How to Do the History of Psychoanalysis:

A Reading of Freud's Three Essays

on the Theory of Sexuality

Arnold I. Davidson

I have two primary aims in the following paper, aims that are inextricably
intertwined. First, I want to raise some historiographical and epistemological
issues about how to write the history of psychoanalysis. Although they
arise quite generally in the history of science, these issues have a special
status and urgency when the domain is the history of psychoanalysis.
Second, in light of the epistemological and methodological orientation
that I am going to advocate, I want to begin a reading of Freud's Three
Essays on the Theoryof Sexuality, one whose specificity is a function of my
attachment to this orientation, to a particular way of doing the history
of psychoanalysis. Despite the enormous number of pages that have been
written on Freud's ThreeEssays,it is very easy to underestimate the density
of this book, a density at once historical, rhetorical, and conceptual. This
underestimation stems in part from historiographical presumptions that
quite quickly misdirect us away from the fundamental issues.
In raising questions about the historiography of the history of science,
I obviously cannot begin at the beginning. So let me begin much further
along, with the writings of Michel Foucault. I think of the work of Foucault,
in conjunction with that of Gaston Bachelard and Georges Canguilhem,
as exemplifying a very distinctive perspective about how to write the

Discussions, both recent and ancient, with Dan Brudney, Nancy Cartwright, Peter
Galison, Erin Kelly, and David Wellbery have greatly benefited this paper. Conversations
with Stanley Cavell about how to approach the texts of Freud were enormously helpful.
A version of this paper was given as a talk to the Institute for Psychoanalytic Training and
Research in New York, and I am grateful for the discussion that followed my presentation.

? 1987 by Arnold I. Davidson. All rights reserved. Permission to reprint may be obtained only from the author.

252

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 253

history of science. In the English-speaking world, perhaps only the work

of Ian Hacking both shares this perspective and ranks with its French
counterparts in terms of originality and quality. No brief summary can
avoid eliding the differences between Bachelard, Canguilhem, Hacking,
and Foucault; indeed, the summary I am going to produce does not
even fully capture Foucault's perspective, which he called "archaeology."'
But this sketch will have to do for the purposes I have in mind here,
whose ultimate aim is to reorient our approach to the history of psy-
choanalysis.
In a 1977 interview, Foucault gave what we might take to be a one-
sentence summary of his archaeological method: "'Truth' is to be under-
stood as a system of ordered procedures for the production, regulation,
distribution, circulation and operation of statements."2 Given this char-
acterization of his standpoint, we should think of Foucault as having
undertaken in his archaeological works to write a history of statements
that claim the status of truth, a history of these systems of ordered
procedures. The attempt to write such a history involves isolating certain
kinds of discursive practices--practices for the production of state-
ments-which will be "characterized by the delimitation of a field of
objects, the definition of a legitimate perspective for the agent of knowledge,
and the fixing of norms for the elaboration of concepts and theories.
Thus, each discursive practice implies a play of prescriptions that designate
its exclusions and choices."3 Foucault's project, announced in the foreword
to the English edition of The Orderof Things, was to write the history of
what Hacking has called the immaturesciences-those sciences that, in
Foucault's words, are "considered too tinged with empirical thought, too
exposed to the vagaries of chance or imagery, to age-old traditions and
external events, for it to be supposed that their history could be anything

1. The sketch that follows reproduces, with some omissions and additions, the beginning
of my "Archeology, Genealogy, Ethics," in Michel Foucault:A CriticalReader, ed. David Hoy
(London, 1986), pp. 221-34.
2. Michel Foucault, "Truth and Power," Power/Knowledge:SelectedInterviewsand Other
Writings, 1972-1977, ed. Colin Gordon, trans. Gordon, Leo Marshall, John Mepham, and
Kate Soper (New York, 1980), p. 133.
3. Foucault, "History of Systems of Thought," Language, Counter-Memory,Practice:
SelectedEssays and Interviews, ed. Donald F. Bouchard, trans. Bouchard and Sherry Simon
(Ithaca, N.Y., 1970), p. 199.

Arnold I. Davidson is assistantprofessor in the department of phi-

losophy, the Committee on the Conceptual Foundationsof Science, and
the Committee on General Studies in the Humanities at the University
of Chicago.He is currentlywritinga book on the historyand epistemology
of nineteenth-century psychiatrictheories of sexuality.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 254 Arnold I. Davidson Freud's Three Essays

other than irregular"4-from the standpoint of an archaeology of discursive

practices.5 Foucault made the claim, perhaps commonplace now, but
bold and even radical when he first wrote, that this kind of knowledge
possesses a well-defined regularity, that a history of this knowledge can
exhibit systems of rules, and their transformations, which make different
kinds of statements possible. These rules are, however, never formulated
by the participants in the discursive practice; they are not available to
their consciousness but constitute what Foucault once called the "positive
unconscious of knowledge."6
If these rules are both relatively autonomous and anonymous, if
they make it possible for individuals to make the claims they do when
they do, then the history of such rules and such knowledge will not look
like the sort of history with which we are most familiar. It will not, for
example, necessarily group sets of regularities around individual works
and authors; nor will it rest content with the ordinary boundaries of what
we think of as a science or a discipline. It will rather force regroupings
of statements and practices into a "a new and occasionally unexpected
unity."'7Because Foucault wanted to describe discursive practices from
the standpoint of archaeology, the theme of discontinuity was prominent
in some of his major works. The unearthing of discontinuities between
systems of knowledge is not an assumption of his method but a consequence
of it. If one sets out to describe the historical trajectories of the sciences
in terms of anonymous rules for the formation and production of state-
ments, then what looked continuous from some other perspective now
may very well appear radically discontinuous. Problems of periodization
and of the unity of a domain may be almost entirely transformed: one
will find, for example, that new kinds of statements which seem to be
mere incremental additions to scientific knowledge are in fact made possible
only because underlying rules for the production of discourse have sig-
nificantly altered. However, the method of archaeology also makes possible
the discovery of new continuities, overlooked because of a surface ap-
pearance of discontinuity. Archaeology makes no presumption about the
predominance of discontinuity over continuity in the history of knowledge;
but it does make it extremely likely that what had been taken to be natural
groupings of thought will turn out, at this new level of analysis, to be
quite unnatural indeed.
In other writings, I have tried to adopt and adapt Foucault's ar-
chaeological perspective, using it to write a history of nineteenth-century

4. Foucault, The Orderof Things:An Archaeologyof theHuman Sciences(New York, 1970),

p. ix.
5. See Ian Hacking, "Michel Foucault's Immature Science," Nouis 13 (Mar. 1979):
39-51.
6. Foucault, The Orderof Things, p. xi.
7. Foucault, "History of Systems of Thought," p. 200.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 255

psychiatric theories of sexuality.8 I have argued that starting around 1870

a new psychiatric style of reasoning about diseases emerges, one that
makes possible, among other things, statements about sexual perver-
sion-about homosexuality, fetishism, sadism, and masochism-that then
quickly become commonplaces in discussions of sexuality. The appearance
and proliferation of these statements were a direct consequence of this
new style of reasoning, which we can think of, in Foucault's terms, as
the birth of a new discursive practice. An epistemologically central con-
stituent of a style of reasoning, as I interpret it, is a set of concepts linked
together by specifiable rules that determine what statements can and
cannot be made with the concepts.9 So to write a history of nineteenth-
century psychiatry by way of this notion requires writing a history of the
emergence of a new system of concepts and showing how these concepts
are internally related by a set of rules to form what we might think of
as a determinate conceptual space. We want to see what concepts, connected
in what particular ways, allowed statements about sexual perversions that
had never been made before, allowed the creation of a new object of
medical discourse-sexuality. Thus, I have urged that we need a conceptual
history of sexuality, without which we cannot know what was being talked
about when the domain of psychiatricdiscourse became fixated on sexuality.
This same kind of method was employed by Heinrich Wolfflin in
his Principles of Art History: The Problemof the Developmentof Style in Later
Art. Wolfflin characterized the differences between classic and baroque
art in terms of two distinct systems of determining concepts. He tried to
show the way in which the features of classic art were linked together to
form a specific classic visual space, while opposing features were linked
together to form a distinctive baroque visual space.'0 It is no surprise
that Paul Veyne, in his inaugural lecture at the Collbge de France, has
conjoined the names of Wolfflin and Foucault; nor is it a surprise, re-
membering the derivative role that great men in the history of science
play in Foucault's work, that Arnold Hauser has referred to Wolfflin's
art history as "art history without names.""
Whatever the plausibility of an art history without names, and whatever
the general applicability of this methodological perspective in the history

8. See my "Closing Up the Corpses: Diseases of Sexuality and the Emergence of the
Psychiatric Style of Reasoning," in Handbookfor the History of Psychiatry,ed. J. Gach and
E. Wallace, forthcoming; and my "Sex and the Emergence of Sexuality," forthcoming.
9. I discuss this notion at length in an unpublished paper, "Styles of Reasoning,
Conceptual History, and the Emergence of Psychiatry."
10. See Heinrich Wolfflin, Principles of Art History: The Problem of the Developmentof
Style in Later Art, trans. M. D. Hottinger (New York, 1950).
11. See Paul Veyne, L'Inventaire des diffrrences:leqon inaugurale au College de France
(Paris, 1976); see Arnold Hauser, The Philosophyof Art History(Evanston, Ill., 1985). Hauser
is referring to Wolfflin's own phrase, "Kuntsgeschichte ohne Namen," which appears in
the foreword to the first edition of Principlesof Art History. This foreword was omitted from
later editions, and from the English translation of Wolfflin's book.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 256 Arnold I. Davidson Freud's Three Essays

of science, a thoroughgoing skepticism about its usefulness for writing

the history of psychoanalysis might well persist. Since psychoanalysis is
so completely intertwined with the name of Sigmund Freud, it is natural
to object that writing its history without his name would not be to write
its history at all. It is, no doubt, a peculiar feature of psychoanalysis, a
feature that requires a more detailed account than I can provide here,
that no matter what one takes as the last word of psychoanalysis, its first
and second words are always the words of Freud. And this is not merely
because Freud was the originator of psychoanalysis but primarily because
the central concepts, claims, and problems of psychoanalysis have not
received deeper specification beyond their congealment in his texts. So
there is an obvious sense in which any history of psychoanalysis must
continuously invoke the name of Freud. This fact, however, does not,
and should not, settle the question of what form this invocation should
take. Wolfflin was not reluctant to discuss the great works of classic and
baroque art; he wanted to demonstrate that this greatness was not in-
compatible with their artists being subject to specifiable constraints. "Not
everything is possible at all times," Wolfflin famously remarked, and his
art history, without names, was meant to conceptualize the limits of the
artistically possible in a given historical period and to show how a change
in constraints could lead to a reorganization of the limits of the possible.12
To do this successfully, W1olfflinhad to operate at a level distinct from
individual biography and psychology. In writing the history of psycho-
analysis, I want to preserve this level, one whose articulation requires a
history of a structurally related system of concepts, a conceptual space,
that lies below, or behind, the work of any particular author, even great
works of great authors.
Two competing myths about Freud have gradually developed. The
first myth, that of official psychoanalysis, depicts Freud as a lonely genius,
isolated and ostracized by his colleagues, fashioning psychoanalysis single-
handedly and in perpetual struggle with the world at large. The history
of psychoanalysis under the sway of this myth has become the story of
Freud as triumphant revolutionary. The second, opposing myth pictures
Freud as getting all of his ideas from someone else-usually Wilhelm
Fliess, although the names of Jean Martin Charcot, Havelock Ellis, and
Albert Moll, among many others, are also mentioned frequently-and
taking credit for what were in fact no more than minor modifications in
previously developed theories. This is the myth of the career discontents,
and the history of psychoanalysis dominated by it has become the story
of Freud as demagogue, usurper, and megalomaniac. To the first myth,
one can reply ex nihilo nihil fit, which is as appropriate a slogan in the
history of science as in theology. The second myth derives its strength
from an impoverished reading of Freud and an equally impoverished

12. Woifflin, Principles, p. ix.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 257

notion of how to read Freud. When applied to the Three Essays on the
Theory of Sexuality, this myth proceeds by showing that, for example,
Richard von Krafft-Ebing employed the idea of libido and Ellis the idea
of autoerotism, that Fliess made central use of the notion of bisexuality,
that Moll discovered infantile sexuality years before Freud, that Iwan
Bloch talked about erotogenic zones, and so on, ad infinitum. Since Freud
was fully aware of these writings, the story continues, how could he be
anything other than a usurper, with a kingdom made of stolen materials?
Both of these myths, mirror images of one another, depend on the
same kind of historiographical presumptions, unacknowledged, prejudicial,
and, in my opinion, misguided. Whether Freud did or did not discover
infantile sexuality, whatever his own changing assessment of his indebt-
edness to Fliess, whether he was the first, second, or third to use the
word Trieb when speaking of sexuality, all of these claims, both pro and
con, are radically inadequate if we want to understand his place in the
history of psychiatry. Both myths rely on an inappropriate invocation of
his name, both misplace the role that such invocation should have in
writing the history of psychoanalysis. Freud's biography, his personal
drama, and whom he read in what year are all topics that, of course, are
interesting and important. But they will not allow us accurately to ground
the question whether he was an originator of thought or merely a conserver,
and sometimes extender, of other people's ideas.
How we characterize Freud's place in the history of psychiatry ought
to depend not on who said what first, but on whether the structure of
concepts associated with Freud's writings continues, extends, diverges
from, or undermines the conceptual space of nineteenth-century psychiatry.
What we need, as I have indicated, is a history of the concepts used in
psychoanalysis, an account of their historical origins and transformations,
their rules of combination, and their employment in a mode of reasoning.13
This task presumes, first, that we can isolate the distinctive concepts of
nineteenth-century psychiatry, articulate their rules of combination, and
thereby discern their limits of the possible. We must then undertake the
very same enterprise for Freud's work, which, with sufficient detail, should
enable us to see more clearly whether Freud's conceptual space continues
or breaks with that of his predecessors. Although Freud may use much
the same terminology as many of the people we know he read, the
structure of concepts he employs, what I have been calling his conceptual
space, may nevertheless deviate to greater or lesser extent from theirs.
These methodological remarks, however brief and abstract, should
stand or fall depending on whether or not they enable us to produce a
philosophically enlightening, historically plausible account of the issues
at hand. If they do not directly guide us toward a more adequate reading

13. I have discussed some of these issues in relation to hysteria in my "Assault on

Freud," London Review of Books, 5-19 July 1984, pp. 9-11.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 258 Arnold I. Davidson Freud's Three Essays

of Freud, then their interest will remain but brief and abstract. So I now
want to turn to some of the historical questions that this archaeological
method dictates. This is the place to acknowledge the somewhat misleading
implications of the title of this essay. I will not even attempt anything
like a complete reading of the Three Essays. I want to focus exclusively
on the problems encountered in reading the first essay, "Sexual Aber-
rations." Given the structure of Freud's book, I will obviously have to
look at passages in the other essays as well, but I will discuss these only
when, and insofar as, they are relevant to excavating Freud's concep-
tualization of the sexual perversions. The scope of my task is limited
here by my desire to approximate to a comprehensive reading only of
this first essay. In order to do even that, I will have to start before Freud,
with the prevailing concept of sexual perversion in the literature of
nineteenth-century psychiatry. So let me try to demarcate the conceptual
space of which perversion was an element that dominated European
psychiatry at the time Freud was writing the ThreeEssays.14
During the second half of the nineteenth century there was a virtual
explosion of medical discussions about the sexual perversions, what Fou-
cault has called an incitement to discourse, an immense verbosity.'5 These
discussions saturated European and, eventually, American psychiatric
concerns, resulting in an epidemic of perversion that seemed to rival the
recent cholera outbreaks. Despite many differences between these lo-
quacious psychiatrists, differences both theoretical and clinical, all shared
the concept of perversion that underpinned these discussions-the per-
versions were a shared objectof psychiatric discourse about which there
were commonly recognized and fully standardized forms of reasoning.
The best way to begin to understand the nineteenth-century conceptual
space encircling perversion is to examine the notion of the sexual instinct,
for the conception of perversion underlying clinical thought was that of
a functional disease of this instinct. That is to say, the class of diseases
that affected the sexual instinct was precisely the sexual perversions. A
functional understanding of the instinct allowed one to isolate a set of
disorders or diseases that were disturbances of its special functions. Moreau
(de Tours), in a book that influenced the first edition of Krafft-Ebing's
PsychopathiaSexualis, argued that the clinical facts forced one to accept
as absolutely demonstrated the psychic existence of a sixth sense, which
he called the genital sense.16 Although the notion of a genital sense may
appear ludicrous, Moreau's characterization was adopted by subsequent
14. In what follows, I recount, with some additional quotations, parts of my article,
"Closing Up the Corpses." More detailed historical documentation for my claims can be
found in that paper.
15. See Foucault, The History of Sexuality, Vol. 1: An Introduction,trans. Robert Hurley
(New York, 1978).
16. See Paul Moreau (de Tours), Des Aberrationsdu sens genesique, 2d ed. (Paris, 1880),
p. 2.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 259

French clinicians, and his phrase "sens genital" was preserved, by Charcot
among others, as a translation of our "sexual instinct." The genital sense
is just the sexual instinct, masquerading in different words. Its charac-
terization as a sixth sense was a useful analogy. Just as one could become
blind, or have acute vision, or be able to discriminate only a part of the
color spectrum, and just as one might go deaf, or have abnormally sensitive
hearing, or be able to hear only certain pitches, so too this sixth sense
might be diminished, augmented, or perverted. What Moreau hoped to
demonstrate was that this genital sense had special functions distinct
from those served by other organs and that, just as with the other senses,
this sixth sense could be psychically disturbed without the proper working
of other mental functions, either affective or intellectual, being harmed.
A demonstration such as Moreau's was essential in isolating diseases of
sexuality as distinct morbid entities.
The OxfordEnglish Dictionary reports that the first modern medical
use in English of the concept of perversion occurred in 1842 in Dunglison's
Medical Lexicon: "Perversion,one of the four modifications of function in
disease: the three others being augmentation, diminution, and abolition."18
The notions of perversion and function are inextricably connected. Once
one offers a functional characterization of the sexual instinct, perversions
become a natural class of diseases; without this characterization there is
really no conceptual room for this kind of disease. It is clear, for instance,
that Krafft-Ebing understood the sexual instinct in a functional way. In
his Text-bookof Insanity Krafft-Ebing is unequivocal in his claim that life
presents two instincts, those of self-preservation and sexuality; he insists
that abnormal life presents no new instincts, although the instincts of
self-preservation and sexuality "may be lessened, increased, or manifested
with perversion."19 The sexual instinct was often compared with the
instinct of self-preservation, which manifested itself in appetite. In his
section on "Disturbances of the Instincts," Krafft-Ebing first discusses
the anomalies of the appetites, which he divides into three different
kinds. There are increases of the appetite ("hyperorexia"), lessening of
the appetite ("anorexia"), and perversions of the appetite, such as a "true
impulse to eat spiders, toads, worms, human blood, etc." (TI, p. 80; see
also pp. 77-81). Such a classification is exactly what one should expect
on a functional understanding of the instinct. Anomalies of the sexual
instinct are similarly classified as lessened or entirely wanting ("anesthesia"),
abnormally increased ("hyperesthesia"), and perversely expressed ("par-
esthesia"); in addition there is a fourth class of anomalies of the sexual
17. See ibid., p. 3.
18. OxfordEnglish Dictionary, s.v. "perversion."
19. Richard von Krafft-Ebing, Text-bookof InsanityBased on Clinical Observations,trans.
Charles Gilbert Chaddock (Philadelphia, 1904), p. 79; further references to this work,
abbreviated TI, will be included in the text. Krafft-Ebing considers abolition to be the
extreme case of diminution.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 260 Arnold I. Davidson Freud's Three Essays

instinct which consists in its manifestation outside of the period of an-

atomical and physiological processes in the reproductive organs ("par-
adoxia") (see TI, p. 81).20 In both his Text-bookof Insanity and Psychopathia
Sexualis, Krafft-Ebing further divides the perversions into sadism, ma-
sochism, fetishism, and contrary sexual instinct (see TI, pp. 83-86 and
PS, pp. 34-36).
To be able to determine precisely what phenomena are functional
disturbances or diseases of the sexual instinct, one must also, of course,
specify in what the normal or natural function of this instinct consists.
Without knowing the normal function of the instinct, everything and
nothing could count as a functional disturbance. There would be no
principled criterion to include or exclude any behavior from the disease
category of perversion. So one must first believe that there is a natural
function of the sexual instinct and then believe that this function is quite
determinate. We might think that questions as momentous as these
would have received extensive discussion during the heyday of perversion
in the nineteenth century. But, remarkably enough, no such discussion
appears. There is virtually unargued unanimity both on the fact that this
instinct does have a natural function and on what that function is. Krafft-
Ebing's view is representative here:

During the time of the maturation of physiological processes

in the reproductive glands, desires arise in the consciousness of
the individual, which have for their purpose the perpetuation of
the species (sexual instinct)....
With opportunity for the natural satisfaction of the sexual instinct,
every expression of it that does not correspond with the purpose
of nature-i.e., propagation--must be regarded as perverse. [PS,
pp. 16, 52-53]21

Should anyone doubt the representativeness of Krafft-Ebing's con-

ception, let me cite a long passage from Moll's Perversionsof the Sex Instinct
(1891), since Moll is considered by Frank Sulloway, among others, to be
a direct anticipator of Freud.22 Although Moll disputed many of Krafft-
Ebing's specific claims, the degree of unspoken agreement on the ap-
propriate conception of perversion is remarkable. Moll believed that
many of the theories of homosexuality with which he was familiar (ho-
mosexuality being the most clinically well documented of the sexual

20. This same classification is given in Krafft-Ebing, PsychopathiaSexualis, with Especial

Referenceto the AntipathicSexual Instinct:A Medico-ForensicStudy, trans. Franklin S. Klaf (New
York, 1965), p. 34; all further references to this work, abbreviated PS, will be included in
the text.
21. See also TI, p. 81.
22. See Frank J. Sulloway, Freud, Biologistof the Mind (New York, 1979), esp. chap. 8.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 261

perversions) did not sufficiently take into account the analogy between
the sexual instinct and other functions:

To understand the homosexual urge we should consider the

genital instinct not as a phenomenon apart from the other functions
but rather as a psychic function. The morbid modifications of the
genital instinct would appear to be less incomprehensible if we
were to admit that almost all the other functions whether physical
or psychic may be susceptible to similar modifications. The sexual
anomalies strike us as singular because most individuals who possess
the attributes of the masculine sex have a sexual urge for women.
But one must not be led astray by the frequence and regularity of
this phenomenon. From a teleological point of view, that is from
the point of view of the reproduction of the species, we consider
natural the urge that the normal man feels for woman. Still in
certain pathological conditions the organs do not meet the end
assigned to them. The teeth are meant to grind food yet there are
men who have no teeth or who have very few of them. The
function of the liver is to secrete bile which is diverted into the
intestine, and in certain disorders of the liver or of the bile ducts
the bile is not secreted and does not reach the intestine. The
function of hunger is to remind the organism that it needs food.
However, there are pathological states in which the sensation of
hunger is absent, although the stomach continues to function nor-
mally. It is the same with the absence of the sexual urge for
women in a man possessing normal genital organs. We can hardly
establish a connection between man's genital organs and his urge
for women except from a teleological point of view. Otherwise,
one does not see why men should be urged to have connections
with women since ejaculation of the sperm may be brought about
in quite other ways. It would be rather surprising to see the genital
instinct not presenting the same morbid anomalies as the other
functions.23

Like that of other late nineteenth-century psychiatrists, Moll's teleological

fapon de parler was mixed with, and presumably grounded in, evolutionary
considerations. But my concern is not with why Moll said what he did
but rather with exactly what he said. In this respect, his conception and
Krafft-Ebing's are quite literally interchangeable.
Nineteenth-century psychiatry silently adopted this conception of
the function of the sexual instinct. It was often taken as so natural as
not to need explicit statement24 since it was the only conception that

23. Albert Moll, Perversionsof the Sex Instinct:A Studyof Sexual Inversion, trans. Maurice
Popkin (Newark, N.J., 1931), pp. 171-72; my emphasis.
24. For some French examples of this understanding, see Maurice Paul Legrain, Des
Anomalies de l'instinct sexuel et en particulier des inversions du sense genital (Paris, 1896), and
Dr. Laupts (pseudonym of Georges Saint-Paul), L'Homosexualitiet les typeshomosexuels: Nouvelle
idition de perversion et perversiti sexuelles (Paris, 1910).

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
262 Arnold I. Davidson Freud's Three Essays

made sense of psychiatric practice. It is not at all obvious why sadism,

masochism, fetishism, and homosexuality should be treated as species of
the same disease, for they appear to have no essential features in common.
However, if one takes the natural function of the sexual instinct to be
propagation, and if one takes the corresponding natural, psychological
satisfaction of this instinct to consist in the satisfaction derived from
heterosexual, genital intercourse, then it becomes possible to see why
they were all classified together as perversions. Sadism, masochism, fet-
ishism, and homosexuality all exhibit the same kind of perverse expression
of the sexual instinct, the same basic kind of functional deviation, which
manifests itself in the fact that psychological satisfaction is obtained pri-
marily through activities disconnected from the natural function of the
instinct. As Moll succinctly states it, emphasizing the psychological con-
stituent of this natural function, "we ought to consider the absence of
heterosexual desires morbid even when the possibility of practicing normal
coition exists."25 This understanding of the instinct permits a unified
treatment of perversion, allowing one to place an apparently heterogeneous
group of phenomena under the same natural disease-kind. Had anyone
denied either that the sexual instinct has a natural function or that this
function is procreation, diseases of perversion, as they were actually
understood, would not have entered psychiatric nosology.
With this conceptual and historical background, we can place the
opening two paragraphs of Freud's first essay in proper perspective:

The fact of the existence of sexual needs in human beings

and animals is expressed in biology by the assumption of a 'sexual
instinct', on the analogy of the instinct of nutrition, that is of
hunger. Everyday language possesses no counterpart to the word
'hunger', but science makes use of the word 'libido' for that purpose.
Popular opinion has quite definite ideas about the nature and
characteristics of this sexual instinct. It is generally understood to
be absent in childhood, to set in at the time of puberty in connection
with the process of coming to maturity and to be revealed in the
manifestations of an irresistible attraction exercised -by one sex
upon the other; while its aim is supposed to be sexual union, or
at all events actions leading in that direction. We have every reason
to believe, however, that these views give a very false picture of
the true situation. If we look into them more closely we shall find
that they contain a number of errors, inaccuracies and hasty con-
clusions.26

25. Moll, Perversionsof the Sex Instinct, p. 180.

26. Sigmund Freud, Three Essays on the Theoryof Sexuality, The Standard Edition of the
CompletePsychologicalWorksof SigmundFreud, ed. and trans. James Strachey, 24 vols. (London,
1953-74), 7:135; further references to this work, abbreviated T, will be included in the
text.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 263

In describing popular opinion about the sexual instinct, Freud's use of

the analogy of hunger indicates, as it did throughout the nineteenth
century, the functional conception of this instinct. Moreover, just as we
should expect, the natural function of the sexual instinct is expressed by
an irresistible attraction of the sexes toward one another, an attraction
whose ultimate aim is sexual union. Freud's use of the phrase "popular
opinion" can easily mislead a reader to think that this conception of the
sexual instinct defines popular as opposed to learned opinion. But however
popular this opinion was, it was exactly the view of those psychiatrists,
listed in the first footnote of this first essay, from whom Freud says his
information has been derived." If the argument of Freud's first essay is
that these views "give a very false picture of the true situation," then we
can expect Freud's conclusions to place him in opposition to both popular
and, more importantly, medical opinion. The problem is how precisely
to characterize this opposition.
In the last paragraph of this preliminary section of the first essay,
Freud introduces what he calls "two technical terms." The sexual objectis
"the person from whom sexual attraction proceeds," while the sexual aim
is "the act towards which the instinct tends" (T, pp. 135-36). Freud's
motivation for introducing these terms is not merely, as he explicitly
states it, that scientific observation uncovers many deviations in respect
of both sexual object and sexual aim. More significantly, these are precisely
the two conceptually basic kinds of deviations we should expect of those
writers who subscribed to the popular conception of the sexual instinct.
Deviations with respect to sexual object are deviations from the natural
attraction exercised by one sex upon the other; deviations with respect
to sexual aim are deviations from the natural goal of sexual union. The
remainder of the first essay is structured around this distinction between
sexual object and sexual aim, and the central role of this distinction is
itself firmly dependent on the view of the sexual instinct that Freud will
argue is false. I emphasize this point because one must recognize that
Freud's opposition to the shared opinion concerning the sexual instinct
is an opposition from within, that his argument unfolds while taking this
shaied opinion as given. Freud's opposition, let me say in anticipation,
participates in the mentality that it criticizes. This decisive starting point,
Freud's immanent criticism, will show itself in his final formulations and
conclusions, specifically in their ambiguities and hesitations.
I want to proceed by reminding you of the general outlines of the
next two sections of the first essay, in many ways the core of this essay.
The next section discusses deviations in respect of the sexual object.
Under this category Freud includes the choice of children and animals

27. In my discussion, I shall leave aside Freud's comments about popular opinion
concerning the absence of infantile sexuality. The question of the relationship between
popular and learned opinion on this issue is too complex to take up here.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
264 Arnold I. Davidson Freud's Three Essays

as sexual objects, but his most detailed discussion is of inversion, the

deviation to which nineteenth-century psychiatristshad themselves devoted
the most attention. After describing different degrees of inversion, Freud
argues that inversion should not be regarded as an innate indication of
nervous degeneracy-an assessment which was widespread, even if not
universal, in the nineteenth century. The overturning of the theory of
degeneracy as the explanation of nervous disorders was of central im-
portance in the history of nineteenth- and early twentieth-century psy-
chiatry, and Freud played a role here, as did many others."28Indeed,
Freud insisted that the choice between claiming inversion to be innate
and claiming it to be acquired is a false one, since neither hypothesis by
itself gives an adequate explanation of the nature of inversion. Freud
immediately turns, in a section both complicated and problematic, to the
role of bisexuality in explaining inversion, and I shall not even attempt
to discuss this section now. Despite the recent attention that has been
given to the notion of bisexuality in the development of Freud's early
psychoanalytic thought, his remarks in this section become more and
more puzzling the more carefully they are studied.
Freud next describes the characteristics of the sexual object and
sexual aims of inverts and ends this whole section on deviations in respect
of the sexual object with an extraordinary conclusion, a conclusion more
innovative, even revolutionary, than I suspect he was able to recognize.

It has been brought to our notice that we have been in the habit
of regarding the connection between the sexual instinct and the
sexual object as more intimate than it in fact is. Experience of the
cases that are considered abnormal has shown us that in them the
sexual instinct and the sexual object are merely soldered together
-a fact which we have been in danger of overlooking in conse-
quence of the uniformity of the normal picture, where the object
appears to form part and parcel of the instinct. We are thus warned
to loosen the bond that exists in our thought between instinct and
object. It seems probable that the sexual instinct is in the first
instance independent of its object; nor is its origin likely to be due
to its object's attractions. [T, pp. 147-48]

In the nineteenth-century psychiatric theories that preceded Freud, both

a specific object and a specific aim formed part and parcel of the instinct.
The nature of the sexual instinct manifested itself, as I have said, in an
attraction to members of the opposite sex and in a desire for genital
intercourse with them. Thus, inversion was one unnatural functional
deviation of the sexual instinct, a deviation in which the natural object
of this instinct did not exert its proper attraction. By claiming, in effect,
that there is no natural object of the sexual instinct, that the sexual object
28. For a useful overview, see Sulloway, Freud, Biologist of the Mind.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 265

and sexual instinct are merely soldered together, Freud dealt a conceptually
devastating blow to the entire structure of nineteenth-century theories
of sexual psychopathology. In order to show that inversion was a real
functional deviation and not merely a statistical abnormality without
genuine pathological significance, one had to conceive of the "normal"
object of the instinct as part of the very content of the instinct itself. If
the object is not internal to the instinct, then there can be no instrinsic
clinico-pathological meaning to the fact that the instinct can become
attached to an inverted object. The distinction between normal and inverted
object will not then coincide with the division between the natural and
the unnatural, itself a division between the normal and the pathological.
Since the nature of the instinct, according to Freud, has no special bond
with any particular kind of object, we seem forced to conclude that the
supposed deviation of inversion is no more than a mere difference.
Indeed, Freud's very language is indicative of the force of this conclusion.
He says, "Experience of the cases that are consideredabnormal," thus
qualifying "abnormal" in a rhetorically revealing manner. 29 These cases
of inversion are consideredabnormal because of a certain conception of
the sexual instinct in which one kind of object is a natural part of the
instinct itself. Unhinged from this conception, these cases cannot be
considered pathological, cannot instantiate the concept of abnormality
employed by Krafft-Ebing, Mill, and others. I think that what we ought
to conclude, given the logic of Freud's argument and his radically new
conceptualization in this paragraph, is precisely that cases of inversion
can no longer be considered pathologically abnormal.
In light of these remarks, I think that we can conclude further that
Freud operates with a concept of the sexual instinct different from that
of his contemporaries, or, better yet, that he does not employ the concept
of the sexual instinct in his theory of sexuality. What is at issue here is
not Freud's choice of words. Commentators are forever remarking that
English-reading readers of Freud are led astray by the translation of Trieb
as "instinct," since Triebis better translated by "drive,"reserving "instinct"
for Instinkt.30However, since many of Freud's contemporaries, among
them, Krafft-Ebing, used Trieb, Freud's terminology did not constitute
a break with previously established terminology. It is not the introduction
of a new word that signals Freud's originality but rather the fact that
Sexualtriebis not the same concept as that of the sexual instinct. We can
see this, to reiterate my main point, by recognizing that Freud's conclusion
is explicitly and directly opposed to any conclusion that could be drawn

29. The German is "Die Erfahrung an den ffir abnorm gehaltenen Fallen lehrt
uns ... " (Freud, GesammelteSchriften [Vienna, 1924], 5:20).
30. For one recent example, see Bruno Bettelheim, Freud and Man's Soul (New York,
1982). I have criticized Bettelheim's claims in "On the Englishing of Freud," LondonReview
of Books, 3-16 Nov. 1983.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 266 Arnold I. Davidson Freud's Three Essays

by using the concept of the sexual instinct. The relationship between the
concepts of sexual instinct and sexual objectfound in nineteenth-century
texts, a rule of combination partially constitutive of the concept of the
sexual instinct, was completely undermined by Freud, and as a consequence
of this cutting away of old foundations, inversion could not be thought
of as an unnatural functional deviation of the sexual instinct. That Sexualtrieb
is not the same as sexual instinct is shown by the fact that the concept
of sexual instinct played a very specific role in a highly structured, role-
governed, conceptual space, a space within which psychiatric theories of
sexuality had operated since about 1870.
If Freud's conclusions are as radical as I have made them out to be,
if his conclusions really do overturn the conceptual structure of nineteenth-
century theories of sexual psychopathology, then we might well wonder
what prepared the way for these conclusions. I think that we can point
to an attitude that prepared the way for Freud, even though there is a
very large gap between this attitude and the new conclusions Freud drew.
Freud himself tells us in a footnote the source of his attitude, and the
fact that he mentions it only in passing should not lead us to underestimate
the depth of its significance: "The pathological approach to the study of
inversion has been displaced by the anthropological. The merit for bringing
about this change is due to Bloch" (T, p. 139 n. 2). In 1902-3, Iwan
Bloch published a two-volume book, Beitriigezur AetiologiederPsychopathia
Sexualis,which was central in establishing the inadequacy of the degeneracy
explanation of perversion. This work is exhaustive in cataloging the utter
pervasiveness of sexual aberrations, which, according to Bloch, have
appeared in all historical periods, all races, and all cultures. His attitude
toward these facts is surprising and, one might say, potentially revolu-
tionary, although his work lacks the conceptual rearticulation that was
a precondition of any radical conclusions.
In the introduction to the first volume of his work, Bloch announces
that he intends to show that "the purely medical view of the sexual
anomalies, which has been stated so well by Casper, von Krafft-Ebing,
A. Eulenburg, A. Moll, von Schrenck-Notzing, [and] Havelock Ellis, ...
[does not suffice] for a fundamental explanation of the phenomena in
this field," and he then opposes to the "clinico-pathological theory" of
the sexual aberrations his own "anthropologic-ethnologic concept of the
facts of so-called 'psychopathiasexualis.'"31 He claims that he will show
that "this general concept of the sexual anomalies as universal human,
ubiquitous phenomena makes it necessary to recognize as physiologic

31. Iwan Bloch, AnthropologicalStudies on the Strange Sexual Practises of All Races and
All Ages (New York, 1933), pp. 5, 6; all further references to this work, abbreviated AS,
will be included in the text. This is a translation of vol. 1 of Bloch's Beitriige zur Aetiologie
der PsychopathiaSexualis (Dresden, 1902-3). I believe that Sulloway does not adequately
see the role of Bloch's work in Freud's ThreeEssays.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 267

much that previously has been regarded as pathologic" (AS, p. 6). (Bloch
follows a standard nineteenth-century medical convention of often using
the contrast physiologic/pathologic instead of normal/pathologic.) Given
his "anthropologic" attitude, Bloch finds no difficulty in making statements
such as the following:

We find minor deviations from the norm of vita sexualis quite

general. There are few persons who have not somewhere touched
the narrow boundary between normal and pathological indulgence.
[AS, pp. 165-66]
There can be no doubt that a normal individual can accustom
himself to the various sexual aberrations so that these come to be
"perversions," which deviations appear in the same form in sound
persons as well as in diseased.32

The narrow boundary between the normal and the pathological, the fact
that sexual aberrations are a universal human phenomenon, was the
primary evidence for Bloch that nervous degeneracy was not an accurate
explanatory or diagnostic rubric under which perversion could be placed.
And his attitude about sexual aberrations was distinctively different-
less unequivocal, less psychiatric-from that of the authors with whom
he was engaged in debate. But this attitude toward inversion and toward
the other perversions, however different it was from the purely medical
view and however unstable it often seemed, never led Bloch to throw
into doubt the concept of the sexual instinct that made possible the
classification of these phenomena as deviant (and that therefore required
some alternative explanation of their status as pathological). Freud might
have taken Bloch's anthropological observations, in conjunction with the
other clinical evidence he cites, to show merely that inversion, if only in
a rudimentary or shadow form, was much more widespread than many
psychiatrists had believed. This claim would still have allowed a conceptual
priority to the "uniformity of the normal picture." Freud might then
have advanced with this conclusion to blur the boundary between the
normal and the pathological even further, thereby providing yet one
more attack on the idea that a distinctive class of degenerate individuals
suffered from inversion. But rather than drawing this limited, though
significant, conclusion, Freud went to the core of the matter and decisively
replaced the concept of the sexual instinct with that of a sexual drive "in
the first instance independent of its object." This was a conceptual in-
novation worthy of the name of genius-although genius need not be

32. Bloch, Anthropologicaland Ethnological Studies in the Strangest Sex Acts in Modes of
Love of All Races Illustrated (New York, 1935), pt. 2, p. 4. This is a translation of vol. 2 of
Bloch's Beitrige zur Aetiologieder PsychopathiaSexualis.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 268 Arnold I. Davidson Freud's Three Essays

conscious of itself as such, as we shall see if we turn to the next section

of the first essay, entitled "Deviations in Respect of the Sexual Aim."
Freud defines perversions as

sexual activities which either (a) extend, in an anatomical sense,

beyond the regions of the body that are designed for sexual union,
or (b) linger over the intermediate relations to the sexual object
which should normally be traversed rapidly on the path towards
the final sexual aim. [T, p. 150]

This definition of perversion is explained by the fact that, as Freud puts

it,

The normal sexual aim is regarded as being the union of the

genitals in the act known as copulation, which leads to a release
of the sexual tension and a temporary extinction of the sexual
instinct-a satisfaction analogous to the sating of hunger. [T, p.
149]

So since the normal sexual aim is copulation and the anatomical region
appropriate to this aim is the genitals, two main kinds of perverse deviations
in respect of the sexual aim are possible. Under the heading of anatomical
extensions, Freud discusses oral-genital sexual activities, anal-genital sexual
activities, kissing, and fetishism, recognizing that the latter might also
have been classified as a deviation in respect of the sexual object. Under
fixations of preliminary sexual aims, Freud discusses touching and looking,
and sadism and masochism. Since I cannot discuss each of these examples
here, let me focus on a few representative ones.
Sexual use of the mucous membrane of the lips and mouths of two
persons, otherwise known as kissing, is, strictly speaking, a perversion,
since, as Freud says, "the parts of the body involved do not form part
of the sexual apparatus but constitute the entrance to the digestive tract"
(T, p. 150). But when the mucous membranes of the lips of two persons
come together, we are not in the habit of classifying the anatomical
extension or resulting aim as a perversion. Indeed, Freud notes that we
hold kissing in "high sexual esteem" and he goes on to claim that kissing
is "the point of contact with what is normal" (T, pp. 150-51). So given
kissing's technical status as a perversion and our refusal to classify it as
such, those who claim oral-genital and anal-genital activities are perversions
must be "giving way to an unmistakable feeling of disgust, which protects
them from accepting sexual aims of this kind" (T, p. 151). Freud im-
mediately adds that "the limits of such disgust are, however, often purely
conventional"(T, p. 151; my emphasis).33
33. The German is "Die Grenze dieses Ekels ist aber haiufig rein konventionell" (Ge-
sammelteSchriften, 5:25).

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 269

In discussing the kind of looking that has a sexual tinge to it, Freud
acknowledges that most normal people linger to some extent over this
form of pleasure, so he gives a number of conditions under which this
pleasure in looking, usually called scopophilia, becomes a perversion.
The most important of these conditions is when "instead of being preparatory
to the normal sexual aim, it [pleasure in looking] supplants it." And
Freud goes on to remark that "the force which opposes scopophilia, but
which may be overridden by it (in a manner parallel to what we have
previously seen in the case of disgust), is shame" (T, p. 157). Similarly,
when the aggressive component of the sexual instinct "has usurped the
leading position" so that sexual satisfaction "is entirely conditional on
the humiliation and maltreatment of the object," we are faced with the
perversion of sadism (T, p. 158). Shame and disgust are the two "most
prominent" forces that keep the sexual instinct "within the limits that
are regarded as normal" (T, p. 162), but Freud also lists pain, horror,
and aesthetic and moral ideals as other normalizing restraints.34
In the conclusion to the third section, after mentioning the importance
of such restraints, Freud insists that since these perversions admit of
analysis, that is, since they "can be taken to pieces," they must be of a
"composite nature":

This gives us a hint that perhaps the sexual instinct itself may be
no simple thing, but put together from components which have
come apart again in the perversions. If this is so, the clinical ob-
servation of these abnormalities will have drawn our attention to
amalgamations which have been lost to view in the uniform be-
haviour of normal people. [T, p. 162]

This passage introduces the concept of component instincts, a notion

that will assume its full role in Freud's conception of sexuality only when
he later connects it with the further idea of pregenital libidinal organi-
zations. Some of these component instincts are specified by their source
in an erotogenic zone, a zone of the body capable of sexual excita-
tion-examples are oral and anal component instincts (see esp. T, pp.
167-69).35 Other component instincts are specified by their aim, inde-
pendent of any erotogenic zone-examples are the component instincts
of scopophilia and cruelty (see T, pp. 191-93).
In the 1905 edition, the first edition, of the ThreeEssays,the component
instincts are thought to function anarchically until the primacy of the
genital zone is established. In his 1913 article, "The Disposition to Ob-

34. Pain and horror are mentioned, respectively, on pp. 159 and 161; aesthetic and
moral ideals are listed on p. 177.
35. Freud uses the term "erotogenic instincts" once on p. 193. The German is "erogenen
Trieben" (GesammelteSchriften, 5:68).

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 270 Arnold I. Davidson Freud's Three Essays

sessional Neurosis," Freud introduces the concept of pregenital orga-

nization, recognizing that there is an anal organization of the libido. In
the 1915 edition of the ThreeEssays,Freud recognizes an oral organization
of the libido, and, finally, in his 1923 article, "The Infantile Genital
Organization," he describes a phallic organization of it. All of these pre-
genital organizations are theoretically incorporated into the 1924 edition
of the ThreeEssaysin the section of the second essay entitled "The Phases
of Development of the Sexual Organization." Though we should not
undervalue the importance of the notion of pregenital organizations of
the libido, it was Freud's articulation of the concept of component instincts
that constituted another one of his major conceptual innovations (without
which the notion of pregenital organizations would have made no sense).
The concept of component instincts made it possible for Freud to say,
to quote from his concluding summary of the Three Essays, that "the
sexual instinct itself must be something put together from various factors,
and that in the perversions it falls apart, as it were, into its components"
(T, p. 231).
The idea that the sexual instinct is made up of components, that it
so combines a multiplicity of erotogenic zones and aims, is a further
radical break with the nineteenth-century medical conceptualization of
the sexual instinct. Freud's argument, his structure of concepts, leads to
the claim that neither the erotogenic zone of the genitals nor the aim of
copulation bear any privileged connection to the sexual instinct. The
"normal" aim of the sexual instinct, genital intercourse, is not part of
the content of the instinct; or, to put it another way, recurring to Freud's
previous conclusions about the sexual object, the sexual instinct and
sexual aim are merely soldered together. If there is no natural aim to
the sexual instinct, no given aim internal to this instinct, then deviations
from the aim of genital intercourse appear to lose their status as genuine
perversions, as pathological aberrations whose status outstrips any supposed
statistical abnormality. If the structure of Freud's argument here, in
conjunction with his argument in the previous section, is to show that
neither a specific aim nor a specific object has any constitutive bond with
the sexual instinct, and if the previously shared concept of the sexual
instinct is thus effectively dismantled, then it is difficult to see how any
conceptual foothold could remain for the concept of unnatural functional
deviations of this instinct. In the case of both sexual aim and sexual
object, it is only the apparent uniformity of normal behavior that directs
us to think otherwise. But this apparently well-entrenched uniformity
actually masks the operations of the sexual instinct, operations which,
when conceptualizedbyFreud, show us that the idea of the natural function
of the instinct has no basis whatsoever.36 We ought to conclude from

36. Freud uses this notion of uniformity in two crucial passages. See T, pp. 148 and
162. For the German uses of "gleichfirmig," see GesammelteSchriften,5:21 and 36.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 271

what Freud says here that there are no true perversions. The conceptual
space within which the concept of perversion functions and has a stable
role has been thoroughly displaced-and displaced in a way that requires
a new set of concepts for understanding sexuality and a new mode of
reasoning about it.
This is the place, obviously enough, at which someone might retreat
to Freud's discussion of disgust and shame, claiming that these reactions
can provide an independent criterion for classifying certain sexual phe-
nomena as perversions. But however we are to understand the role of
these reactions, it is absolutely clear from Freud's remarks that even
though he believes that some of these phenomena are such that "we
cannot avoid pronouncing them 'pathological' " (T, p. 161), these pro-
nouncements, our shame, disgust, and moral and aesthetic ideals, cannot
provide an appropriate criterion of perversion. The tone of his example
that we may be disgusted at the idea of using someone else's toothbrush,
which follows his claim that these reactions are "often purely conventional,"
permits no other intelligible reading (see T, pp. 151-52). And, of course,
it almost goes without saying that such a last-ditch attempt to save the
concept of perversion would be at odds with the structure of the Three
Essays as a whole and would make most of its content completely beside
the point.
Even if Freud's conclusions in effect overturn the conceptual apparatus
of perversion, it is well known that he did not embrace these conclusions
unambiguously or unhesitatingly. The language of Freud's discussion
sometimes reads as if he is unaware of the conceptual innovations he
has wrought, as if nineteenth-century theories of sexual psychopathology
can remain secure in their conceptual underpinnings. In the section of
the third essay entitled "The Primacy of the Genital Zones and Fore-
Pleasure," Freud can be found referring to "the appropriate [geeignete]
stimulation of an erotogenic zone (the genital zone itself, in the glans
penis) by the appropriate [geeignetste]object (the mucous membrane of
the vagina)" (T, p. 161). But the whole point of Freud's argument in the
first essay has been that no particular zone of the body and no particular
object is specially suitable to, or qualified for, stimulation. The notion of
appropriateness has lost all of its conceptual plausibility because the
concept of the sexual drive is detached from that of a natural object and
aim. And whatever transformations of puberty Freud may want to sketch
in the third essay, these transformations cannot reinstate the old concept
of the sexual instinct, the concept which gives a place to the notions of
appropriate object and stimulation. The uneasy attitude of Freud's dis-
cussion is highlighted again in the next section of the third essay, "Dangers
of Fore-Pleasure" (Gefahrender Vorlust),where Freud talks of the "normal
sexual aim" as being "endangered by the mechanism in which fore-
pleasure is involved." The danger in question consists in the fact that
one may become fixated on the pleasure of the preparatory acts of the

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 272 Arnold I. Davidson Freud's Three Essays

sexual process, and these acts may then take the place of the normal
sexual aim. Such displacement, Freud tells us, is the "mechanism of many
perversions" (T, p. 211). But again this is dangerous, in the sense of
pathogenic, only if it exhibits some kind of unnatural deviation from the
normal aim of the sexual instinct;37and given Freud's previous argument,
he cannot maintain this latter claim. He dimly indicates his awareness
of this fact when he introduces the distinction between forepleasure and
endpleasure. The first is the "kind of pleasure due to the excitation of
erotogenic zones," while the second is the kind of pleasure "due to discharge
of the sexual substances" (T, p. 210). Since no conceptual space remains
for the distinction between, as it were, natural and unnatural pleasure,
or normal and abnormal pleasure, Freud is left merely with the difference
between two kinds or degrees of pleasure, shorn of any pathological
implications. This is not the only place where Freud hesitates to believe
what he has said.
Let me focus on just a few more passages that will reinforce still
further the complexity of this problem. The first passage comes from
Freud's discussion of "The Perversions in General" in the first essay, and
I want to notice especially the attitude embodied in this passage.

If a perversion, instead of appearing merely alongside the normal

sexual aim and object, and only when circumstances are unfavorable
to them, and favorable to it-if, instead of this, it ousts them com-
pletely and takes their place in all circumstances--if, in short, a
perversion has the characteristics of exclusiveness and fixation-
then we shall usually be justified in regarding it as a pathological
symptom. [T, p. 161]38

The phrase "we shall usually be justified in regarding it as a pathological

symptom" shows that we find here the attitude of, let us say, a pathologist,
apparently the very same kind of medical attitude found in Krafft-Ebing,
Moll, and their fellow-travelers. The rhetoric of this passage emphasizes
the characteristics of exclusiveness and fixation, as though perversions
are harmless until they become exclusive and fixed, as though this is the
real criterion of pathology. But it is clear enough that the tendency
toward exclusiveness and fixation on genital activity is not only non-
pathological but a central component of Freud's conception of normal,
healthy sexuality. It is only when sexual activity is divorced from the
normal sexual aim and object that it can become a perversion and so
qualify for pathological status. The moment the concept of perversion
is introduced, with its corresponding concepts of normal aim and object,
we are prepared for the attitude that treats perversion as pathological.

37. Freud uses the word pathogenic (pathogene) in this context.

38. The German passage appears in GesammelteSchriften,5:34-35.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 273

The crucial move, the moment that makes the medicalizing attitude
inevitable, is not the explicit listing of characteristics that make perversions
pathological, but the use of the concept in the first place.
Freud's problemization of perversion is shown by the fact that in
the first essay the words "normal," "pathological," and "perversion" often
appear in scare quotes or qualified by a phrase like "what we would
describe as"; as we move through the other essays, the scare quotes
become scarcer and the qualifications less emphatic, until, in the concluding
summary of the book, these terms appear simpliciter.Indeed, in the para-
graph preceding the one I have just quoted, "pathological" does appear
in scare quotes, but by the end of the next paragraph the fact that we
regard perversions as pathological is something which is unqualified and
justified.
Although I could discuss the only later, detailed passage of the Three
Essays in which Freud returns to inversion, as opposed to perversion, a
passage where the same questions of attitude arise (see T, pp. 229-30),
it will be more useful, I think, to concentrate on a remarkable passage
in the concluding summary to the book. The passage appears during
Freud's discussion of the various factors that can interfere with the de-
velopment of a normal sexual instinct.

Writers on the subject, for instance, have asserted that the necessary
precondition of a whole number of perverse fixations lies in an
innate weakness of the sexual instinct. In this form the view seems
to me untenable. It makes sense, however, if what is meant is a
constitutional weakness of one particular factor in the sexual instinct,
namely the genital zone-a zone which takes over the function of
combining the separate sexual activities for the purposes of repro-
duction. For if the genital zone is weak, this combination, which is
required to take place at puberty, is bound to fail, and the strongest
of the other components of sexuality will continue its activity as a
perversion. [T, p. 237]39

We find in the writings of both Moll and Ellis the claim that an innate
weakness of the sexual instinct is often responsible for the failure of
normal heterosexual development, with perversion being the manifest
result. In fact many writers before Freud used the terms "sexual instinct"
and "genital instinct" interchangeably, as if the latter were simply a more
precise name for the former. This identification was not in the least bit
arbitrary, since the sexual instinct was conceived of as psychically expressing
itself in an attraction for members of the opposite sex, with genital in-
tercourse as the ultimate aim of this attraction. And since these features
specified the natural operation of the sexual instinct, the common use

39. The corresponding German passage appears in GesammelteSchriften,5:113.

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 274 Arnold I. Davidson Freud's Three Essays

of the alternative phrase "genital instinct" was not conceptually misplaced.

But once Freud reconceived the sexual instinct as having no natural
operation, once any specific aim and object of the drive were thought to
be merely soldered to it, the genital zone lost the conceptual primacy
that was a precondition of its principled identification with the instinct
itself. When the sexual instinct is conceived of as an amalgamation of
components, the genital zone being one such component but without
any natural privilege, then to single out this zone as Freud does in this
passage, to claim that a constitutional weakness of it is responsible for
the perversion, is to maintain an attitude toward genitality that is no
longer appropriate. Freud in effect reintroduces, behind his own back,
an identification that he has shown to be untenable. His claim that these
writers are mistaken in asserting that an innate weakness of the sexual
instinct is responsible for perversion, but that their assertions would make
sense "if what is meant is a constitutional weakness of one particular
factor in the sexual instinct, namely the genital zone," is astonishing,
since this is, of course, exactly what they meant, and had to mean, given
their conception of the sexual instinct. It is Freud who cannot mean to
say that the absence of this particular factor, the primacy of the genital
zone, is a condition of perversion. The last sentence of this paragraph
reads, "For if the genital zone is weak, this combination, which is required
to take place at puberty, is bound to fail, and the strongest of the other
components of sexuality will continue its activity as a perversion." But
the system of concepts Freud has been working with in the first essay
requires a slightly different conclusion, one whose subtle modulation
from Freud's actual conclusion must be emphasized. The appropriate
formulation of the conclusion should read, "For if the genital zone is
weak, this combination, which often takes place at puberty, will fail, and
the strongest of the other components of sexuality will continue its activity."
The differences between these two formulations represent what I have
been calling Freud's attitude.
Although it is a central feature of commentary that it can go on
indefinitely, I want to stop and return in conclusion to issues of histo-
riographical orientation. I should perhaps first describe the sense in
which I think that my reading of the ThreeEssays is a history of psycho-
analysis without names. It is not, of course, that I have refused to invoke
Freud's name, or the names of Bloch, Moll, and others. It is rather that
I have treated their names as, so to speak, place-holders for certain sets
of concepts and the way these concepts fit together to constitute a conceptual
space. We see that the concept of perversion in nineteenth-century psy-
chiatry was part of a conceptual space in which, for example, the concept
of the sexual instinct combined, according to definite rules, with those
of sexual object, sexual aim, unnatural functional deviation, and so on.
It was this conceptual space, itself a nineteenth-century invention, that
made it possible for psychiatrists to make the statements about perversion

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 275

that so dominated the period. These statements were thus set within a
shared discursive practice. Freud's Three Essays on the Theoryof Sexuality
provided the resources to overturn this conceptual space by fundamentally
altering the rules of combination for concepts such as sexual instinct,
sexual object, sexual aim-with the consequence that these shared concepts,
among others, were destroyed. The conclusion forced upon us is that
perversion is no longer a legitimate concept, that the conceptual pre-
conditions for its employment no longer exist in Freud's text. So that if
Freud, despite himself, said that such and such phenomena were per-
versions, he could not have meant what Krafft-Ebing, or Moll, or Charcot
had meant. We will not be able to arrive at this conclusion if we focus
simply on whom Freud read, on who before him used what words in
which contexts. We must turn rather to the issue of conceptual articulation,
to reconstructing nineteenth-century and Freudian concepts of sexuality
and determining their points of contact and dissociation. Many writers
before Freud possessed bits and pieces of his terminology and exhibited
an inchoate, unself-possessed grappling with the problems brought to
light by the Three Essays. But it was Freud who ascended to the level of
concepts, who systematically and lucidly thought what had previously
remained in a kind of precognitive blockage, who turned what had been,
at most, a creeping anxiety into a conceptual mutation.
Yet we know that Freud continued to use the idea of perversion, as
if he failed to grasp the real import of his own work. And so now we
must directly invoke Freud's name, and wonder about the accessibility
of his achievement to Freud himself. I have said that what prepared the
way for Freud's achievement was a certain attitude, one that was most
clearly appropriated from the writings of Bloch but that could no doubt
be found in other authors as well. This attitude allowed a sort of opening
so that perversions might no longer be treated as unambiguously patho-
logical. This notion of attitude, which I cannot elaborate theoretically
here, is one component of the concept of mentalit6,a concept that has
been put to extraordinarily fertile use by recent historians, especially in
France.40 A mentality includes, among other constituents, a set of mental
habits or automatisms that characterize the collective understanding and
representations of a population. Bloch's BeitriigezurAetiologiederPsychopathia
Sexualis exhibit the tremors of a shift in mentality in which what was
taken for granted begins to become dislodged. But this displacement

40. Useful introductions to the history of mentalities can be found in Jacques LeGoff,
"Les mentalites: Une histoire ambigue," in Faire de l'histoire:Nouveaux objets,ed. LeGoff
and Pierre Nora (Paris, 1974), and Roger Chartier, "Intellectual History or Sociocultural
History? The French Trajectories," in Modern European IntellectualHistory:Reappraisalsand
New Perspectives, ed. Dominick LaCapra and Steven L. Kaplan (Ithaca, N.Y., 1982), pp.
13-46. The notion of mentality is invoked for the history of science in some of the essays
in Occult and Scientoic Mentalities in the Renaissance, ed. Brian Vickers (London, 1984).

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 276 ArnoldI. Davidson Freud'sThree Essays

could only be partial, and one was alwaysin danger of falling back into
the old mentality, preciselybecause there was no conceptualbackingfor
this change of attitude. That Bloch never pushed this attitude into con-
ceptualinnovationstartsto explainwhy his attitudewasinherentlyunstable,
why his work often reads like a kind of unsteady bridge between the old
and new mentalities, a bridge always ready to collapse because still in
need of completion.
Freud's genius consisted not simply in appropriating this attitude
but in seizing and exploiting it. He provided a conceptual foundation
for the newly emerging mentalitythat made it possible, once and for all,
for us to change decisivelyour old mental habits. So why, one wonders,
did Freud himself not so change his own mental habits, why did he
exhibit an attitudevirtuallyno less ambiguousand unstablethan Bloch's?
Any answer to this question is bound to be complicated,so in lieu of an
answer let me provide the structurefor what I take this answer to consist
in. Automatismsof attitude have a durability,a slow temporality,which
does not match the sometimes rapid change of conceptual mutation.
Mental habits have a tendency toward inertia, and these habits resist
change that, in retrospect, seems conceptuallyrequired. Such resistance
can take place not only in a scientificcommunitybut even in the individual
who is most responsible for the conceptual innovation. Freud was a
product of the old mentality that regarded perversionsas pathological,
a mentality whose first real signs of disintegration can be found at the
beginning of the twentieth century. Freud's ThreeEssaysought to have
stabilizedthe new mentality,speeding up its entrenchmentby providing
it with a conceptual authorization. But given the divergent temporality
of the emergence of new concepts and the formationof new mentalities,
it is no surprise that Freud's mental habits never quite caught up with
his conceptual articulations.The attitudes that comprise a mentalityare
sufficientlyimpervious to recognition, so much like naturaldispositions,
that many decades may intervene before habit and concept are aligned.
However,withoutsome appropriateconceptualbackdrop,it is veryunlikely
that a new scientific mentality can genuinely displace an old one, since
concepts, especiallyin science, are one fundamentalhabit-formingforce,
one force which, even if over a long span of time, makes possiblea stable
set of firm mental habits. Although social, cultural, institutional, and
psychologicalfactors may all delay the definitive formation of these new
habits,it is conceptual innovationof the kind Freud produced that marks
one place of genius. But we must rememberthat genius too has its habits,
its inert tendencies, that create a form of friction between what could be
said and what is said, so that genius is alwaysahead even of itself.
The hesitationsand ambiguitiesof Freud'sThreeEssayson the Theory
of Sexuality are not the result of some deconstructive indeterminacy or
undecidability of the text but are rather the consequence of the dynamics
of fundamental change. Mentality and concept are two different aspects

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
 Critical Inquiry Winter 1987 277

of systems of thought, and we should not expect them to be coherently

connected all at once, as if forms of experience could be dissolved and
reconstituted overnight. Sidney Morgenbesser is said to have asked the
following question on an exam at Columbia University: "Some people
argue that Freud and Marx went too far. How far would you go?" Whether
Freud went too far or not far enough, this is exactly the right range of
question. How far can you go? How far will you go?

This content downloaded from 128.123.044.023 on September 15, 2016 09:04:39 AM

All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).