized its threateningother in the criminal body, recognized its own acquisitive and aggressive impulses unchecked, and sought to reassure itselfin two contradic- tory ways. The firstwas the invention of an exceptional criminal who was indistinguishablefrom the bourgeois, save for a conspicuous lack of moral in- hibition: herein lay the figureof the criminal genius.25The second was the in-

25. On thispoint see Michel Foucault,"Prison Talk," in Power/Knowledge:Selected Interviews and OtherWritings, 1972-1977, ed. Colin Gordon,New York, Pantheon,1980, p. 46.

IR NDIX 174 .TION.\.LE. ' 17 .'IfI

lIl-:ADS1 l ES \ ILI:i 'I','l

'II lw i,],mvimlrdI.;1viw l sI r; introduc'l t"rl ihpii.4

- w(VinLg IC t-lk nllg conlrast litweent the,f erc' ral ,h vlI -iweiuhi Ic , a wc have i ecitie.erilii intlnd! ti.. Mll,, ;irr,ell ,~\l mt:l.'t jilh si lcrio ll lpower.-roi intellect andl gclt- I0IIlt.

I".

B. F. is one of lie iia of itiL,.,, r f itil 1"ruIi. 1i is partially idiwic. mid th very nimlprfi-ect t(i'loi e wti superior portion of tho lrail. withI i twill .i/ Cf ithe wIh .. clearly indicates IIIth h rnet.r, "fhi m1 nt l ,'t; iti . afords a strikili c !'~I-i it riI' -tatl A., i. ti ii.in . . harmony wvilitic :a t.It l lil .. ltiw . tflt.1I1i1 II ,' itI twvoitldi id1ual[ 13. F'. I- \ n'tI -. drI'.I ; oIpl jIj-l..itttlY.K l,'; J (ti c +,alazly cluvoti-.,l ,,w?l,,m1 t,ll,++ -elitim eltt.

I.,i ti i ; 111. 1 ,1- I .I ,

FromEliza Farnham,Appendix to Marmaduke Sampson, Rationaleof Crime, 1846. i-i-i -::-:-;-::-_I II-: --. -- ~:i: ~-: ir: iI---i .....i ! ii-::ii:I ". :::i--i~ ii

iii1:ii-i-i i:-: :::: ::::-: : iiiii!+ ...... :::;:i:-~-

: i :i "i l : ? i : i !::i ;:ii'

FromAlphonse Bertillon, Service d'identification. Expositionuniverselle de Chicago, 1893. (Album collectionNational Galleryof Canada, Ottawa.)

ventionof a criminalwho was organicallydistinct from the bourgeois: a biotype. The science of criminologyemerged fromthis latter operation. A physiognomiccode of visual interpretationof the body's signs- specifi- cally the signs of the head- and a technique of mechanized visual representation intersectedin the 1840s. This unifiedsystem of representationand inter- pretationpromised a vast taxonomic orderingof images of the body. This was an archival promise. Its realizationwould seem to be groundedprimarily in the technicalrefinement of strictlyoptical means. This turnsout not to be the case. I am especially concerned that exaggerated claims not be made for the powers of optical realism, whetherin a celebratoryor criticalvein. One danger lies in constructingan overly monolithic or unitary model of nineteenth- centuryrealist discourse. Within the ratherlimited and usually ignored fieldof instrumentalscientific and technicalrealism, we discover a house divided. No- where was this division more pronounced than in the pursuit of the criminal body. If we examine the manner in which photographywas made usefulby the late-nineteenth-centurypolice, we findplentiful evidence of a crisis of faithin optical empiricism. In short, we need to describe the emergence of a truth- apparatus that cannot be adequately reduced to the optical model provided by the camera. The camera is integratedinto a larger ensemble: a bureaucratic- clerical-statisticalsystem of "intelligence."This systemcan be described as a sophisticatedform of the archive. The central artifactof this systemis not the camera but the filingcabinet. The Bodyand theArchive 17

II.

The institutionof the photographic archive received its most thorough early articulation in precise conjunction with an increasinglyprofessionalized and technologicalmode of police work and an emergingsocial science of criminology. This occurred in the 1880s and 1890s. Why was the model of the archive of such import for these linked disciplines? In structuralterms, the archive is both an abstract paradigmatic entity and a concrete institution.In both senses, the archive is a vast substitutionset, providing fora relation of general equivalence between images. This image of the archive as an encyclopedic repositoryof exchangeable images was articulated most profoundlyin the late 1850s by the American physician and essayist Oliver Wendell Holmes when he compared photographs to paper currency.26 The capacity of the archive to reduce all possible sightsto a singlecode of equivalence was grounded in the metrical accuracy of the camera. Here was a medium fromwhich exact mathematicaldata could be extracted,or as the physicist Frangois Arago put it in 1839, a medium "in which objects preserve mathe- matically their forms."27 For nineteenth-centurypositivists, photography doubly fulfilledthe Enlightenmentdream of a universal language: the universal mimeticlanguage of the camera yielded up a higher,more cerebral truth,a truththat could be utteredin the universal abstract language of mathematics. For thisreason, photographycould be accommodated to a Galilean vision of the world as a book "writtenin the language of mathematics."Photography promised more than a wealth of detail; it promised to reduce nature to its geo- metricalessence. Presumably then, the archive could provide a standard physiognomic gauge of the criminal, could assign each criminal body a relative and quantitative position within a larger ensemble. This archival promise was frustrated,however, both by the messy contin- gency of the photographand by the sheer quantityof images. The photographic archive's components are not conventional lexical units, but ratherare subject to the circumstantialcharacter of all that is photographable. Thus it is absurd to imagine a dictionaryof photographs, unless one is willing to disregard the specificityof individual images in favor of some model of typicality,such as that underlyingthe iconographyof Vesalian anatomy or of most of the plates accompanying the Encyclopidieof Diderot and d'Alembert. Clearly, one way of "taming"photography is by means of this transformationof the circumstantial and idiosyncraticinto the typical and emblematic. This is usually achieved by stylisticor interpretivefiat, or by a sampling of the archive's offeringsfor a

26. Oliver Wendell Holmes, "The Stereoscope and the Stereograph," AtlanticMonthly, vol. 3, no. 20 (June 1859), p. 748. For a more extensive treatmentof this issue, see my 1981 essay, "The Trafficin Photographs," in Photographyagainst the Grain, pp. 96-101. 27. Frangois Arago, letter to Duchatel, in Gernsheim, Daguerre,p. 91. 18 OCTOBER

"representative"instance. Another way is to invent a machine, or rather a clerical apparatus, a filingsystem, which allows the operator/researcher/editor to retrievethe individual instance fromthe huge quantity of images contained within the archive. Here the photograph is not regarded as necessarilytypical or emblematic of anything,but only as a particular image which has been isolated forpurposes of inspection. These two semantic paths are so fundamental to the culture of photographic realism that their very existence is usually ignored. The differencebetween these two models of photographic meaning are played out in two differentapproaches to the photographic representationof the criminal body: the "realist" approach, and by realism here I mean that venerable (medieval) philosophical realism that insists upon the truthof general propositions,on the realityof species and types, and the equally venerable "nominalist"approach, which denies the reality of generic categories as anything other than mental constructs.The firstapproach can be seen as overtly theoreticaland "scientific"in its aims, ifmore covertlypractical. The othercan be seen as overtlypractical and "technical"in its aims, ifonly covertlytheoreti- cal. Thus the would-be scientistsof crime sought a knowledge and masteryof an elusive "criminal type." And the "technicians"of crime sought knowledge and masteryof individual criminals. Herein lies a terminologicaldistinction, and a divisionof labor, between "criminology"and "criminalistics."Criminology hunted "the"criminal body. Criminalisticshunted "this"or "that"criminal body. Contraryto the commonplace understandingof the "mug shot"as the very exemplar of a powerful,artless, and whollydenotative visual empiricism,these early instrumentaluses of photographicrealism were systematizedon the basis of an acute recognitionof the inadequaciesand limitationsof ordinaryvisual empiricism. Thus two systemsof descriptionof the criminal body were deployed in the 1880s; both sought to ground photographic evidence in more abstract statisticalmethods. This merger of optics and statisticswas fundamental to a broader integrationof the discourses of visual representationand those of the social sciences in the nineteenthcentury. Despite a common theoreticalsource, the intersectionof photographyand statisticsled to strikinglydifferent results in the work of two differentmen: Alphonse Bertillon and Francis Galton. The Paris police officialAlphonse Bertillon invented the firsteffective modern systemof criminalidentification. His was a bipartitesystem, positioning a "microscopic" individual record within a "macroscopic" aggregate. First, he combined photographic portraiture,anthropometric description, and highly standardized and abbreviated writtennotes on a singlefiche, or card. Second, he organized these cards within a comprehensive, statisticallybased filing system. The English statistician and founder of eugenics, Francis Galton, invented a method of composite portraiture.Galton operated on the peripheryof criminology.Nonetheless, his interestin heredityand racial "betterment"led The Bodyand theArchive 19

him tojoin in the search fora biologicallydetermined "criminal type." Through one of his several applications of composite portraiture,Galton attempted to constructa purelyoptical apparition of the criminal type. This photographicim- pression of an abstract,statistically defined, and empiricallynonexistent criminal face was both the most bizarre and the most sophisticatedof many concur- rentattempts to marshall photographicevidence in the search forthe essence of crime. The projects of Bertillon and Galton constitutetwo methodologicalpoles of the positivist attempts to define and regulate social deviance. Bertillon sought to individuate. His aims were practical and operational, a response to the demands of urban police work and the politicsof fragmentedclass struggle during the Third Republic. Galton sought to visualize the generic evidence of hereditarian laws. His aims were theoretical, the result of eclectic but ulti- matelysingle-minded curiosities of one of the last Victorian gentleman-amateur scientists.Nonetheless, Bertillon'swork had its own theoreticalcontext and implications,just as Galton's grimlyplayful research realized its practical implications in the ideological and political program of the international eugenics movement. Both men were committedto technologiesof demographic regulation. Bertillon'ssystem of criminal identificationwas integral to the effortsto quarantine permanentlya class of habitual or professionalcriminals. Galton sought to intervene in human reproduction by means of public policy, en- couraging the propagation of the "fit,"and discouragingor preventingoutright that of the "unfit." The idealist proclivities,territorialism, and status consciousness of intellectual history have prevented us from recognizing Bertillon and Galton's shared ground. While Galton has been considered a proper, if somewhat ec- centric,object of the historyof science, Bertillonremains an ignored mechanic and clerk, commemorated mostlyby anecdotal historiansof the police. In order to explore this terrain shared by a police clerk and gentleman statistician,I need to introduce a third figure. Both Bertillon's and Galton's projects were grounded in the emergence and codificationof social statisticsin the 1830s and 1840s. Both relied upon the central conceptual categoryof social statistics:the notion of the "average man" (l'hommemoyen). This concept was invented (I will argue shortlythat it was actually reinvented) by the Belgian astronomerand statisticianAdolphe Quetelet. Although less well remembered than Auguste Comte, Quetelet is the most significantother early architectof sociology. Certainlyhe laid the foundationsof the quantitativeparadigm in the social sciences. By seeking statisticalregularities in rates of birth, death, and crime, Quetelet hoped to realize the Enlightenmentphilosopher Condorcet's proposal fora "social mathematics,"a mathematicallyexact science thatwould discover the fundamentallaws of social phenomena. Quetelet helped to establish some of the firstactuarial tables used in Belgium, and to found in 1853 an internationalsociety for the promotion of statisticalmethods. As the philoso- 20 OCTOBER

pher of science Ian Hacking has suggested, the rise of social statisticsin the mid-nineteenthcentury was crucial to the replacement of strictlymechanistic theories of causality by a more probabilisticparadigm. Quetelet was a determinist, but he invented a determinismbased on iron laws of chance. This emergentparadigm would lead eventually to indeterminism.28 Who, or what, was the average man? A less flippantquery would be, how was the average man? Quetelet introducedthis compositecharacter in his 1835 treatiseSur l'homme.Quetelet argued thatlarge aggregatesof social data revealed a regularityof occurrence that could only be taken as evidence of determinate social laws. This regularityhad political and moral as well as epistemological implications: The greater the number of individuals observed, the more do individual peculiarities, whether physical or moral, become effaced, and leave in a prominentpoint of view the general facts,by virtueof which society exists and is preserved.29

Quetelet sought to move from the mathematicizationof individual bodies to that of society in general. In Sur l'hommehe charted various quantitative biographies of the productive and reproductive powers of the average man and woman. For example, he calculated the fluctuationof fecunditywith respect to female age. Using data from dynamometer studies, he charted the average muscular power of men and women of differentages. At the level of the social aggregate, lifehistory read as a graphic curve. (Here was prefiguration,in extreme form,of Zola's naturalism: a subliterary,quantitative narrative of the generalized social organism.) Just as Quetelet's early statisticalcontributions to the life insurance in- dustrycan be seen as crucial to the regularization of that organized formof gambling known as financecapital, so also his chartingof the waxing and wan- ing of human energiescan be seen as an attemptto conceptualize thatHercules of industrial capitalism, termed by Marx the "average worker,"the abstract embodimentof labor power in the aggregate.30And outside the sphereof waged work, Quetelet invented but did not name the figureof the average mother, crucial to the new demographic sciences which sought nervouslyto chart the relative numeric strengthsof class against class and nation against nation. For Quetelet the most emphatic demonstrationof the regularityof social

28. See Ian Hacking,"How ShouldWe Do theHistory of Statistics?" Ideology and Consciousness, no. 8 (Spring 1981), pp. 15-26; and "Biopowerand the Avalanche of PrintedNumbers," Humanitiesand Society, vol. 5, nos. 3-4 (Summerand Fall 1982), pp. 279-295. 29. Adolphe Quetelet, A Treatiseon Man and theDevelopment of His Faculties,trans. R. Knox, Edinburgh,Chambers, 1842, p. 6. 30. Karl Marx, Capital: A Critiqueof Political Economy, trans. Ben Fowkes, London, New Left Books,vol. 1, 1976, pp. 440-441. The Bodyand theArchive 21

phenomena was given by crime statistics."Moral statistics"provided the linch- pin forhis constructionof a "social physics"that would demolish the prestigeof moral paradigms groundedin freewill. The criminalwas no more than an agent of determiningsocial forces. Furthermore,crime statisticsprovided the synec- dochic basis fora broader descriptionof the social field.As Louis Chevalier has argued, Quetelet inaugurated a "quantitativedescription which took criminal statisticsas the startingpoint for a descriptionof urban living as a whole."31 Chevalier has argued furtherthat criminal statisticscontributed thus to a pervasive bourgeois conception of the essentiallypathological character of metropolitan life, especially in the Paris of theJuly Monarchy. Quetelet's terminological

Tailles des Belges de 18 20 ans.

11,55,3I1m,387 Im,459 im,591 Im,,643 m,695 Im,747 1m,799 1m,851

FromAdolphe Quetelet, Physique sociale, ou Essai sur le developpementdes facultesde l'homme,1869. contributionto this medicalization of the social fieldis evident in his reference to the statisticalstudy of crime as a formof "moral anatomy." Quetelet refinedhis notion of the "average man" with conceptual tools borrowed fromastronomy and probabilitytheory. He observed that large ag- gregates of social data- notably anthropometricdata- fell into a pattern cor- respondingto the bell-shaped curve derived by Gauss in 1809 in an attemptto determineaccurate astronomicalmeasurements from the distributionof random errorsaround a central mean. Quetelet came to regard this symmetricalbino-

31. Chevalier,p. 10. 22 OCTOBER

mial curve as the mathematicalexpression of fundamentalsocial law. While he admitted that the average man was a statisticalfiction, this fictionlived within the abstract configurationof the binomial distribution. In an extraordinary metaphoricconflation of individualdifference with mathematical error, Quetelet defined the central portion of the curve, that large number of measurements clustered around the mean, as a zone of normality.Divergent measurements tended toward darker regions of monstrosityand biosocial pathology.32 Thus conceived, the "average man" constitutedan ideal, not only of social health, but of social stabilityand of beauty. In interestingmetaphors, revealing both the astronomical sources and aesthetico-politicalambitions inherent in Quetelet's "social physics,"he definedthe social norm as a "centerof gravity," and the average man as "the type of all which is beautiful--of all which is good."33 Crime constituteda "perturbingforce," acting to throw the delicate balance of this implicitlyrepublican social mechanism into disarray. Although Quetelet was constructinga quantitative model of civil society and only indi- rectlydescribing the contours of an ideal commonwealth,his model of a gravi- tational social order bears strikingsimilarity to Hobbes's Leviathan.34 Like Hobbes, Quetelet began with atomized individual bodies and returnedto the image of the body in describingthe social aggregate. Quetelet worked, however, in a climate of physiognomicand phrenologic enthusiasm, and indeed early social statisticscan be regarded as a variant of physiognomy writlarge. For example, Quetelet accepted, despite his republicanism,the late- eighteenth-centurynotion of the cranialangle, which, as George Mosse has argued, emerges fromthe appropriationby preevolutionaryEnlightenment anthropologyof the classicist idealism of Wincklemann.35Based in part on the art-historicalevidence of noble Grecian foreheads, this racist geometrical fictiondefined a descending hierarchyof head types, with presumably upright Caucasian brows approaching this lost ideal more closely than did the presumably apelike brows of Africans. For his part, Quetelet was less interestedin a broadlyracist physical anthropology than in detectingwithin European society patternsof bodily evidence of deviation from"normality." It is understandable that he would be drawn to those variants of physiognomic thought which sought to systematizethe body's signs in terms of a quantifyinggeometrical

32. AdolpheQuetelet, Lettres sur la thioriedes probabilitis, Brussels, Academie Royale, 1846. (Letterson the Theory of Probability, trans. 0. G. Downes, London,Layton, 1849). See also Georges Canguilhem,On theNormal and thePathological, trans. Carolyn Fawcett, Boston, Reidel, 1978, pp. 86-104. 33. Quetelet,Treatise on Man, p. 100. 34. See note8. Of course,Quetelet's extreme determinist view of thesocial field was diamet- ricallyopposed to the contractualmodel of humanrelations advanced by Hobbes. 35. See George Mosse, Towardthe Final Solution.A Historyof European Racism, New York, Fertig,1978, pp. 17-34. The Bodyand theArchive 23

schema. From Quetelet on, biosocial statisticiansbecame increasinglyabsorbed withanthropometrical researches, focusingboth on the skeletal proportionsof the body and upon the volume and configurationof the head.36 The inherited idealist fascinationwith the uprightforehead can be detectedeven in Quetelet's model of an ideal society:he argued that social progresswould lead to a dimin- ished number of defectiveand inferiorcases, thus increasing the zone of normality. If we consider what this utopian projection meant in terms of the binomial curve, we have to imagine an increasinglypeaked, erect configuration: a classical ideal to a fault. Certainly physiognomyprovided a discursive terrainupon which art and the emergingbio-social sciences met during the middle of the nineteenthcentury.Quetelet's explicitlystated enthusiasm forthe model of artisticpractice is understandable in thiscontext, but the matteris more complicated. Despite the abstract characterof his procedures, Quetelet possessed the aestheticambition to compare his project to Diirer's studies of human bodily proportion. The statisticianargued thathis "aim had been, not only to go once more throughthe task of Albert [sic] Dfirer, but to execute it also on an extended scale."37Thus visual empiricism retained its prestige in the face of a new object- society- which could in no way be effectivelyor comprehensivelyvisualized.38

36. See Adolphe Quetelet, Anthropomitrie,ou mesuredes diffrentsfacultis de l'homme,Brussels, Muquardt, 1871. Quetelet sufferedfrom aphasia after 1855, and his later works tend to be repetitiousand incoherent(see Frank H. Hankins, AdolpheQuetelet as Statistician,New York, Co- lumbia University Press, 1908, pp. 31-32). On the intersection of anthropometryand race science, see Stephen Jay Gould, The Mismeasureof Man, New York, Norton, 1981. 37. Quetelet, Treatiseon Man, p. v. 38. Here are some ways in which Quetelet's position in relation to idealist aesthetic theory become very curious. The "average man" can be regarded as a bastard child of Kant. In the "Critique of AestheticalJudgement" Kant describes the psychological basis of the constructionof the empiricallybased "normal Idea" of human beauty, arguing that "the Imagination can, in all probability,actually though unconsciously let one image glide into another, and thus by the con- currence of several of the same kind come by an average, which serves as the common measure of all. Every one has seen a thousand full-grownmen. Now ifyou wish to judge of the normal size, estimatingit by means of comparison, the Imagination (as I think) allows a great number of images (perhaps the whole thousand) to fall on one another. If I am allowed here the analogy of optical presentation,it is the space where most of them are combined and inside the contour, where the place is illuminated with the most vivid colors, that the averagesize is cognizable; which, both in heightand breadth, is equally far removed fromthe extremebounds of the greatestand smallest stature. And this is the stature of a beautiful man" (Immanuel Kant, CritiqueofJudgement, trans. J. H. Bernard, London, Macmillan, 1914, pp. 87-88). This passage prefiguresnot only Quetelet but also-as we shall see--Galton. However, Kant was careful to respect differencesbetween normal Ideas of beauty appropriate to differentraces. On an empirical level, he constructedno hierarchy. Furthermore,he distinguished between the empirically based normal Idea, and the "Ideal of beauty," which is constructedin conformitywith a concept of morality. Quetelet can be accused of unwittinglycollapsing Kant's distinctionbetween the normal Idea and the Ideal, and thus fusingaesthetics and moralityon a purely quantitative basis, preparing thus the ground for Galton's plan for the engineering of human reproduction. Although Kant's more general proposal for a science of the human species based on the model of the natural sciences was known to Comte, Quetelet, "a stranger to all philosophical 24 OCTOBER

By the end of the nineteenthcentury, this essentiallyorganismic model of a visiblesocial fieldwas in crisis. The terms of Quetelet's honorificlinkage of an emergent statisticsto a venerable optical paradigm were explicitlyreversed. The French sociologistGabriel Tarde argued in 1883 that "a statisticalbureau mightbe compared to an eye or ear," claiming furtherthat "each of our senses gives us, in its own way and fromits special point of view, the statisticsof the externalworld. Their characteristicsensations are in a certainway theirspecial graphical tables. Every sensation . . . is only a number.""39Here the transition is made fromthe prestigeof the visual and the organic to the prestigeof institu- tionalized, bureaucratic abstraction. Tarde was a centralfigure, not only in the demise of organismicmodels of society, but also in the development of a French school of criminological thoughtduring the 1880s. Tarde was a magistrateduring his early career, and by 1894 became the head of the Bureau of Statisticswithin the Department of Justice in Paris, which made him the abstractoverseer of the quantitativeebbs and flowsof a regulated criminality.His background in legal theoryand practice led him to attempta criticismand modificationof Quetelet's extremedeter- minism, which had absolved the criminal of all responsibility. After all, classical legal theory was not about to abandon its ideological capacity to uphold the state's rightto punish criminals fortheir deeds. In 1890, Tarde advanced a notion of "criminal responsibility"based upon the continuityof individual identitywithin a shared social milieu, a milieu of "social similarity." Tarde's psychologicalmodel of individualityassumed an essential internalnar-

speculation," seems never to have read Kant (Joseph Lottin, Quetelet,statisticien et sociologue, Louvain, Institut supdrieur de philosophie, 1912, p. 367). Quetelet's persistentlikening of his project to the work of the visual artistcan certainlybe taken as emblematic of the fusion of idealist aesthetics with Enlightenment theories of social perfection. More specifically,however, Quetelet's evocations of art history- which extended to the measurement of classical sculpture and to long chronological tables of artistswho had dealt with problems of bodily proportion- can be seen as a legitimatingmaneuver to ward offaccusa- tions that his strictdeterminism obliterated the possibilityof a human creativitybased on the exercise of free will. (It was also an attempt to compare the average bodily types of "ancients" and "moderns.") Thus Quetelet colors his gray determinismwith a self-justifyinghint of romanticism. But this maneuver also converts the visual artist into a protoscientist,linking Quetelet to the emerging discourse of artisticrealism. (See his Anthropomitrie,pp. 61-169. In this work Quetelet constructeda visual diagram of the biographical course of an average body type frominfancy to old age, based on anthropometricaldata.) 39. Gabriel Tarde, "Archaeology and Statistics,"in The Laws ofImitation, trans. Elsie Parsons, New York, Henry Holt, 1903, pp. 134-135 (this essay firstappeared in the Revuephilosophique, October 1883). In an extraordinarypassage of the same essay Tarde compares the graphical curve for criminal recidivism with the "curve traced on [the] retina by the flightof [a] swallow," metaphoricallylinking within the same epistemological paradigm the work of Bertillonwith that of the physiologistEtienne Jules Marey, chronophotographerof human and animal locomotion (ibid., p. 133). The Bodyand theArchive 25

rative coherence of the self: "Identityis the permanence of the person, it is the personalitylooked at fromthe point of view of its duration."40 Tarde's rather nominalist approach to the philosophy of crime and punishment paralleled a more practical formulationby Alphonse Bertillon, directorof the IdentificationBureau of the Paris Prefectureof Police. In 1893, Bertillon offeredthe followingintroduction to his system,then in use for ten years, known variously as "Bertillonage"and the "signaletic notice": In prison practice the signaletic notice accompanies every reception and everydelivery of a human individuality;this registerguards the trace of the real, actual presence of the person sought by the admin- istrativeor judicial document. ... [The] task is always the same: to preserve a sufficientrecord of a personalityto be able to identifythe presentdescription with one which may be presentedat some future time. From this point of view signalment is the best instrumentfor the proofof recidivation, which necessarilyimplies the proofof identity.41 In effect,then, Bertillon'spolice archive functionedas a complex biographical machine which produced presumably simple and unambiguous results. He sought to identifyrepeat offenders,that is, criminalswho were liable to be considered "habitual" or "professional"in theirdeviant behavior. The concern with recidivismwas of profoundsocial importancein the 1880s. Bertillon,however, professedno theoryof a criminaltype, nor of the psychiccontinuities or discon- tinuities that might differentiate"responsible" criminals from "irresponsible" criminals. He was sensitive to the status hierarchybetween his Identification Bureau and the more "theoretical"mission of the Bureau of Statistics.(Bertillon was the son of a prominentanthropometrician, Louis Adolphe Bertillon, and seems to have labored mightilyto vindicate himselfafter an inauspicious start as a mere police clerk.) He was more a social engineer, an inventive clerk- technician, than a criminologist.He sought to ground police work in scientific principles,while recognizingthat most police operatives were unfamiliarwith consistentand rigorous empirical procedures. Part of his ambition was to ac- celerate the work of processingcriminals and to employ effectivelythe labors of unskilled clerks. He resembles in many respects his American contemporary, Frederick Winslow Taylor, the inventor of scientificmanagement, the first system of modern factorydiscipline. Bertillon can be seen, like Taylor, as a prophet of rationalization. Here is Bertillondescribing the rapidityof his process: "Four pairs of police officerssuffice, at Paris, forthe measurement,every

40. Gabriel Tarde, PenalPhilosophy, trans. Rapelje Howell, Boston, Little, Brown, 1912, p. 116. 41. Alphonse Bertillon,Identification anthropomttrique; instructions signalitiques, Paris, Melun, 1893,p. xiii. I have modifiedthe translation given in theAmerican edition, Signaletic Instructions, trans.R. W. Mclaughry,Chicago, Werner,1896. 26 OCTOBER

mob"wr?:.::::-:i:w WOW:~:::::::::::::::::::~: NOWii-r'~~-iiii~il'ig:-::-:::::::_::::::i_:::~ii:i~::i::~:WK ~j:: ??-,~:' :::I-i::-:::::::.:::::: :::::::---:::::::~~ :;one& ...... -:~:::::Mi:i:~:iiiiii?~ikiiiij :'':::l':-':iiil'i::lii~~i~i~::: sMa,:::i: iiiiili 12li; i:::~i:::?::::I:::::::::: ?::,i::: :--: i

FromAlphonse Bertillon, Service d'identification. Classificationcabinets, Paris Prefecture ofPolice. Expositionuniverselle de Chicago, 1893.

morningbetween nine o'clock and noon, of from100 to 150 men who were ar- restedthe day before."42Ultimately, this was not fastenough, and thereinlay a principal reason forthe demise, some thirtyyears later, of the Bertillonsystem. How did the Bertillonsystem work? The problems with prior attemptsat criminalidentification were many. The early promiseof photographyhad faded in the face of a massive and chaotic archiveof images. The problem of classifica- tionwas paramount:

The collection of criminal portraitshas already attained a size so considerable that it has become physically impossible to discover among them the likeness of an individual who has assumed a false name. It goes fornothing that in the past ten years the Paris police have collected more than 100,000 photographs. Does the reader believe it practicable to compare successivelyeach of these witheach one of the 100 individuals who are arresteddaily in Paris? When this was attemptedin the case of a criminalparticularly easy to identify, the search demanded more than a week of application, not to speak of the errorsand oversightswhich a task so fatiguingto the eye could

42. AlphonseBertillon, "The BertillonSystem of Identification,"Forum, vol. 11, no. 3 (May 1891), p. 335. The Bodyand theArchive 27

i~:i:i;-i~iiii:i

'''',iiiiiii:~ici-iii~""sS:iiF3ii;i'i ''i.iiiiB:ii'iiiii?iiiiiiiiiiiaiii.:iii:iiiiiiii .-:---? :':::::-:':':::?::':"''':''iiii~iiii-ii_::::-:-:-::.:i:;::::::::::: -riii-?::::_i?i:i:?:I-:.iiiii:iiiiii iiiii:"ii:'ii2'.'ii'::.::'::':"''::-::::iiibia"iD:i-8i:D'~:i~;~iiiiii:i;i-i--::::::i?--::--:i:?:-i:i-:--_-:-i-i:i.i:i:i:i:::::::?:':':i.i:iiiiiiiiiiiii-iii_:-i:iii.i-ii _.i:i:? :.i i-i:i-i-i:i:i: ::'::::::::::::: ::::::-:?;:-'''''::'';l:i-i:iiiiii'iiiii:iiiji';:ij :::'::::::::: ::: :::: :: '~.sli:ii~':~~i:':i:ili:---:i-i-i-ii-i:-i-i--:iiiii.i-ii :~~.i:i:i-.cc-~.l-i;il~~~ilii~:;.i-:I;,I:i:iiiiiii~:~:-i'iii'::ii$i'ii'i'i'iiiiiiiiliji'i-'i'ii:iiii-i:i:iii:iij iai-~-i-_~:,-i:~,_:I,_:-i:ilsii-iiri`l":''~'i ilii:i_-::-:::::-:':':_::?:?:--:::':?:?:?:::::( --:1::::: -i-i:i'i:i:il:i:::ii-i-i:i--:- :i-zi?~ii~iis':i:i:iai-~:jl:i?:i-iiiii?:::':::::::::::::::::::::::::: :i:'~iiiiiliiiiiiiiiiii-i:iiiiiiiiii-iii::i-i:ii-i:i ----:--_::-i:i--i:i-i-:--?:?ii ~::-:::iiii~i:iii~i~ii~j~~iij-iiiHi~_i~:... -jii:_:i_:~~:~'j:li_~-~iii~$::-~:~i.~~i p-.~-:~:_~:::::-::::i---: _-:::::i:i:i:::::i::i:::::_::-::::... -:.: ::.:::- : :: : ii~~,,i::i:i-iii:_iiiii:iiiiii:i.:-:(--iiiiiiiiiii:iiiiiiiii--iiiii:-::.:-::-ii:..::::(i:i-:i--::iii: ::':... : ::-:::--~1::"R:[j~~ iii: :i;-:iii'i;iiil-i-i:i:iiiri :: ..:::._:i:::-:_:i:i .:::,i''''''''''''' ;i~:~i?i-il-iiiii~:i,::-----:-::: ISI::::i::::i::::-::-:::::::::"::::---iil,-;:;:;,ili'' ::::::::'::"::I~:-~---sciiii:~iai;.i:~-1:~':':' ~ :'::~ii-:'a;:l-i~~:I-.iiiiiiiiiiiiiiiii ?--i-ii:Wi-ij:i:i-_i:i:i?-:ii-i-i-i a i-i? -i:-i--i i _'i::,;iiai~iii::i:i~i~iisii:i--i-i-:iii:iiiiiiiii "?:?1::i:iijiiiiii-iii:::i':iiiiii:~Yiiii~i-?~s~ii~iiiii~jijjj_::::::?:c:ii-:-:----:::'?::i:ii:i:i:i-_,l_':---:::::_:::::::~:iiiii:--i-:;-:_-:-::_--_: :---:--- :-:i?:_i.-::?:---:;: --:-:~::-___:_:_-;l-::l:iiiii:- :i-i:ii-i -?~i~i-ii:::i:::::::.::.: i:?:?-::-:--i:-i?i:i:i-i:i:i-..i-:-i?i _i-::i_, _i:.i::::::::i: :..:::....-_::::::::-. i-i:::,:::1,- :;?i:?:_:- :i:i_:_: i-i:::::::::::~:::: ~_e-iiiiii--i:ii:iiIlci:--i?::?:? ::: iii-i iiii-i-i:ii:i:i:iii:i::ili:iiiii:iii:ii i-:i;:~l???:?~.:?:i?ii-;,?,i:~.~~~: -:i-i :::sa?il~i;~i:s-i-;ii::i-i-i-i-i-i-:::::::::-:ii::-:--::i:::::--~:-:--::--a :: ii-iii:ibjiiii:iiiZiiil"l;:----?-__.~- i~_s-:?--.-:-m_-:--:::---:-_::__--I:::iii~iiiiii:_-ii ii:.??:::i: --i-i--:i :::::::::::-?i ---.---.:. --:--.... :-----i? _:?-::I: iiii: iiiii-ii-i : ::i::- ' iii-----:----. ?-:----::'.:--.::--:-,-_:,ii?__- Biii-i-i:i:i:iii-::i:ii '---:-:::::-:-i-i-ii:-:-. .: ...... --_:--i-?--:-:---:--.:-_-:- :::::i:-::::::-:_:;-:::-::?::ii:i~--~ i-i-i-i:i-:i-i:_::::i--i-i-i-i.; ii-iili:iiii:i_ : _::iii::i-i-::_i:i-_iiii '::." ::::::::::::::.-.-.-::. ii:,.ia:i'i:i:ii:iiiiiiiiiiiii:iiiii:ii:-: :::: - iiii:li-iiiiii-:-i-i-;iii-i:iiiii: ii...., iiiiij:i-iiii~iiiiia?i--i-ii I~~-~.5i-:;-isii:i-i:~i-i-i -:i:l-.:-:i_:~. -i:il.i -~-ii:,i iiii :,:::-;_....::..:..... -:i:::::? ~:'-iZiii------::----iiii:ili---i:i::-l -:::-:-: :?::: ,.,,,,~;._lli-.....-: ' ?i-:iii;i:i::::~::iiiii:iii-i ...... ::. i i-_---_ .:::: :-?-"7-:i;iii:~--:~:::i-i--.:::::::t-'li-i-it-i~-:--(:s---:-::;iiiiii-iiiiiiii-ii-iiiiiiiiii-iijiiii'iiiii-iiii; -?::-?:----~-:--:-:?'?::ii---i~i:i-i;:::i:-ii-:-:::::_:_::_: --:iii-''-i?!--::~:::::l__li:-::--:- ::-? :-I- --:'.-?'---:i : :::I:::-:i-i ::::;::::::::::::- : : -- - : i~i-ii- -':':::i--i '' ~ -::-' -'''"'" :: '-'----'- :'-::::-:::::::::-:'--'-'"'' ''''"'' ?-'::::'-:':"?':::' ~-::-::------:::;-...i-iii?---,iil--, ..--. ;:i-_;i-i--~---i"i:i-::i-i i-i:i-i.i.i iiiiiii:iiiiii:iii--:-iiiii i?iiiii:iii?::iii:il,:iiii:i:i-iii--::-iiii???iiiii:i:i:'- :iiili-i-.-i' -::-i::: -:-:---:- '::': :: :..' -:'--'' :'-::::--i:-ii:-l-iiii:-i:ii _ii ::-:-_-ii: iiii~- i:i. .-ii:i : . -:::?i-Li4i;-:-;- . -~i-i-:ili -I--:i--i-:-i:-:------: - -:---i-i_-:-il-_ii-;il:-i:i-~i:i~i-i-,i-~ii i-i i--i-ii-:: ...i-i-:-i-ii-ii --iii:-;ii.ii-' .:..:-: ii~i ... i:''i:i:iiiiii:iil~.iii:i:iiii '-.- -.i- -i:i- .:.._:~ii--iiiiiii:::_ __,_;_~ -:r-:-:-::--:-:::::::--::;:-:l-:i-i-i-i- hii'iBiii~:i:ili:ilI`iiiiiii-i'-::: :I~iii---iiii iii??i:ili -iii-:-::::::::: ::::: 'i'ii?i:i;i:i?ii-i iii~ii-i:iiilii.iiii:i:i :::::::?:::::::-:::::"iii::::ii:i:i:-ii-:::::::::j:::::::::-: -i- :--:-:-:-:::: :--.: --- ~:i-~?:~ li:i-~.-,: 1!- -... -::-:-::::::--::::------.-'1::''-1:::---:--::::::--::-:-:-::- '-'--"-:':' '--_-i -:"-----::':':-''i-: ii~ ~-i:i;;iii-iii-iiiiii~i;:ii::i -?il-:l~:-i:iiii::-iiiiiI : ... :iii,-:ii--... ii-: --~I- -i i-iiii;-i::::ii ::- :--:-ii -:i i:i-i:: ili-i?i-i-:-i-i.i-i-,i :.. : : i:;-~~----:.:;:::iiili ...-i:ii--i-i i :i;ieii-i~'i-:i::iii'iiiiii ..:._:? : i-i--:-:_:::::-::_-:: : :- ..:...-.:::: :::--:-::: --_--'-ia'-ll- .:: :-:-:::: ---:--:;::: .: -:-:::ii:-i-: -l"iiI1rl.i?-:-1-i:::~::::::;::l:-:-j:::: ,iii~iiiiii-iiiii_iiii:i-iiiiiiiiiiiiiiiiiiiiii _?iiii;ii-ii~iiii:.. i~:iiii_.;-_i ::::-:::-:::::::---':: : i-i-iii-i-::iii::iiii:: iiiii~:liiii:iiiiiii':i:ii'i~iiii:i _--::i--ii_ ii_-:ii::i--i::i--i:-iii:-::::::: ------..- ii-l::;l:aliiiaii:I:::l:l_~?;:,llji::_::::i: iijij. ... :-:-::-:_::-:::::_::::::::: iiiiiiiiiiiiiiiiiiiiii i:... ::.-:-::::::::::::;iii-i:i:i- :.:---:-:--iii:i:li-iiiiiiiiiii i i:i-.- ii'il I ri:-..-::.-j:i_ iiiiiiiii;:ii:::iiii--:ii-;i~-iiii:i:i.:...: :ii:ii:::?ri?iiii-i;~~':::-:::-:--:-:I:~::::":::::::::::`i?i:iii. :::' : ji-i-iii:i-i- i:-:i-i:i::i? _:::----:----:-::::::-- -...... -:--.: :::i::--... ::::: :-: ::: i-iiii,:-iiii ii:ii--i:iii:iiii-:i-ii:_:i-- iiii-iiiii:iii-:iii-i., iii:-.". i:i :::::~::_~:::-i::::i:_:_:::-:-:::-_-i ::i ::.::.:-:-i-i? :-::::: :_::i:ii--i: -:-:. ;.~ii~iiuiii"iiiiii.aiiii~:-::-?~ :_::::-:-:::-::::::. _-.-.----: iiiii:iiii:--:-ii: :-iii:i:iiii::i-ii:_iii:-:i:i:i:::: -::-::-:---::--:::::-::--~z: 'iiiiii:-iii:iii~iiiiiiiii~iiiiii-ii.:'- ::--:::----:--::::::--:-_iiiiiiji:-iii'i:ii::i-ii:::: -:-_:iiii:i li-~i:iiii':iiiii:-81:'-:1:-:iiiiiiiiiii ii --iiii :-i--:iiiii-::i-:'iii:i:iiii:ii-ii:i :: ::i:i-:- iii:.iiiiiiii; ii a-i--i-a-il-i:_:,::_:_:,:_:,:_j::::::::-:;_:-:---:-:: .::.:::.i-iiijiiii ::--:i-iii:--iii- i ~i-iiiiiiiii-jiii:iii . : -i---_: ii.:i::ii-li-i: -:-:-- 'iiiiii::'-iiiiiii-i':i:-:-ii~iiiiii:liii-iiiiiiii:-i?;i.... iii i-:-:l ~-iiiii:l:- .. ~:r::-ijl::-::~:~::-, i_,ii-i:ii--i:i:iii~:i-i:-i!:-;iii-i-i~ iji:ijiiii:i_~l:~ i?-i-i:i::-i:i-i::-?-:i:--i-i-i:i:::::i:. ::i. .. i:-i::-ii::i:ji---i.-:i:::::-:::-:-: :i--:--: -:::~:i:i:-:-::: :i: : : -:-:---i::: ::::::::::::: ::::---:::-_:::j?:-:-:-::-l----_i:-:- :_:::::::i?::.iiiji-:-:_:-;iii~i: i:iiiiiiiiiiiiiiiiiiii-ii-iiii---iiiiii--i-i:::ijii i-iiiiiiiiiii:ii-iii~.i.iiiiiiiiiiiiii i::..: iiit ..: . .: I :i:i-:: -:i::ii_-i i:--::::: : ::: : --i-:-:- iiiiiiiiiiiiiiiii..' ijii-iiiiii iii:. ii-:__i:-:i;( :::::--:-_::i- .:.:: ,iii~:_ i-i-ii,-::i?- _-_::--:-::-::::-_::::iiii_:iii::::---i_-iil-_---::-::_:_:2::-::::.... :-:: . _:-:::.-.. :::: ...:.-..-...::: :~~~f!..r;~:i~~'--'':iiiiii ::- -._ :_:- :::-_:::::: _:::: :::: :-----'--H::---:::_-::-::::-::-:-:_:_:__::___:___:___:_:_:___::____:::::::: ::::-::: :.: . :: -..._:--:-?:_-:: _:_:. ::::`:: : : i;:i ii:iiiii i:;-i--i- -i_-ii iiiiii;ii--iiiiiiiiii :-ii:i:~i--:_--_--::..--... -:- : ~~~l-?~8~r~s~la~8T--~1~ I ~:,::::::-_:- -::- :---.: : ::-- ii~ -R::::-:_-':i:~~-:j~-8::::::-i: WPi; ~ii-ii:i:iii-i:i:i:--i iiii-i-' ,,,_iii.ii:::i:i;;?i?:!:ili;iaii??ii-iiiis--?--:ia-,--::--i--iiHiaiiaiiP''~:::,::::::::::::-:::ii--::-i::I:ii::-:ii:::::?:::?::::::?-:;;::;ll:.~:~iii:?.l-.-;::i; :i ::: ...... ii-i:i-i-j:i :::::i-:i'i:i-i-i-ii-i:::::::::::: :::::iI:i:::: ~:i:.:::::::_ i-ii::iiiiiilii:i;-li-iii:i:i .:...:...::::?..~ii; -::iiii:-i:--iii-ii-iii:i :::::. :: --. -i"ii:i:i-~:~-i':.::..:-"i;---:r-::?:-:---:---~--:~-::~::::-:-:?- ::iiiiii.i-i-i-i.i-iiiiiii-i-ii-i:i-:r?iil-~_;i:i:iiii~iQii'::''-ii--i:i~'i"i-i-i:i.-__::::-.- ::::::ii:iiiiii iiiiiiiiiiiiiiii ..:.:::.. :.._ i:i;i;iiii:~lil~'i" i-i:i-i:i:iii-i-ii:i'i:i-i:i:ii-i-ii-i:i-ii:'-i-i-i:r...'"' .....--iiii:--iii:ji-iiiii:~i:i:iiii ~i:~l~ui-i-i'i:i:i-.i-i--i-i~i:i-i-i::i?_--::-_..ii:i;ii-i~iiii-i'i-i?i-iiii:iiiiiiiiiii:-i-i-i:::i~:i?ii:---:-:-:::: ;:i:i~i:::ii:?'::-:--::--::.-:------"'- '-::'~--':"''''''i-i?--i::_~-i-i--i:iii:i-?:--- iiii~i-i-i :: ::::- :::?-'-:------:---?:-~--~::-:---::'----:--:-:?-'------~:::::: :iiijiiiiiiiii;iijiiiiiiii :'::-::i?-i::--:? :ii-i:ii-iliiSi'i:iiiii-i.---ii-i-i:i- :-:: .:.-.-.-_ :-::-:-':-:-.--::i-i:-?:-:ii.;.. .- i~iiii-iiiiiii-iii-:_:i:_.i:i-i:,_i:-i-.-i . .: ---::_-_--:?:-:c:i-.i-i-;i--i:i:ir::. .. __:-: : :_----: :: i:i-i--i-i-ii:_:-_ :..: _::. :._-:::_:i-:, ~jiiiiiiii.. _iiiiiiiiiii'i' ~iiiii-'iiii:i'iii~8iiii:i''-iii-iii-ii _iij -::?::::;:::::::::: :::?::?:::;:::?:-:----::---:-:---::iiii::iiiii i--i-i;ii:i:i:i:i-i;-i:i-i:i--:-i--:_:i- :,ll-::~-'-'~i:ii~~I:::ii ~:??:aii?ii?~:i';'i2:-,ii~:::-?iiil:i:-'i'-~"~'~~'::ii":'ir:i:;iiiiiiiiii-:----::::::::-:: :i-i-i::::::::::::::::--_:_- i-i i:i --.-_-_-::-:-- .,. ..: -i-: iiiii .. : iiiiiii-i.iiii-i:ii:-::..:-. -i:i~iii.i:; i_------:::---::-ijiiii i::--:---ii-:- :- ::-:::::::-:: -:: -:- :--:?-::-:; :-:-: - :::::::::--::::: : :;i~i:u:i:_:;::~:i:i:is-*:i:l_::::::::-i:?:-:::-:--_::::::_,::::?1-:-:-:-:::-,--::_:::::::,,::::::_:?--_....:iiii~:i-.iiiiiii -i?i~i:i-i:i--:'i-ii-::-'-:::-::...-: :i-:i:i:: ::-::'--':i-iiiii'i-i-i-i::i:i:'- : :- -?:::'---::- ::--:-::-:---::: :::":::--::::- _:_-::: -:i-i-- _-: ::::::::- ::-::jiii:iii:_iii- -ili_:l:ii- :::::::j_::::::::_:-::::i i:. : -:-:-::_:::_--:::::-:::_:_: -:i:i-:i-i:.i-i::::- :_i:i:__i:-:: ii:_:_i:i,:-. -_i-ii:-iiiii~i:-liiilii:ii:is-il :--:-::-:::--:--:-:--:i ::-'-""-:-::-:-:--i-i:i'i-ii-i-i--::i-i- ~':i~-':ii-:liFi: :-: iii:::i:::: :- i i:ii:ij-:i:i__~iji -ii~.~l- ::::- -: :-i;i- --:-iiiiii- -iiisiiiiii~-;l.:lii:i::ii'i~~ii:-:::::::iiiii::iii~ii`Fi-i'ci:i ::i:ii:iiiii-ii-i-:iii-i:iiliiiii:i::-:.ii-i-i':::::::'iiiiiiiiiiii~iiijii ::::::::::I:_::-:---::::--:--:.i;:iiiiiiiiijijiliiiiiiiiiiiii-_:::: ----:i:- :::::::::: :::: ii:ii-:.:- ::-i-ii::-iii-: -i--::i::-::iii::i-i- -_i:i-_:- -i_-iiii:i-:iii- . -.::::ii::-:i:::iiii -iii-:::::::::: Displayof apparatus, Chicago Exposition.

not fail to occasion. There was a need for a method of elimination analogous to that in use in botany and zoology; that is to say, one based on the characteristicelements of individuality.43 Despite the last part of this remark,Bertillon sought not to relate individual to species, but to extractthe individual fromthe species. Thus he inventeda clas- sifyingscheme that was based less upon a taxonomic categorization of types than upon an orderingof individual cases withina segmented aggregate. He had failedmiserably in an earlier attemptto classifypolice photographsaccord- ing to the genre of offense,for obvious reasons.44 Criminals may have constituteda "professionaltype," as Tarde argued, but they did not necessarily observe a narrow specialization in their work. Bertillonsought to break the professionalcriminal's mastery of disguises, false identities,multiple biographies, and alibis. He did thisby yokinganthro- pometrics,the optical precision of the camera, a refinedphysiognomic vocabu- lary, and statistics. First Bertillon calculated, without a very sophisticated grasp of the calculus of probabilities,that the chance that two individuals mightshare the same series of eleven bodily measurements ran on the order of one in four 43. Ibid., p. 331. 44. AlphonseBertillon, L'identiti des ricidivistes etla loide religation, Paris, Masson, 1883,p. 11. RELEVE DU SIGNALEMENT ANTHROPOMETRIQUE

IP I I I

K I i I ,.

I. *

_ I ~/ ~ 1 'K. 7 8;r? 9

I. Taille. - 2. Envergure.- 3. Buste. -- 4. Longueurde la tete. - 5. Largeurde la ttec. - 6. Oceille droite.- - 7. Pied gauche. - 8. MNdius gauche. 9. CoudJc gauche. Frontispiece(left)and figures (right) from Alphonse Bertillon,Identification anthropom6trique, 1893. million.45 He regarded these eleven measurements as constant in any adult body. His signaleticnotice linked this"anthropometrical signalment," recorded as a numerical series, with a shorthand verbal description of distinguishing marks, and a pair of photographicportraits, both frontaland profileviews. Bertillon'ssecond problem was the organization of individual cards in a comprehensivesystem from which recordscould be retrievedin shortorder. To this end, Bertillon enlisted the prodigious rationalizing energies of Quetelet's "average man." By organizing his measurements into successive subdivisions, each based on a tripartiteseparation of below-average, average, and above- average figures,Bertillon was able to file 100,000 records into a grid of file drawers, with the smallest subset withinany one drawer consistingof approx- imatelya dozen identificationcards. Having thus separatelyprocessed 100,000

45. Bertillon,Identification anthropomitrique, pp. xvii-xviii. SIGNALEMENT DESCRIPTIF XLI

V141525 M196:;65 616 7017t 777 881 trespefiirsc~proprcmT?Vfletljldite ImullnrT.1) PLUC)Iz~w,~r Txllenwjww rmlrfimnTiall, Tillie qrrmd Taille tre51-di

Fig. 4. - COURSEDINOUIALE DE LA sur laquelle on a s9par6 par des verticales l'emplacomentdesTAIL.. sept cat6goriesdo taille.

236

198 198

0.. 7

I( "1 71 53 "59 7-63 1"67 171 7"77 1"89 !

1"-0 1' 52 1"62 1'"66 1"70 1`76 1"88 Co, 1'58.

Fig. 5. - DIAGRAMMEDE LA TAILLEindiquant par la hauteurproportionnelle descolonnes le nombredes sujets ressortil,it i chacune des sept catigoriesde taille indipques surla courbe binomiale (Fig. ).

male and 20,000 female prisoners over the decade between 1883 and 1893, Bertillon feltconfident in boasting that his systemwas "infallible."He had in the process "infallibly"identified 4,564 recidivists.46 Bertillon can be said to have realized the binomial curve as officefur- niture. He is one of the firstusers of photographicdocuments to comprehend fullythe fundamentalproblem of the archive, the problem of volume. Given his recourse to statisticalmethod, what semantic value did he find in photographs? He clearly saw the photograph as the final conclusive sign in the process of identification.Ultimately, it was the photographedface pulled fromthe filethat had to match the rephotographedface of the suspect, even if this final "photographic"proof was dependent upon a series of more abstract steps. Bertillonwas criticalof the inconsistentphotography practiced by earlier

46. Ibid., pp. xxi-xxiii, lxxiv. 30 OCTOBER

police techniciansand jobbers. He argued at lengthfor an aestheticallyneutral standard of representation: In commercial and artisticportraits, questions of fashion and taste are all important.Judicial photography,liberated fromthese con- siderations, allows us to look at the problem from a more simple point of view: which pose is theoreticallythe best forsuch and such a case?47

Bertilloninsisted on a standard focal length,even and consistentlighting, and a fixeddistance between the camera and the unwillingsitter. The profileview servedto cancel the contingencyof expression;the contourof the head remained consistentwith time. The frontalview provided a face that was more likelyto be recognizable withinthe other,less systematizeddepartments of police work. These latterphotographs served betterin the search for suspects who had not yetbeen arrested,whose faces were to be recognized by detectiveson the street. Just as Bertillonsought to classifythe photographby means of the Vitru- vian registerof the anthropometricalsignalment and the binomial curve, so also he sought to translate the signs offeredby the photograph itself into another, verbal register.Thus he was engaged in a two-sided, internaland ex- ternal, taming of the contingencyof the photograph. His inventionof the portrait-parld-the "speaking likeness"or verbal portrait- was an attemptto overcome the inadequacies of a purely visual empiricism. He organized voluminous taxonomic grids of the features of the male human head, using sectional photographs. He devoted particular attentionto the morphologyof the ear, repeating a physiognomicfascination with that organ that extended back to Lavater.48 But on the basis of this comparative anatomy, Bertillon sought to reinvent physiognomy in precise nonmetaphysical, ethnographic terms. Through the constructionof a strictlydenotative signaleticvocabulary, this project aimed forthe precise and unambiguous translationof appearance into words. For Bertillon,the criminal body expressed nothing. No characterological secrets were hidden beneath the surface of this body. Rather, the surface and the skeletonwere indices of a more strictlymaterial sort. The anthropometrical signalment was the registerof the morphological constancy of the adult

47. Alphonse Bertillon, La photographiejudiciare,Paris, Gauthier-Villars, 1890, p. 2 (my translation). 48. In 1872, 0. G. Rejlander suggested that photographs of ears be used to identifycriminals ("Hints Concerning the Photographing of Criminals," BritishJournal Photographic Almanac, 1872, pp. 116-117). Carlo Ginzburg has noted the coincidence of Bertillon's attention to the "individuality" of the ear and Giovanni Morelli's attempt to construct a model of art-historical authenticationbased on the careful examination of the renderingof the ear by differentpainters ("Morelli, Freud, and Sherlock Holmes: Clues and ScientificMethod," HistoryWorkshop, no. 9 [Spring 1980], pp. 5-29). iriiiiiiiiiii ::::-::-::'::-::il:il

!ii--i -iiii-'iiiPiil-i??l iiijilli:i-ii.liiij-jjiii~ii:L::i--:-ii-3iii

iaiiii-ii!iiiiiiiiiiiiiiii Siiiii:i:,iiiiiiii`?::: :.:::?:::j:,:i::: iiiii: i:: ii : !::" iiiiiiiaiii-iiili-c-ii-iiiiiiiiiiiiii:::::?.::iii'i '''''':':''':-:'::''''''::::::::':'':'::i?'iii::-iici-iii -l:iiiliii'iiiif;ii:i I'-i: ::;l -iiiiiiiiiiiiiii-iiiiiiiii--ili: !!!!!!!!!!!!!!iiii:iiii::iiSi l,'ii:i::" i:i:- .ir- ,,, -"1 "*' alir i :i~iiiiiii~-:ii'i~~-'---i~-~ii::i~~bi~iiiiiiii:?;l.;ii: liiiiii':iiiili: ,iili:/?:;i:ii~iiiiii:::::::.': ':-';::::::::i_:i:::::::??:i~i::iii-li jiil i~~liiii~ ....iiiii:'i'-iiiii~ii~iiiiiii :::il-:-ii i:: ii' i:iiiii :-.iiiiiii:?iil'j ...... Eii?9Lliiil "i--" ,iii~:iiii iiii-?Iiiiiii:i'si'~iii~~iiii~li-i~i~iiiiii-iiiiii1 iiiiiiiiii(i~i~i!i~iiii~iiii~i-..::-i;iiiiiI

,Tiiitiliiiiii~liiilii ~ I I ss~s8~sss~iii'l' 8l ~ ~ 3~~~~sa~ig~~6- .i..i.iiii. -;

cii-ii-:iiiiiiiiiiiiiiii :iiii~iiiiiiiiiiiiiiii.:iill ri~i~iiii-i!ii~li-

biiii~i~iiiiiiiiiiii~ liiiiiii~si:iii~iiiiiiiiC4 :ii~iiiii~i~iiiiiiii!......

iii-i:Iii~i~iii'iiiiiii

i.iiiiiii!i-iiiiiiiiiiii

-i-i-ii-i-i-i:iiii:iiil i

!i!i~iii~iii~iii~i:;i

i!iiiiiiiii~iiiiiiii

ii~iiigii~i;i-ii !ii

...... i

Plate41 fromAlphonse Bertillon, Identification anthropomrntrique,1893...... iiiiiiiiiiii~i::iiiil~i~iiiiii!iiii! i-iiii:--i-i-i;:_-ii iiii :::--:-:-:- Iji i-li:::ii:-;:Il? ii-iiiiii-iiii i-iii:i iiii-ii-iiiiiiiiii------:'---' iiiiiiiii-i~iii::. iiili:: i--i:ii-i-iiiiiiiiii:: :-i

-'"'''':-~:a-::::-i::'-ii.$.l~r~;?':i?::?c:-:i::--:j.:-::i:::NOW*-

-: :-iiiiiiiiiiii--i--ii-i~i-i:_iii:--::-i:_::----_------:i:: '--':'-:'-~1-':--:?;:::--:::--::-;--:-i-i:iii-i-i-:LAii i~ :::-?ii : i :i: - ii. i-i~!i~iiii -:ililiiii i::iii:iiiiii i...... :i~:::ili-::::::::::: iI-:i--ii- -? i~:i;:y:---:~:-:_--:::::: ::::::ANti~-iii~:iii?;?ii~i lai Iaii-:iil .;:??:,_:'?:::!ii?:liij-::. :'''-';'I'- iiiiiiiiiii ii-iiiii .:. :::.:...:. ... l ...... iiiiii:iii.iiii ii i ii!!!! iiiiiiiiii'

i !iiiiii~iii~iii!!!iiiii......

Plate56from Alphonse Bertillon, Identification anthropomitrique,1893. The Bodyand theArchive 33

skeleton,thus the key to biographical identity.Likewise, scars and otherdefor- mations of the fleshwere clues, not to any innate propensityfor crime, but to the body's physical history:its trades, occupations, calamities. For Bertillon, the mastery of the criminal body necessitated a massive campaign of inscription,a transformationof the body's signs into a text,a text that pared verbal descriptiondown to a denotative shorthand,which was then linked to a numerical series. Thus Bertillonarrested the criminal body, deter- mined its identityas a body thathad alreadybeen definedas criminal,by means that subordinated the image-which remained necessary but insufficient-to verbal text and numerical series. This was not merelya self-containedarchival project. We can understand another, more global, imperativeif we remember that one problem for the late-nineteenth-centurypolice was the telegraphic transmission of informationregarding suspects. The police were competing with opponents who availed themselvesof the devices of modernityas well, including the railroad. Why was the issue of recidivismso importantin France during the 1880s? Robert Nye has argued recentlythat the issue emerged on the political agenda of GambettistRepublicans during the Third Republic, leading to the passage of the Relegation Law of 1885, which established a Draconian policy of colo- nial transportfor repeat offenders.The bill worked out a variable quota of misdemeanors and felonies,including vagabondage, that could lead to perma- nent exile in Guyana or New Caledonia. The French agriculturalcrisis had led to a renewed massive urban influxof displaced peasants during the 1880s. The recidivism debate focused on the social danger posed by the vagrant, while also seeing the milieu of the chronicallyunemployed urban poor as a source of increased criminality.Not least in provokingthe fearsof the defendersof order was the evidence of renewed working-classmilitancy in the strikewave of 1881, aftera decade of peace purchased by the slaughterof the Communards. At its most extreme,the debate on recidivismcombined the vagabond, the anarchist, and recidivistinto a single composite figureof social menace.49 Bertillon himselfpromoted his system within the context of this debate. Having only succeeded in identifyinghis firstrecidivist in February of 1883, he quickly argued that his binomial classificationsystem would be essential to the application of any law of relegation. He described a Parisian working-class milieu that was undergoing what might facetiouslybe called a "crisis of identity."During the Commune, all city records prior to 1859 had been burned; any Parisian over twenty-twoyears old was at libertyto inventand reinventan

49. See Robert Nye, Crime,Madness, and Politicsin ModernFrance.: The Medical Conceptof National Decline, Princeton, Princeton University Press, 1984, pp. 49-96. Although Nye mentions Bertillon's project only in passing, I have relied upon his social historyfor an understanding of the politics of French criminologyduring the late nineteenthcentury. A more directlyrelevant study of Bertillon, Christian Pheline's L'image accusatrice(Paris, Cahiers de la Photographie, 1985), unfortunatelycame to my attention only after this essay was going to press. 34 OCTOBER

entirelybogus nativity.Furthermore, Bertillon claimed that there was an ex- traordinarytraffic in false documents, citing the testimonyof foremenat the more "insalubrious" industrial establishment- white lead and fertilizerfac- tories, forexample- thatjob applicants frequentlyreappeared two weeks after being rejected with entirely new papers and differentnames.50 In effect, Bertillonsought to reregistera social fieldthat had exploded into multiplicity. One curious aspect of Bertillon'sreputation lies in the way in which his method, which runs counter to any metaphysicalor essentialistdoctrine of the self, could be regarded as a triumphof humanism. One biographer put it this way: "A man of his type inevitablyfound a kind of romance in a technique the aim of which was to individualize human beings."51 Bertillon himself con- tributedto this "humane" reading of his project: "Is it not at bottom a problem of this sort that formsthe basis of the everlastingpopular melodrama about lost, exchanged, and recovered children?"52But in more technicaland theoretical contexts,the degree to which Bertillonage actually eroded the "uniqueness" of the selfbecame clear. Writingwith a coauthor in 1909, Bertillonnoted that according to the logic of the binomial curve, "each observationor each group of observations is to be defined, not by its absolute value, but by its deviation fromthe arithmeticmean."53 Thus even the nominalistBertillon was forcedto recognize the higherreality of the "average man." The individual could only be identifiedby invokingthe powers of this genie. And the individual only existed as an individualby being identified. Individuality as such had no meaning. Viewed "objectively,"the self occupied a position that was wholly relative. The Bertillonsystem proliferated widely, receivingan enthusiasticrecep- tion especially in the United States and contributingto the internationalization and standardizationof police methods. The anthropometricsystem faced com- petitionfrom the fingerprintsystem, a more radically synecdochicprocedure, invented in part by Francis Galton, who had interestsin identificationas well as typology.With the advent of fingerprinting,it became evidentthat the body did not have to be "circumscribed"in order to be identified.Rather, the key to identitycould be found in the meresttrace of the body's tactile presence in the world. Furthermore,fingerprinting was more promising in a Taylorist sense, since it could be properlyexecuted by less-skilledclerks. By the late nineteen- tens, the Bertillonsystem had begun to yield to thismore efficientand less cumbersome method, although hybrid systemsoperated for some years.54

50. Bertillon, L'identitides ricidivistes,pp. 2, 5. 51. Henry Rhodes, AlphonseBertillon: Father of ScientificDetection, London, Abelard-Schuman, 1956, p. 83. 52. Bertillon,"The BertillonSystem of Identification,"p. 330. 53. A. Bertillonand A. Chervin,Anthropologie mnitrique, Paris, ImprimerieNationale, 1909, p. 51 (my translation).The same textdrolly likens the shape of thebinomial curve to thatof a "gendarme'shat." 54. Bertillonnoted that his systemwas adopted by 1893 in the United States, Belgium, Bertilloncard, 1913.

POLICE DEPARTMENT, SAN DIEGO, CALIFORNIA

Il egwit I Im dFF Age"__

v lMA gt 4 ubit PLuAOt+ ii aiii

...... HeighD!toMNasilomN":'

WidthTe.neightgh oh Wih j----- .....---P-Beard Chilld

A 14 k rtied els

Switzerland,Russia, muchof SouthAmerica, Tunisia, theBritish West Indies,and Rumania (Identificationanthropomitrique, p. lxxxi). Translations of Bertillon'smanuals of signaleticinstruc- tionsappeared in Germany,Switzerland, England, and Peru, as well as theUnited States. On theenthusiastic American reception of the Bertillon system, see Donald Dilworth,ed., Identifica- tionWanted.: Development of the American Criminal Identification System, 1893-1943, Gaithersburg, Maryland,International Association of Chiefs of Police, 1977. The IACP promotedthe general adoptionof Bertillonageby the geographicallydispersed and municipallyautonomous police forcesof the United States and Canada, and the establishmentof a National Identification Bureau in Washington,D.C. This officewas absorbedinto the Federal Bureau of Investigation in 1924. (Canada adoptedBertillonage with the Criminal Identification Act of 1898.) Startingin 1898,a quasi-officialmonthly publication of the IACP, called TheDetective, carried Bertillon mea- surementsand photographsof wanted criminals. This publicationprovides a reasonablegauge of the ratioof relianceby Americanpolice on the Bertillonand fingerprintsystems over the next twenty-fiveyears. The Britishresisted Bertillon's method, largely because the fingerprint system S r a c i...a .k..ad..

:: ::? ,il-:-

From Yawmanand ErbeMfg. Co., Criminal Identificationby "Y and E": Bertillon and Finger Print Systems, 1913.

was of Britishorigin. Nonetheless, regulations were establishedin 1896 underthe Penal Ser- vitudeAct of 1891 forthe photographing, fingerprinting, and Bertillon measurement of criminal prisoners(Great Britain,Statutory Rules and Orders, London, H. M. StationaryOffice, 1896, no. 762, pp. 364-365). By 1901, however,the anthropometricsignalment was abandoned. Bertillonand Galtontraded jibes at theirrespective systems. Bertillon faulted Galton for thedifficulties encountered in classifyingfingerprints ("The BertillonSystem of Identification," p. 331). Galtonfaulted Bertillon for his failure to recognize that bodily measurements were corre- lated and not independentvariables, thus grossly underestimating the probabilityof duplicate measurements(Francis Galton, Memories ofMy Life, London, Methuen, 1908, p. 251; see also his "PersonalIdentification and Description,"Journal of the Anthropological Institute, vol. 18 [May 29, 1888], pp. 177-191). The two men'sobsession with authorship may have been a bit misplaced,however. In "Morelli,Freud, and SherlockHolmes" (cited in note48, above), Carlo Ginzburghas suggested thatthe whole enterprise of rationalized criminal identification rested on thetheft of a morepopu- lar,conjectural form of empiricism, grounded in huntingand divining.Sir William Herschel had appropriatedfingerprinting in 1860 from a usagecustomary among Bengali peasants under his co- lonialadministration. The sourceof police methods in whatGinzburg describes as "lowintuition" was obliquelyacknowledged by Bertillonin a passagein whichhe arguesfor a rigorouslyscientific policing,while invoking at thesame time the distinctly premodern image of the hunter: "Anthropol- ogy,by definition,is nothingbut thenatural history of man. Have nothunters in all timesbeen interestedin naturalhistory? And, on theother hand, have not naturalists something of the hunter in them?No doubtthe police of the future will apply to theirparticular form of the chase the rules ofanthropology and psychology,just as theengineers of our locomotivesare puttingin practice thelaws of mechanicsand thermodynamics"("The BertillonSystem of Identification,"p. 341). Ginzburghas proposeda modelof observation and descriptionthat is moreopen to multiplicity and resistancethan thatadvanced by John Tagg, who subsumesall documentarywithin the paradigmof thePanopticon (Tagg, "Powerand Photography,"p. 55). The Bodyand theArchive 37

For Bertillon,the type existed only as a means forrefining the description of individuality. Detectives could not affordnot to be nominalists. Bertillon was not alone in this understanding of the peculiarities of the policeman's search for the specificityof crime. For example, the New York City detective chiefThomas Byrnes published in 1886 a lavish "rogues' gallery"entitled Profes- sionalCriminals of America. Although Byrnes practiced a less systematicmode of photographythan did Bertillon,he clearlyarticulated the position thatclassical physiognomictyping was of no value whatsoever in the hunt for the "higher and more dangerous order" of criminals, who "carried no suggestion of their calling about them."55In Bertillon'scase, the resistanceto the theoryof a biologi- callygiven criminal type was also in keeping with the general driftof late-nineteenth-centuryFrench criminologicaltheory, which stressedthe importanceof environmental factors in determining criminal behavior. Thus the "French school," notablyGabriel Tarde and Alexandre Lacassagne, opposed thebiolog- ical determinismof the "Italian school"of criminalanthropology, which centered on the anatomist-craniometricianCesare Lombroso's quasi-Darwinian theory of the criminalas an "atavisticbeing who reproduces in his person the ferocious instinctsof primitivehumanity and the inferioranimals."56 Against this line of reasoning, Lacassagne argued that "the social milieu is the mother culture of criminality;the microbe is the criminal."57(In this context, it is worth noting the mutual admiration that passed between Pasteur, the microbe-hunter,and Bertillon,the hunterof recidivists.58)The French were able to medicalize crime while simultaneouslypointing to environmentalfactors. A range of positions emerged, some more medical, some more sociological in emphasis. Tarde in- sistedthat crime was a professionthat proliferated through channels of imitative behavior. Others argued that the criminal was a "degenerate type," suffering more than noncriminals fromthe bad environmentaleffects of urbanism.59 Despite the acute differencesbetween the warringfactions of the emerging criminologicalprofession, a common enthusiasm forphotographic illustration of the criminal type was shared by almost all of the practitioners,with the no- table exception of Tarde, who shunned the lowly empiricismof the case study formore lofty,even ifnominalist, meditations on the problem of crime. Before looking at Francis Galton's peculiar contributionto the search for a criminal type, I will note that during the 1890s in particular, a profusionof texts ap-

55. Thomas Byrnes, "Why Thieves are Photographed," in ProfessionalCriminals ofAmerica, New York, Cassell, 1886, p. 53. 56. Cesare Lombroso, "Introduction,"to Gina Lombroso-Ferrero, CriminalMan, New York, Putnam, 1911, p. xxv. 57. Quoted by Nye, p. 104. 58. Rhodes, p. 190. 59. See Nye, pp. 97-131. 17

R O .

.O L..H...

. PC E .. POCK.T

ji,

From ThomasByrnes, Professional Criminals of America, 1886. P YSIONOMIFof L LA ,. .. CRAI..INE

... .."......

...... ii ::......

Fo;.Iti). -Clayes. t.1It Cla,c

...... )

:gg ii-- ijiiiiiii

): : i ...... g ......

FromCharles Marie Debierre,Le crane des criminels, 1895. 40 OCTOBER

peared in France and Italy offeringphotographic evidence of basic criminal types. Although the authors were frequentlyat odds with one another over the "atavistic"or "degenerate"nature of the criminal, on a more fundamentallevel they shared a common battle. This was a war of representations.The photograph operated as the imageof scientifictruth, even in the face of Bertillon's demonstrationof the inadequacies of the medium. Photographs and technical illustrationswere deployed, not only against the body of the representative criminal, but also against that body as a bearer and producer of its own, in- feriorrepresentations. These textscan be seen as a battle between the camera and the tattoo, the erotic drawing, and the graffitiof a prison subculture. For Lombroso, tattooingwas a particular mark of atavism, since criminals shared the practice with presumablyless evolved tribalpeoples. But even workswhich sought to demolish Lombroso's dogmatic biologism established a similar hierarchy. Scientificrationalism lookeddown at the visual products of a primitive criminality.This was a quasi-ethnologic discourse. Consider, for example, a work which argued against atavism and for degeneracy, Charles Marie Debierre's typologicallytitled Le crdnedes criminels.This book contained an il- lustratedchapter treating"les beaux-arts dans les prisons"as subject matterfor the psychological study of the criminal. A subsequent chapter offereda set of photographs of the severed heads of convicts, "taken one quarter of an hour after decapitation." Faced with these specimens of degeneracy, this physiog- nomist of the guillotine remarked: "Degroote and Clayes . . . theirdull faces and wild eyes reveal that beneath theirskulls thereis no place forpity." Works of this sort depended upon an extreme form of statistical inference: basing physiognomicgeneralizations on very limited samples.60 This brings us finallyto Francis Galton, who attemptedto overcome the limitationsof this sort of inferentialreading of individual case studies. Where Bertillonwas a compulsive systematizer,Galton was a compulsive quantifier.While Bertillonwas concerned primarilywith the triumphof social order over social disorder,Galton was concerned primarilywith the triumphof established rank over the forcesof social leveling and decline. Certainly these were not incompatible projects. On a theoreticalplane, however, Galton can be linked more closely to the concerns of the Italian school of criminal anthropology and to biological determinismin general. Composite images based on Galton's procedure, firstproposed in 1877, proliferatedwidely over the following three decades. A composite of criminal skulls appears in the albums of the

60. CharlesMarie Debierre,Le crdnedes criminels, Lyon and Paris, Storckand Masson, 1895, p. 274. The otherimportant illustrated works are by membersof the Italian school: Lombroso's revisedFrench and Italianeditions of his 1876 L'uomodelinquente included separate albums of il- lustrations(Paris, Alcan, 1895and Turin,Fratelli Bocca, 1896-97). The platesof criminal types in thesealbums were taken from materials prepared for Enrico Ferri, Atlante antropologico-statistico dell'omicidio,Turin, FratelliBocca, 1895. PlateXXXIX from Cesare Lombroso, L'homme criminel, 1895. C

lo41,I PACE J5. 2"j W'5

(I d If 4tc~?~ *VA61,,I >zzJ .. \fJyj~~&/&#

)\) *. ? ,,,

LIUK5UBIK

TATOUAGES DE CRIMINELS. THE CRIMINAL.

HAVELOCK ELLIS.

ILLWSTRATED.

LONDO N: WALTER SCOTT, 24 WARWICK LANE, PATERNOSTER ROW 890.

Galtoniancomposite.

1895 French edition and the 1896-97 Italian edition of Lombroso's Criminal Man. Likewise, Havelock Ellis's The Criminal,which adhered to the positionsof the Italian school and marked the high tide of Lombrosoism in England, bore a Galtonian frontispiecein its first,1890 edition.61 Both Galton and his quasi-officialbiographer, the statisticianKarl Pear- son, regarded the composite photograph as one of the central intellectualin- ventions of Galton's career. More recent studies of Galton have tended to neglect the importanceattached to what now seems like an optical curiosity.62 Galton is significantin the historyof science fordeveloping the firststatis- tical methods for studyingheredity.63 His career was suspended between the triumphof his cousin Charles Darwin's evolutionaryparadigm in the late 1860s and the belated discoveryin 1899 of Gregor Mendel's work on the geneticratio underlyinginheritance. Politically, Galton sought to constructa program of social bettermentthrough breeding. This program pivoted on a profoundly ideological biologizationof existingclass relations in England. Eugenicistsjusti- fiedtheir program in utilitarianterms: by seekingto reduce the numbers of the

61. HavelockEllis, TheCriminal, London, WalterScott, 1890. 62. The exceptionis David Green,"Veins of Resemblance:Photography and Eugenics,"The OxfordArt Journal, vol. 7, no. 2 (1984), pp. 3-16. 63. See RuthSchwartz Cowan, SirFrancis Galton and the Study of Heredity inthe Nineteenth Century, New York, Garland, 1985. ~~ ~ ~ ~ ~ ~ ~ ~ ~ .. :(:.A?:m: ...... :::... alXX!71

Fiotograli~eomuposite (4--dltoli~ane: di: eu ani di d~e~inquenti.

.r.r ..d.\.e. loirae

di Anoiuiiia~i:inl Vre er~mi deli~nquou:ti.

PlateXXVII fromCesare Lombroso, L'uomo delinquente, 1896-97. 44 OCTOBER

"unfit"they claimed to be reducing the numbers of those predestinedto unhap- piness. But the eugenics movement Galton founded flourishedin a historical context- similar in this respect to Third Republic France - of declining middle-classbirthrates coupled withmiddle-class fears of a burgeoningresiduum of degenerate urban poor.64 Galton's early, 1869 work HereditaryGenius was an attemptto demonstrate the priority,in his words, of "nature"over "nurture"in determiningthe quality of human intelligence.In a rathertautological fashion, Galton set out to demonstrate that a reputation for intelligenceamounted to intelligence,and that men with(reputations for) intelligence begat offspringwith (reputations for) intelligence. He appropriated Quetelet's binomial distribution,observing that the entrance examination scores of militarycadets at Sandhurst fellinto a bell- shaped pattern around a central mean. On the basis of this "naturalizing" evidence, he proposed a general quantitativehierarchy of intelligence,and ap- plied it to racial groups. This hierarchywas characterized by a distinctclassi- cist longing: "The average abilityof the Athenian race is, on the lowest possible estimate, very nearly two grades higherthan our own - that is, about as much as our race is above that of the Africannegro."65 Eugenics can be seen as an attempt to push the English social average toward an imaginary, lost Athens, and away froman equally imaginary, threateningAfrica. Galton's passion forquantification and numerical ranking coexisted with a qualified faithin physiognomicdescription. His writingsdemonstrate a remarkable parallelism and tension between the desire to measure and the desire to look. His composites emerged fromthe attemptto merge optical and statistical procedures within a single "organic" operation. Galton's Inquiriesinto Human Facultyof 1883 began by suggestingsome of the limitationsof prior- and subsequent - attemptsat physiognomictyping: The physiognomicaldifference between differentmen being so numerous and small, it is impossible to measure and compare them each to each, and to discover by ordinarystatistical methods the true physiognomyof a race. The usual way is to select individuals who are judged tQ be representativeof the prevalent type, and to photograph them; but this method is not trustworthy,because the judg- ment itselfis fallacious. It is swayed by exceptional and grotesque featuresmore than by ordinaryones, and the portraitssupposed to be typical are likely to be caricatures.66

64. See GarethStedman Jones, Outcast London: A Studyin theRelationship between Classes in Vic- torianSociety, Oxford, Clarendon, 1971. 65. FrancisGalton, Hereditary Genius, London, Friedman,1978, p. 342. 66. FrancisGalton, Inquiries into Human Faculty and Its Development, London, Macmillan, 1883, pp. 5-6.

Frontispiecefrom Francis Galton, Inquiries into Human Faculty, 1883. SPEIMESPERSNALOFCOMPOSTEAND FAMI"rOTATR

\/bf/i/Ii /~ I tvbIrb / Jr)/L4m litufflexde

HEALT// f DS EAXSfE.,i?CRIMINALITY. 10' .1/110 "(((("NXfY t/ t f&/na CONSUMPTIONANDOTHEMAL ADE 11 20

(1wsup/rthe w. 1 // DIAGRAM SHOWING THE ESSENTIAL PARTS. A

! 0:1i w 'ruse rI:z OR CAS

Side View.

End View. is A The body of the camera,which is fixed. e The eye-holethrough which the image B Lens on a carriage,which can be moved viewed on g; a thinpiece of glass iim- to and fro. mediatelybelow e, reflectsthe illumin- C Frame for the transparency,on a car- ated fiducial lines in the transparency of riage that also supports the lantern; at!, and gives themthe appearance the whole can be moved to and fro. lying upon g,-the distancesfe and r The reflectorinside the camera. g e being made equal, the angle fe g mThe arm outside the camera attached to beingmade a rightangle, and theplane made the axis of the reflector;by moving of the thin piece of glass being it, the reflectorcan be moved up or to bisectfe g. down. f Framework,adjustable, holding the trans- on it. A screenon the roof,which parencywith the liducial lines g ground-glass trans- receives the image when the reflector t Framework,adjustable, holding the is turneddown, as in the diagram. parencyof the portrait.

This book was a summary of Galton's researches over the preceding fifteen years. From this initial criticismof a more naive physiognomicstance, Galton moved directlyto an outline of his composite method. The composite frontispiece and the recurrentreferences in various contextsthroughout the book to lessons to be learned fromthe composites suggest that Galton believed that he had invented a prodigious epistemological tool. Accordingly, his interestin composite imagery should not be regarded as a transparentideological stunt, but as an overdeterminedinstance of biopositivism. How did Galton produce his blurred, fictitiousapparitions? How did he understand them? He acknowledged at the outset of his experimentsHerbert Spencer's prior proposal fora similar process of superimposition.Spencer's or- Thecomposite apparatus, from Francis Galton, Inquiries intoHuman Faculty,1883.

ganismic conception of societycan be seen as fertilesoil forthe notion of a generalized body, although in this case Spencer seems to have been drawn to the notion of a composite through a youthfulfascination with phrenology.67But Galton was concerned also with the psychologyof the visual imagination, with the capacity of the mind to constructgeneric images fromsense data. Here he foundhis inspirationin Thomas Huxley. He claimed in factthat the composite photographicapparatus shared, and ultimatelysurpassed, the capacity of ar- tistic intelligence to generalize. Here, as with Quetelet, one witnesses the statisticianas artist manque. Galton fabricated his composites by a process of successive registration and exposure of portraitsin frontof a copy camera holding a single plate. Each successive image was given a fractionalexposure based on the inverse of the total number of images in the sample. That is, if a composite were to be made froma dozen originals, each would receive one-twelfthof the required total exposure. Thus, individual distinctivefeatures, features that were unshared and idiosyncratic,faded away into the nightof underexposure. What remained was the blurred, nervous configurationof those featuresthat were held in common throughout the sample. Galton claimed that these images constituted legitimate averages, and he claimed further that one could infer larger generalitiesfrom the small sample that made up the composites. He proposed that "statisticalconstancy" was attained after"thirty haphazard pictures of the same class [had] been combined."'68 Galton made more expansive claims forhis process,which he has described as a formof "pictorial statistics": Composite pictures are . . . much more than averages; they are rather the equivalents of those large statisticaltables whose totals, divided by the number of cases and entered on the bottom line, are the averages. They are real generalizations,because theyinclude the whole of the materialunder consideration.The blur of theiroutlines, which is never great in trulygeneric composites, except in unimpor-

67. Galtonacknowledged Spencer in an 1878 paperread beforethe Anthropological Institute, extractedin ibid.,p. 340. Spencer'spreviously unpublished 1846 proposalfor producing and superimposingphrenological diagrams of the head, "On a Proposed Cephalograph,"can be foundas an appendixto his An Autobiography,vol. 1, New York, Appleton,1904, pp. 634-638. Like Quetelet,Spencer appears not to have read Kant on the notionof an averagetype, or on any othertopic forthat matter (see David Wiltshire,The Social and PoliticalThought of Herbert Spencer,Oxford, Oxford University Press, 1978, p. 67). Spencer'sorganismic defense of a hierar- chicalsocial division of labor is articulatedin a reviewof the collected works of Plato and Hobbes: "The Social Organism,"The WestminsterReview, New Series,vol. 17, no. 1 (January1860), pp. 90-121. This extendedmetaphor goes so faras to comparethe circulation of blood withthat of money(p. 111). On the connectionsbetween Spencerian social Darwinismand eugenics,see GretaJones, Social Darwinism and English Thought, Sussex, Harvester,1980. 68. Galton,Inquiries, p. 17. 48 OCTOBER

tant details, measures the tendencyof individuals to deviate fromthe central type.69 In this passage the tension between claims forempirical specificityand claims for generalityreaches the point of logical rupture: what are we to make of this glib slide from"they include the whole" to "except unimportantdetails"? In his search fora type, Galton did not believe that anythingsignificant was lost in underexposure. This required an unacknowledged presupposition: only the gross featuresof the head mattered. Ears, forexample, which were highlymarked as signs in other physiognomic systems, both as individuating and as typical features, were not registeredat all by the composite process. (Later Galton sought to "recapture"small differencesor "unimportantdetails" by means of a technique he called "analyticalphotography," which superimposed positive and negative images, therebyisolating theirunshared elements.70) Just as he had acknowledged Quetelet as a source forhis earlier rankingof intelligence,so Galton claimed thatthe composite photographproduced an improved impression of l'hommemoyen: The process . . . of pictorial statistics[is] suitable to give us generic pictures of man, such as Quetelet obtained in outline by the ordi- nary numerical methods of statistics,as described in his work on An- thropomitrie. . . . By the process of composites we obtain a picture and not a mere outline.7" In effectGalton believed that he had translated the Gaussian error curve into pictorial form. The symmetricalbell curve now wore a human face. This was an extraordinaryhypostatization. Consider the way in which Galton conve- niently exiled blurring to the edgesof the composite, when in fact blurring would occur over the entire surface of the image, although less perceptibly. Only an imagination that wanted to seea visual analogue of the binomial curve would make this mistake, findingthe type at the center and the idiosyncratic and individual at the outer periphery. The frontispieceto Inquiriesinto Human Faculty consists of eight sets of composites. Galton describes these images as an integratedensemble in his text, in what amounts to an illustratedlecture on eugenics. The first,upper leftcom-

69. Francis Galton, "On Generic Images," Proceedingsof the Royal Institution, vol. 9 (1879), p. 166. 70. Francis Galton, "Analytical Photography,"Nature, vol. 18 (August 2, 1890), p. 383. 71. Francis Galton, "Generic Images," NineteenthCentury, vol. 6, no. 29 (July 1879), p. 162. In the related, previouslycited paper "On Generic Images," Galton stated that Quetelet was the first to give "the idea of type"a "rigorous interpretation"(p. 162). Ruth Schwartz Cowan has argued, followingKarl Pearson, that Quetelet was of no particular import in Galton's development as a statistician; but Cowan is interestedin Galton's position as a statisticianin the lineage of hereditarian thought,and not in his attemptto negotiate the merger of optical and statisticalmethods. That is, Cowan prefersto definebiostatistics as a science which began with Galton, a science having no prehereditarian precursor in Quetelet (see Sir FrancisGalton, pp. 145-200). FrancisGalton. Criminal Composites. c. 1878. PlateXXVII fromKarl Pearson,The Life, Letters and Labours of Francis Galton, vol. 2, 1924.

eoMPONENTS 8 CMPO NENN ONET 3,O

T f . .. .

-M MN N S : :. : : ...... ".... 50 OCTOBER

posite of six portraitmedallions of Alexander the Great serves Galton as an in- troductory,epistemological benchmark, not only to the series, but to the entire book. Oblivious to issues of style or artisticconvention, Galton assumed that individual engravers had erred in various ways in their representations.The composite, according to a Gaussian logic of averaged measurements, would contain a "truer likeness." An unspoken desire, however, lurks, behind this construction. Galton made many composites of Greek and Roman portrait coins and medallions, seeking in the blurred"likenesses" the vanished physiognomy of a higher race. Galton's next two sets of composites were made from members of the same family.With these he charged into the active terrainof eugenic research and manipulation. By exhibitingthe blending of individual characteristicsin a single composite image, Galton seems to have been searching for a ratio of hereditaryinfluence. He extended these experimentsto composites tracingthe lineage of race horses. The next composite was probably the most democratic constructionof Galton's entire career: a combination of portraitsof twelve officersand eleven enlisted men of the Royal Engineers. This was offeredas a "clue to the direction in which the stock of the English race might most easily be improved."72This utopian image was paired with its dystopian counterparts,generic images of disease and criminality. While tuberculosisseemed to produce a vaguely wan physiognomy,crime was less easy to type. Galton had obtained identificationphotographs of convicts fromthe Director of Prisons, Edmund Du Cane, and these were the source of his firstcomposites in 1878. Despite this early startin the search forthe biological criminal type, Galton came to a position that was less enthusiasticthan that of Lombroso: "The individual faces are villainous enough, but they are villainous in differentways, and when theyare combined, the individual pecu- liaritiesdisappear, and the common humanityof a low type is all that is left."73 Thus Galton seems to have dissolved the boundary between the criminal and the working-classpoor, the residuumthat so haunted thepolitical imaginationof the late-Victorian bourgeoisie. Given Galton's eugenic stance, this meant that he merely included the criminal in the general pool of the "unfit." Later, followingCharles Booth's sociological stratificationof the London population, Galton classified "criminals, semi-criminals, and loafers" as the worst of the eugenically unfit:the bottom one percent of the urban hierarchy. On this basis, he supportedlong sentences for"habitual criminals,"in hopes of "restrictingtheir opportunitiesfor producing low-class offspring."74 Galton concluded the introductorysample of composite portraitsin his In-

72. Galton, Inquiries,p. 14. 73. Ibid., p. 15. 74. FrancisGalton, Essays in Eugenics, London, Eugenics Education Society, 1909, pp. 8-9, 62. The Bodyand theArchive 51

quirieswith contrastedsets of composites made fromvery large samples: repre- senting"consumptive" and "not consumptive"cases. With these he underlined both the statisticaland the socialhygenic ambitions behind his optical process and his political program. Galton harbored other psychologicaland philosophicalambitions. In his earlier essays on "genericimages" he examined "analogies" between mental images, which he claimed consisted of "blended memories," and the genera produced by his optical process. Citing the Weber-Fechner Law of psychophysics, which demonstratedthat relative perceptual sensitivitydecreased as the level of stimulusincreased, Galton concluded that"the human mind is thereforea most imperfectapparatus forthe elaboration of general ideas," when compared with the relentlessand untiringquantitative consistencyof "pictorialstatistics."'75 In Inquiries,he returnedto this theme: "The ideal faces obtained by the method of composite portraitureappear to have a great deal in common with.., so-called abstract ideas." He wondered whether abstract ideas might not be more cor- rectlytermed "cumulative ideas."''76Galton's ratherreified notions of what con- stitutedthought is perhaps most clearly, if unwittingly,expressed in his off- hand definitionof introspection:"taking stock of my own mental furniture."''" The composite apparatus provided Galton with a model of scientificintel- ligence, a mechanical model of intellectuallabor. Furthermore,this intelligence answered to the logic of philosophicalrealism. Galton argued thathis composites refutednominalist approaches to the human sciences, demonstratingwith cer- tainty the reality of distinct racial types. This amounted to an essentialist physical anthropologyof race.78 It is not surprising,then, that Galton would come to regard his most suc- cessfulcomposite as that depicting"the Jewish type." In a historicalcontext in which therewas no clear anthropologicalconsensus on the racial or ethnicchar- acter of modernJews, Galton produced an image that was, according to Karl Pearson, "a landmark in composite photography":"We all know the Jewish boy, and Galton's portraiturebrings him before us in a way that only a great work of art could equal - scarcely excel, forthe artistwould only idealise from onemodel."79 This applause, ominous enough as it is, takes on an even more sinistertone in retrospectwhen one considers the line of influencewhich led fromAnglo-American eugenics to National Socialist Rassentheorie.80

75. Galton, "Generic Images," p. 169. 76. Galton, Inquiries,p. 183. 77. Ibid., p. 182. 78. Galton, "Generic Images," pp. 163-164. 79. Karl Pearson, The Life, Lettersand Laboursof Francis Galton, vol. 2, Cambridge, Cambridge University Press, 1924, p. 293. 80. On the role played by eugenics in Nazi racial policy, see Allan Chase, TheLegacy of Malthus: The Social Costs of the New ScientificRacism, Urbana, University of Illinois Press, 1980, pp. 342-360. Galton was asked to make the composites in 1883 byJosephJacobs, who was attemptingto 52 OCTOBER

...... " -......

FrancisGalton. The JewishType. 1883. PlateXXXV fromPearson.

Galton's composite process enjoyed a wide prestigeuntil about 1915. De- spite its origins in a discourse of racial essentialism, the composite was used to make a varietyof points, some of which favored"nurture" over "nature." For

demonstratethe existenceof a relativelypure racial typeof modernJew, intactdespite the Diaspora. For theportraits, Jacobs recruited boy studentsfrom the Jews' Free Schooland from theJewish Working Men's Club in London.Galton andJacobs both agreed that a racialtype had been produced,but theydisagreed profoundly on themoral essence of that type. Galton, the great quantifier,met his imaginaryOther: "The featurethat struck me most,as I drovethrough the... Jewishquarter, was the cold scanninggaze of man, woman,and child. . . . I felt,rightly or wrongly,that every one of themwas coolyappraising me at marketvalue, withoutthe slightest interestof any otherkind" ("Photographic Composites," The Photographic News, vol. 29, no. 1389 [April17, 1885]). Jacobsresponded to Galton'santi-Semitism with a morehonorific reading of thecomposites, suggesting that "here we have something.., morespiritual than a spirit. . . The compositeface mustrepresent this Jewish forefather. In theseJewish composites we have the nearestrepresentation we can hope to possessof the lad Samuel as he ministeredbefore the Ark, or theyouthful David whenhe tendedhis father'ssheep" ("The Jewish Type, and Galton'sCom- positePhotographs," The Photographic News, vol. 29, no. 1390 [April24, 1885]). Thus Jacobs countersGalton's myth of theJew as the embodimentof capitalwith a proto-Zionistmyth of origins.(On themedical and racial stereotypingofJews in thelate nineteenthcentury, and the Jewishreaction, see SanderGilman, "The Madnessof the Jews," in Differenceand Pathology: Stereo- typesof Sexuality, Race, and Madness, Ithaca, CornellUniversity Press, 1985, pp. 150-162.) The Bodyand theArchive 53

SO ...... IF

VIP no.

LewisHine. Compositephotograph ofchild laborers employedin cottonmill. 1913. (NationalGallery of Canada,Ottawa.) example, Lewis Hine made a number of crude composite prints of girl mill- workersin 1913, in what was evidentlyan attemptto trace the general effectsof factoryworking conditions on young bodies. And, in a curious twist,the book which provided the conclusive refutationfrom within criminologyof Lom- broso's theoryof the innate criminalwith the telltaleskull, Henry Goring's The EnglishConvict, opened its attackwith a comparison between compositesof free- hand drawingsand compositesof tracingsfrom photographs of criminalheads. The formerhad been used by Havelock Ellis to make his physiognomiccase in The Criminal.The discrepancybetween these and the tracingsrevealed a great degree of caricature in Ellis's pictures.81With both Hine and Goring, a faithin the objectivityof the camera persisted. However, withthe general demise of an optical model of empiricism,Galton's hybridizationof the camera and the statistical table approached extinction. Photographycontinued to serve the sci-

81. HenryGoring, The English Convict: A StatisticalStudy, London, H. M. StationeryOffice, 1913. Lombroso'stheoretical fixation with convict head size had alreadybeen undercutwithin physicalanthropology by Franz Boas. See his 1910-1913essay, "Changesin BodilyForm of Descendantsof Immigrants,"in Race, Languageand Culture,New York, Macmillan, 1949, pp. 60-75. 54 OCTOBER

ences, but in a less grandiose and exalted fashion,and consequentlywith more modest- and frequentlymore casual- truthclaims, especially on the periphery of the social sciences. In retrospect,the Galtonian composite can be seen as the collapsed version of the archive. In this blurred configuration,the archive attemptsto exist as a potent single image, and the single image attemptsto achieve the authority of the archive, of the general, abstract proposition. Galton was certainlya vo- ciferous ideologue for the extension and elaboration of archival methods. He actively promoted familial self-surveillancefor hereditarian purposes, calling forhis readers to "obtain photographsand ordinarymeasurements periodically of themselves and their children, making it a familycustom to do so."82 His model here was the British Admiralty'svoluminous registryof sailors. Here again, eugenics modeled itselfon the military.Galton founded an Anthropo- metrical Laboratory in 1884, situated firstat the InternationalHealth Exposi- tion, then moving to the Science Museum in South Kensington. Nine thousand visitorswere measured, paying three or fourpence each for the privilege of contributingto Galton's eugenic research.83 Althoughmarried formany years, Galton leftno children. Instead, he left behind an immense archive of documents. One curious aspect of Karl Pearson's massive pharaonic biographyof Galton is its profusionof photographicillustra- tions, including not only Galton's many photographicexperiments, but also a kind of intermittentfamily album of more personal pictures. Eugenics was a utopian ideology, but it was a utopianism inspired and haunted by a sense of social decline and exhaustion. Where Quetelet had approached the question of the average withoptimism, finding in averages both a moral and an aesthetic ideal, Galton's eugenicist hope for an improved racial stock was always limited by his early discovery that successive generationsof eugenically bred stock tended to regress back toward the mean, and "medi- ocrity."84Thus the fantasyof absolute racial bettermentwas haunted by what must have seemed a kind of biological entropy.85Later, in the twentiethcen- tury, eugenics would only operate with brutal certaintyin its negative mode, throughthe sterilizationand exterminationof the Other. What can we conclude, finally, about the photographic problems encountered and "solved" by Bertillon,the nominalist detective, and Galton, the essentialistbiometrician? The American philosopher and semiotician Charles

82. Galton, Inquiries,p. 43. 83. Pearson, Life, Lettersand Labours,vol. 2, p. 357. 84. Galton, HereditaryGenius, pp. xvii-xviii. 85. On the cultural resonance of the concept of entropy in the nineteenthcentury, see Anson Rabinbach, "The Body without Fatigue: A Nineteenth Century Utopia," in Seymour Drescher et al., eds., PoliticalSymbolism in ModernEurope: Essays in Honorof GeorgeMosse, New Brunswick, Transaction Books, 1982, pp. 42-62. The Bodyand theArchive 55

Sanders Peirce, their contemporary,made a useful distinctionbetween signs that referredto theirobjects indexically, and those that operated symbolically. To the extent that photographs are "effectsof the radiations fromthe object," they are indexical signs, as are all signs which registera physical trace. Sym- bols, on the other hand, signifyby virtue of conventions or rules. Verbal language in general, and all conceptual thought,is symbolicin Peirce's system.86 Paradoxically, Bertillon, in taming the photograph by subordinatingit to the verbal text of the portraitparld, remained wedded to an indexicalorder of meaning. The photographwas nothingmore than the physical traceof its contingent instance. Galton, in seeking the apotheosis of the optical, attemptedto elevate the indexical photographiccomposite to the level of the symbolic,thus express- ing a generallaw through the accretion of contingentinstances. In so doing, Galton produced an unwittingcaricature of inductive reason. The composites signified,not by embodyingthe law of error,but by being rhetoricallyannexed to thatlaw. Galton's ambition, although scientistic,was not unlike thatof those otherelevators of photography,the neosymbolistsof the Photo Secession. Both Galton and Stieglitzwanted somethingmore than a mere trace, somethingthat would match or surpass the abstract capabilities of the imaginative or general- izing intellect. In both cases, meaning that was ferventlybelieved to emerge fromthe "organic"character of the sign was in fact certifiedby a hidden fram- ing convention. Bertillon, on the other hand, kept his (or at least his under- lings') eye and nose to the ground. This made him, in the prejudiced and probably inconsequential opinion of one of his biographers, Henry Rhodes, "the most advanced photographerin Europe."87 Despite theirdifferences, both Ber- tillon and Galton were caught up in the attempt to preserve the value of an older, optical model of truthin a historicalcontext in which abstract, statistical procedures seemed to offerthe high road to social truthand social control.

III.

The firstrigorous systemof archival cataloguing and retrievalof photographs was that invented by Bertillon. Bertillon's nominalist system of iden- tificationand Galton's essentialist system of typologyconstitute not only the two poles of positivistattempts to regulate social deviance by means of photography, but also the two poles of these attemptsto regulate the semantic traffic in photographs. Bertillon sought to embed the photograph in the archive. Galton sought to embed the archive in the photograph. While their projects were specialized and idiosyncratic, these pioneers of scientificpolicing and

86. Charles Sanders Peirce, The PhilosophicalWritings of Peirce, ed. Justus Buchler, New York, Dover, 1955, pp. 99-119. 87. Rhodes, p. 191. 56 OCTOBER

eugenics mapped out general parameters for the bureaucratic handling of visual documents. It is quite extraordinarythat historiesof photographyhave been writtenthus far with little more than passing referenceto their work. I suspect that this has somethingto do with a certain bourgeois scholarlydiscre- tion concerningthe dirtywork of modernization, especially when the status of photography as a fine art is at stake.88 It is even more extraordinarythat historiesof social documentaryphotography have been writtenwithout taking the police into account. Here the issue is the maintenance of a certain liberal humanistmyth of the whollybenign originsof sociallyconcerned photography.89 Roughly between 1880 and 1910, the archive became the dominant institutional basis for photographic meaning. Increasingly, photographic archives were seen as central to a bewildering range of empirical disciplines, ranging fromart historyto militaryintelligence.90 Bertillon had demonstratedthe use- fulnessof his model forpolice purposes, but otherdisciplines faced significantly differentproblems of image cataloguing. An emergent bibliographicscience provided the utopian model of classificationfor these expansive and unrulycollec- tions of photographs. Here again Bertillonwas prescientin his effortto reduce the multiple signs of the criminal body to a textual shorthand and numerical series. At a varietyof separate but related congresseson the internationalization and standardizationof photographicand bibliographicmethods, held between 1895 and 1910, it was recommended that photographsbe catalogued topically according to the decimal system invented by the American librarian Melvil Dewey in 1876. The lingering prestige of optical empiricism was sufficiently strong to ensure that the terrain of the photographable was still regarded as roughlycongruent with that of knowledge in general. The Institutefor Interna- tional Bibliographybuilt on the universalistlogic of the eighteenth-centuryen- cyclopedists. But appropriate to the triumphal years of an epoch of scientific

88. Compare Josef Maria Eder, Historyof Photography, trans. Edward Epstean, New York, Co- lumbia UniversityPress, 1945, with Beaumont Newhall, Photography.:A Short Critical History, New York, Museum of Modern Art, 1938. Eder, very much part of the movement to rationalize photography during the firstdecade of this century, is quite willing to treat police photographyas a proper object of his narrative. Eder in fact wrote an introduction to a German edition of Bertillon's manual (Die gerichtlichePhotographie, Halle a. S., Knapp, 1895). Newhall, on the other hand, wrote a modernist historyin 1938 that privileged technical photography, including First World War aerial reconnaissance work, without once mentioning the use of photographyby the police. Clearly, Newhall found it easier to speak of the more glamorous, abstract, and chivalrous state violence of early air power than to dwell on the everyday state violence of the police. 89. An exception would be Sally Stein's revisionistaccount ofJacob Riis, "Making Connections with the Camera: Photography and Social Mobility in the Career ofJacob Riis," Afterimage,vol. 10, no. 10 (May 1983), pp. 9-16. 90. Compare Bernard Berenson, "Isochromatic Photography and Venetian Pictures," The Na- tion,vol. 57, no. 1480 (November 9, 1893), pp. 346-347, with Fred Jane, "Preface," Fighting Ships, London, Marsten, 1905-1906, p. 2. However differenttheir objects, these texts share an enthusiasm for large quantities of well-definedphotographs. The Bodyand theArchive 57

positivism and the early years of bureaucratic rationalization, a grandiose clerical mentalityhad now taken hold.91 The new scientificbibliographers articulated an operationalist model of knowledge, based on the "general equivalence" established by the numerical shorthandcode. This was a systemfor regulating and accelerating the flowof texts,profoundly linked to the logic of Taylorism. Is it surprisingthat the main reading room of that American Beaux-Arts temple of democratic and imperial knowledge, the Library of Congress, built during this period of bibliographic rationalization, should so closely resemble the Panopticon, or that the outer perimeterof the building should bear thirty-three"ethnological heads" of various racial types?92Or is it any more surprisingthat the same American man- ufacturingcompany produced Bertillon cabinets, business files, and library card catalogue cabinets?93 Photographywas to be both an objectand meansof bibliographicrational- ization. The latter possibilityemerged fromthe development of microfilmre- productionof documents.Just as photographswere to be incorporatedinto the realm of the text, so also the text could be incorporated into the realm of the photograph. If photographyretained its prestigeas a universal language, it increasingly did so in conjunction with a textual paradigm that was housed within the library.94 The grand ambitions of the new encyclopedists of photography were eventually realized but not in the grand encyclopedic fashion one mighthave expected. With the increasing specialization of intellectualdisciplines, archives tended to remain segregated. Nonetheless, the dominant culture of photography did rely heavily on the archival model for its legitimacy. The shadowy presence of the archive authenticatedthe truthclaims made forindividual pho-

91. The Institut International de Bibliographie, founded in 1895 with headquarters in Brus- sels, campaigned for the establishment of a bibliographiauniversalis registered on standardized filing cards. Following Dewey, the Institute recommended that literature on photography be assigned the seventh position within the graphic arts, which were in turn assigned the seventh position withinthe categories of human knowledge. The last subcategorywithin the classification of photographywas to hold photographicprints. See the Institute'sfollowing publications: Manuel pour l'usagedu ripertoirebibliographique de la photographieitabli d'apresla classificationdicimale, Brussels (copublished with the Socidte Frangaise de la Photographie), 1900; Codepour l'organisationde la documentationphotographique, Brussels, 1910. 92. I am gratefulto Daniel Bluestone for pointing out this latter architecturaldetail. For a contemporarydescription of the heads, see Herbert Small, Handbookof the New Libraryof Congress, Boston, Curtis and Cameron, 1901, pp. 13-16. 93. See the followingcatalogues published by the Yawman and Erbe Mfg. Co.: Card Ledger Systemand Cabinets,Rochester, N.Y., 1904; CriminalIdentification by "Y and E"- Bertillonand Finger PrintSystems, Rochester, 1913; and "Y and E" LibraryEquipment, Rochester, 192-?. 94. On early microfilm,see Livre microphotographique:le bibliophoto ou livreia projection, Brussels, InstitutInternational de Bibliographie, 1911. On the more recent conversion of the photograph from library-documentto museum-object, see Douglas Crimp, "The Museum's Old/The Li- brary's New Subject," Parachute,no. 22 (Spring 1981), pp. 32-37. 58 OCTOBER

tographs, especially within the emerging mass media. The authorityof any particular syntagmatic configurationwas underwrittenby the encyclopedic authorityof the archive. One example will suffice.Companies like Keystone Views or Underwood and Underwood seriallypublished shortpictorial group- ings of stereographcards. Although individual sequences of pictureswere often organized according to a narrativelogic, one sees clearlythat the overall struc- ture was informednot by a narrativeparadigm, but by the paradigm of the archive. Afterall, the sequence could be rearranged; its temporalitywas indeter- minate, its narrativityrelatively weak. The pleasures of this discourse were grounded not in narrativenecessarily, but in archival play, in substitution,and in a voracious optical encyclopedism.There were always more images to be ac- quired, obtainable at a price, froma relentlesslyexpanding, globally dispersed picture-gatheringagency.95 Archival rationalization was most imperative for those modes of photographic realism that were instrumental,that were designed to contributedi- rectly or indirectlyto the practical transformationor manipulation of their referent.Can any connectionsbe traced between the archival mode of photography and the emergence of photographicmodernism? To what degree did self- conscious modernistpractice accommodate itselfto the model of the archive?To what degree did modernistsconsciously or unconsciously resist or subvert the model of the archive, which tended to relegate the individual photographerto the statusof a detail worker,providing fragmentary images foran apparatus be- yond his or her control?Detailed answers to thisquestion are clearlybeyond the scope of this essay. But a few provisional lines of investigationcan be charted. The protomodernismof the Photo Secession and its affiliatedmovements, extendingroughly to 1916, can be seen as an attemptto resistthe achival mode through a strategyof avoidance and denial based on craft production. The elegantfewwere opposed to the mechanized many,in termsboth of images and authors. This strategyrequired the ostentatiousdisplay of the "honorificmarks of hand labor," to borrow the phrase coined by the American sociologistThor- stein Veblen in 1899.96 After 1916, however, aestheticallyambitious photog- raphers abandoned the painterlyand embraced pictorial rhetoricsmuch closer to those already operative within the instrumentalrealist and archival paradigms. Understandably, a variety of contradictoryattitudes to the archive emerge within photographic discourse in the 1920s. Some modernists em-

95. This suggeststhat the historiography ofphotography will have to approachthe question of an "institutionalmode" in differentterms than those already developed for the historiography of cinema. See, forexample, Noel Burch,"Film's Institutional Mode of Representationand the SovietResponse," October, no. 11 (Winter1979), pp. 77-96. 96. ThorsteinVeblen, TheTheory of the Leisure Class: An Economic Study of Institutions, New York, Modern Library,1934, pp. 163-164. The Bodyand theArchive 59

braced the archival paradigm: August Sander is a case in point. Others resisted throughmodernist reworkings of the antipositivismand antirationalismof the Photo Secession: the later Stieglitzand Edward Weston are obvious examples. In many respects the most complicated and intellectuallysophisticated response to the model of the archive was thatof Walker Evans. Evans's book sequences, especially in his 1938 AmericanPhotographs, can be read as attemptsto counterpose the "poetic" structureof the sequence to the model of the archive. Evans began the book with a prefatorynote reclaiminghis photographsfrom the various archival repositorieswhich held copyrightto or authorityover his pictures.97Furthermore, the firstphotograph in the book describes a site of the archival and instrumentalmode's proliferationinto the spaces of metropolitan daily life in the 1930s: License-PhotoStudio, New York,1943. We now know that Evans was fascinated with police photographs during the period in which he made the photographsin this book. A terse topical list on "New York societyin the 1930s" contains a central, telegraphic,underlined inscription:"This project getpolice cards."98 Certainly Evans's subway photographs of the late 1930s and early 1940s are evidence of a sophisticateddialogue withthe empirical methods of the detectivepolice. Evans styledhimself as a flaneur,and late in lifelikened his sensibilityto that of Baudelaire. Though Walter Benjamin had proposed that "no matterwhat trail thefidneur may follow, every one will lead him to a crime,"'99Evans avoided his final rendezvous. This final detour was explicitly described in a 1971 interviewin which he took care to distinguishbetween his own "documentarystyle" and a "literaldocument" such as "a police photograph of a murder scene."100He stressed the necessary element of poetic transcen- dence in any art photograph of consequence. The elderly Evans, transformed into the senior figureof modernistgenius by a curatorialapparatus with its own archival imperative,could no longer recognize the combative and antiarchival stance of his earlier sequential work. Evans was forcedto fallback on an organ- icistnotion of style,searching for that refinedsurplus of stylisticmeaning which would guarantee his authorship,and which in general served to distinguishthe art photographerfrom a flunkyin a hierarchyof flunkies. With the advent of postmodernism,many photographershave abandoned any serious commitmentto stylistictranscendence, but they fail to recognize the degree to which theyshare Evans's social fatalism,his sense of the immuta- bilityof the existingsocial order. Modernism offersother models, however, in-

97. Walker Evans, AmericanPhotographs, New York, Museum of Modern Art, 1938. 98. Reproduced in JerryThompson, ed., WalkerEvans at Work,New York, Harper and Row, 1982, p. 107. 99. Walter Benjamin, "The Paris of the Second Empire in Baudelaire," in CharlesBaudelaire, p. 41. 100. Leslie Katz, "Interview with Walker Evans," Art in America,vol. 59, no. 2 (March-April 1971), p. 87. WalkerEvans. License-Photo Studio, New York, 1934. Plate 1 fromAmerican Photographs, 1938.

SVrn

NIi

...... i...... ? : x + ... . " AP'I(AIN ; , : i -! i

(AR FO RAD ES

cluding more militantand equally intelligentmodels of photographicpractice. Consider Camille Recht's reading of the photographs of Eugene Atget, a photographer of acknowledged import in Evans's own development. Recht commentson interiorviews "whichremind us of a police photographof a crime scene" and then on "the photographof a worker'sdwelling which testifiesto the housing problem." For Recht, the proximityof a "nuptial bed and an unavoid- able chimneyflue," provided grimlycomic testimonyof everydaylife in an ex- ploitative social formation.101This emphasis on the telling detail, the metonymicfragment that points to the systemiccrimes of the powerful,would be repeated and refinedin the writingsof Walter Benjamin.102Our tendencyto associate Benjamin with the theoryand practice of montage tends to obscure the degree to which he built his modernismfrom an empiricistmodel, froma model of careful, idiosyncraticobservation of detail. This model could argue both forthe photographeras monteur,and forthe photographeras revolutionary spy or detective, or, more "respectably,"as criticaljournalist of the working class.

101. Camille Recht, introductionto Eugene Atget,Lichtbilder, Paris and Leipzig, Henri Jonquieres,1930, pp. 18-19 (my translation). 102. See Benjamin's1931 essay "A ShortHistory of Photography,"trans. Stanley Mitchell, Screen,vol. 13, no. 1 (Spring1972), p. 25. EugeneAtget. Plate 12 fromLichtbilder, 1930.

-II....

Metricalphotograph and planimetricsketch. From A. Bertillonand A. Chervin,Anthropologie metrique, 1909.

THitNSORNIATION OF.iLk PHOTOf:RAPHIRCI-CO)TRE

( 110 0 015 PLA i ik 'it Q: iPi~ii-i-i'a-i~iiui-;'i-iii:=iioiii~iit i i i 1 T i i l TI s ty lii

!!!ii!!i~i~!!!i:iii! !!i~i!!!!!!!!!!! O ll1?,iiiiiiiiiiiiiiii~iiiiiiiiiiiii

"41~~ ii~!!~i

...... iiii iii!i i!iii!i! ...... iiii iiii 62 OCTOBER

This essay could end with this sketchof modernistresponses to the prior institutionalizationof the instrumentalrealist archive. Social historywould lead to art history,and we would arriveat a safe archivalclosure. Unfortunately, Bertillon and Galton are still with us. "Bertillon"survives in the operations of the national securitystate, in the condition of intensiveand extensive surveillance that characterizesboth everydaylife and the geopolitical sphere. "Galton" lives in the renewed authorityof biological determinism,founded in the increased hegemony of the political Right in the Western democracies. That is, Galton lives quite specificallyin the neo-Spencerian pronouncementsof Rea- ganism, Thatcherism, and the French National Front.103 Galton's spirit also survives in the neoeugenicist implicationsof some of the new biotechnologies. These are political issues. As such, their resonance can be heard in the aesthetic sphere. In the United States in the 1970s, a number of works, pri- marily in filmand video, took an aggressive stance toward both biological de- terminismand the prerogativesof the police. Martha Rosler's video "opera" The VitalStatistics of a Citizen,Simply Obtained (1976) retainsits forceas an allegorical feministattack on the normalizinglegacy of Quetelet and Galton. Other, more nominalist works, took on the police at the level of counter-testimonyand counter-surveillance.I am thinkinghere of a number of documentary films: Howard Gray's and Michael Alk's The Murderof Fred Hampton(1971), Cinda Firestone's Attica (1973), and the Pacific Street Film Collective's Red Squad (1972). These examples tend to be forgottenor overlooked in a contemporary art scene rife with a variety of what can be termed "neophysiognomic"concerns. The body has returned with a vengeance. The heavily expressionist character of this return makes the scientisticand racialist underpinnings of physiognomyseem rather remote. In photography, however, this lineage is harder to repress. In one particularlytroubling instance, this returnedbody is specificallyGaltonian in its configuration.I referhere to the computergenerated composites of Nancy Burson, enveloped in a promotional discourse so appall- inglystupid in its fetishisticbelief in cybernetictruth and its desperate desire to remain grounded in the optical and organic that it would be dismissable were it not forits smug scientism.For an artistor criticto resurrectthe methodsof biosocial typologywithout once acknowledging the historicalcontext and conse- quences of these procedures is naive at best and cynical at worst.104 In the interestsof a certain internationalism,however, I want to end with a story that takes us outside the contemporaryart scene and away fromthe simultaneouslyinflated and deflated figureof the postmodernistauthor. This anecdote might suggest somethingof the hardships and dilemmas of a photo-

103. For an example of the high regardfor Galton among contemporaryhereditarians, see H. J. Eysenck'sintroduction to the 1978 editionof HereditaryGenius previously cited. 104. See NancyBurson et al., Composites:Computer Generated Portraits, New York,William Morrow, 1986. The Bodyand theArchive 63

graphic practice engaged in frombelow, a photographicpractice on ground patrolled by the police. In 1967, a young Black South African photographer named Ernest Cole published a book in the United States called House ofBond- age. Cole's book and his storyare remarkable. In order to photographa broad range of South Africansociety, Cole had firstto change his racial classification fromblack to colored, no mean feat in a world of multiplebureaus of identity, staffedby officialswho have mastered a subtle bureaucratic taxonomyof even the offhandgestures of the differentracial and ethnicgroups. He counteredthis apparatus, probably the last physiognomicsystem of domination in the world, with a descriptivestrategy of his own, mapping out the various checkpointsin the multiple channels of apartheid. Cole photographed during a period of relative political "calm" in South Africa, midway between the Sharpeville massacre of 1960 and the Soweto students' revoltof 1976. At a time when black resistancewas fragmentedand sub- terranean in the wake of the banning of the main opposition groups, he dis- covered a limited, and by his own account problematic,figure of resistancein young black toughs, or tsotsis,who lived lives of pettycriminality. Cole photo-

oo? m-F,Y. ...

......

FromErnest Cole, House of Bondage, 1967. 64 OCTOBER

graphed tsotsismugging a white workerfor his pay envelope, as well as a scene of a white man slapping a black beggar child. And he regularlyphotographed the routine passbook arrests of blacks who were caught outside the zones in which theywere permittedto travel. As mightbe expected, Cole's documenta- tion of the everyday flowsof power, survival, and criminal resistance got him into trouble with the law. He was questioned repeatedly by police, who assumed he was carryingstolen camera equipment. Finally he was stopped after photographingpassbook arrests. Asked to explain himself,he claimed to be making a documentary on juvenile delinquency. Sensing his criminological promise, the police, who then as now operated througha pervasive systemof informers,invited him to join the ranks. At that point, Cole decided to leave the country while he still could. House of Bondagewas assembled from the negatives he smuggled out of South Africa. Since publishing his book in exile, Cole has disappeared fromthe world of professionalphotojournalism.s05 The example of Cole's work suggests that we would be wise to avoid an overly monolithicconception of realism. Not all realisms necessarilyplay into the hands of the police, despite Theodor Adorno's remark, designed to lam- poon a Leninist epistemologyonce and forall, that"knowledge has not, like the state police, a rogues' galleryof its objects."'06 If we are to listen to, and act in solidaritywith, the polyphonic testimonyof the oppressed and exploited, we should recognizethat some of thistestimony, like Cole's, will take the ambiguous form of visual documents, documents of the "microphysics"of barbarism. These documents can easily fall into the hands of the police or theirintellectual apologists. Our problem, as artistsand intellectualsliving near but not at the center of a global systemof power, will be to help prevent the cancellation of that testimonyby more authoritativeand officialtexts.

105. ErnestCole (withThomas Flaherty),House of Bondage, New York,Random House, 1967. For the accountof Cole's own struggleto producethe pictures in thebook, I have reliedupon JosephLelyveld's introduction, "One of the Least-KnownCountries in theWorld," pp. 7-24. 106. TheodorAdorno, Negative Dialectics, trans. E. B. Ashton,New York,Seabury, 1973, p. 206.