Memory and Representation

Author(s): Norman Malcolm

Source: Noûs, Vol. 4, No. 1 (Feb., 1970), pp. 59-70
Published by: Wiley
Stable URL: http://www.jstor.org/stable/2214294 .
Accessed: 12/05/2014 04:45

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].

Wiley is collaborating with JSTOR to digitize, preserve and extend access to Noûs.

http://www.jstor.org

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
 FIFTH SYMPOSIUM

Memoryand Representation1

NORMAN MALCOLM
CORNELL UNIVERSITY

Commentators:STANLEY MUNSAT UNIVERSITY OF

CALIFORNIA
AT IRVINE
GARETH MArrHEWS UNIVERSITY OF
MASSACHUSETTS

1. In philosophers' accounts of what occurs in remembering, one

item that is nearly always specified as being present is an image,
picture, or copy of what is remembered. Aristotle says that when
we remember "there is in us something like a picture or impression"
(De Memoria et Reminiscentia, 450 b). James Mill says that
"Nothing is remembered but through its Idea" (Analysis of the
Phenomena of the Human Mind, I, 320). When you remember a
previous sensation your idea of it is not itself a sensation, but still
it is "more like the sensation, than anything else can be; so like,
that I call it a copy, an image of the sensation; sometimes, a repre-
sentation, or trace of the sensation" (ibid., pp. 51-52). William
James thought that a necessary ingredient of memory is "the revival
in the mind of an image or copy of the original event" (Principles
of Psychology, I, 649). And Russell says: "Memory demands an
image"(The Analysisof Mind,p. 186).
1 I am greatly indebted to Bruce Goldberg for helping me in my
thinking on the topics of this paper. I owe a debt of a different kind to the
Center for Advanced Study in the Behavioral Sciences, which provided me
with the opportunityto devote the year 1968-69 to the study of Memory.
59

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
 AO^~~~~~~~~~~~~~~~~~~~~~~~~~
U0 NOUS

Beyond the general agreement of memory theorists on the

necessity for an image, some of them have held that "the present
mental occurrencein remembering"must contain additional com-
ponents. Certain feelings must accompany or characterize the
image. One can deduce the nature of the feelings from the prob-
lems that have to be solved. For example, if I have an image is it
an image of somethingI expect or of somethingI remember?Should
I refer the image to the past or the future? Why should I give it
one referencerather than another?Now if there were such a thing
as "a feeling of pastness"it would, apparently,solve the problem.
For if that feeling should accompany an image then I shall refer
the image to the past; if not, not.
Suppose I have now made my image refer to the past. Why
should I believe it is an image of somethingthat actually occurred?
Why should I "trust"the image? If there were such a thing as "a
feeling of familiarity"that can characterizean image, this problem,
too, would apparently be solved. For I shall "trust"a familiar
image-that is, I shall believe it to be a true portrayalof reality.
One can think of still other feelings that would seem to be required
for memory.For example, even if I refer an image to the past and
also am confident of its accuracy, why should I believe it is an
image of something that I witnessed or experienced?James says:
Memoryrequiresmore than mere datingof a fact in the past. It
must be dated in my past. In other words, I must think that I
directly experienced its occurrence (op. cit., p. 650).
This problem is solved by assuming a characteristicof an image
called "warmthand intimacy".
When everything required for remembering is added to-
gether, it turns out that your rememberingyour having witnessed
some past incident comprises a "presentmental content' of many
components.As James says, the memory of something is "a very
complexrepresentation"(ibid., p. 651).

2. These rich accounts of what goes on in rememberingreceive a

jolting set-back when we abstain from considerations of what
must occurin memory,but instead describesome ordinaryexamples
of memory.Supposethat a clerkin a grocerystore is busily occupied
with customers,hurriedlyfetching articlesfrom the shelves, rapidly
adding figures, replying to questions, making change. In the midst
of this activity the manager calls to him, "Wheredid you put the

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
MEMORY AND REPRESENTATION 61

package for Mrs. Casey?". Without a pause in his activity the clerk
immediately replies, "Under the meat counter".
Does anyone think it likely that the clerk had an image char-
acterized by a feeling of "familiarity," or of "pastness", or of
"warmth and intimacy'? In the quiet of his study a philosopher
will undoubtedly note delicate shades of experience, as he tries
to detect the felt difference between remembering and, say, expect-
ing or believing. But it is a mistake to assume that those nuances
of consciousness have anything to do with the concept of memory.

3. We are entitled to laugh a little, but not too much, at these

fantastic conceptions of what goes on in remembering. Our merri-
ment should be restrained, because all of us have the inclination to
feel that something important and hard to describe occurs when
we remember. Part of our feeling could be put like this: When
we remember something we must represent it in some way. Recently
this reputed requirement has been affirmed by Martin and
Deutscher. They say:
Somebody may have observed an event, but unless he is recounting
it to himself, telling others or in some other way representing it,
then, roughly speaking, he is not remembering that event....
nobody actually remembers anything until he comes to the point
of representing in some way what he has observed or experienced.
("Remembering,"Philosophical Review, LXXV (1966), p. 172)
Surprisingly, these authors believe that they were the first to per-
ceive the necessity for a representation in remembering. They de-
clare that "no philosophical writing on memory has so much as
recognized the problem" (ibid.). The truth is just the opposite. The
assumption of the necessity of a representation has been a staple of
the philosophy of memory since Plato.
If we unfold our "intuitive" notion of this representation of a
past fact or occurrence, I think we shall be inclined to say that
it is a complex thing, that it has a structure, and that this structure
is identical with the structure of what is remembered. Harrod
expresses this idea very nicely. His example is one of remembering
a street with houses, gardens and traffic. Not everything that was
seen is remembered.
The materialof the houses may be rememberedbut not the number
of storeys, and so on. The memory event must then be as complex
as what is said to be remembered, which is a selection of a total
experience, for how otherwise can we say how much of it is re-

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
62 NOeT

membered?There must be as many terms and relations as there are

in what is remembered. And these must be properly disposed in
relation to one another. It will not do if the paving stones are on
the roofs and the slates in the street. In fact, the memory must be
in some sense a copy of the thing remembered. ("Memory",Mind,
51 (1942), 49.)
In other words, for every remembered feature of the rememberec
thing, there must be a corresponding feature in the present memory
event. This reminds us of Wittgenstein's Tractatus:
In a propositionthere must be exactly as much to be distinguished
as in the situationthat it represents.The two must possess the same
logical (mathematical) multiplicity. (4.04).
In the Tractatus a proposition is conceived of as a picture of reality.
In a picture of reality the elements of the picture stand for things
in reality, and the fact that the picture-elements are related in a
certain way represents that the things are related in the same way
(ibid., 2.15). The fact that the picture-elements are related in a
certain way is the structure of the picture. When the things in reality
have the same structure, the picture is a true picture of reality.
In the thinking of philosophers about memory "the memory
event' has been conceived of as a picture in the sense of the
Tractatus. Harrod says that plainly.
A Tractatus picture would be, in respect to its structure, a
copy of the reality it depicts. This is implied by the view that in a
true picture the connection of picture-elements is the same as the
connection of reality-elements. Aside from differences in "quality"
between picture-elements and reality-elements, a true picture would
be a facsimile of the reality it depicts.

4. The surprising insistence of memory theorists on the need for

images in memory, probably derives from the assumption that "the
memory event" must "constitute or contain a copy of what is sup-
posed to be remembered" (Harrod, op. cit., p. 50). If we assume
that when we remember something there must be a copy of it in
the mind, and if we then try to specify what kind of mental thing
could have the character of being a copy of the remembered fact, a
mental image would seem to be the only candidate. In The Analysis
of Mind Russell asks:
Why do we believe that images are, sometimes or always, approxi-
mately or exactly, copies of sensations? What sort of evidence is
there? (p. 158)

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
MEMORY AND REPRESENTATION 63

The only answer he provides is to invent a property of images (the

feeling of familiarity) which, when it is present, makes us confi-
dent that the image is an "accurate copy". That the memory event
must contain a copy, and that the copy is an image, goes without
question. It is as if Russell, Price, James, Rume and the rest thought
like this: There certainly has to be in our remembering a copy of
the past event we remember, and what could this copy be if not
an image?

5. What motivates the assumption that in remembering there has

to be a copy-something that is identical in structure with what is
remembered? One answer is that the copy is believed to provide
guidance.
Locke assumed that t-he "ideas" of memory serve as guides.
Speaking of "the faculty of retention" he says:
This faculty of laying up and retaining the ideas that are brought
into the mind, several other animals seem to have to a great degree,
as well as man. For, to pass by other instances, birds learning of
tunes, and the endeavors one may observe in them to hit the notes
right, put it past doubt with me, that they have perception, and
retain ideas in their memories, and use them for patterms.For it
seems to me impossible that they should endeavor to conform their
voices to notes (as it is plain they do) of which they had no ideas.
(Essay, II, x, 10)
The birds are guided in their singing by mental patterns of the
notes: otherwise, how could they even try to get it right?
Suppose a person is shown a green piece of cloth and told to
fetch a bolt of cloth of that color from the next room. He walks
into the next room, looks around at the bolts of various colors there,
and picks out the right one without any hesitation. Locke would
think that he must have had in his mind an idea of the color he was
shown. The idea, presumably, would be an image. It served as a
pattern to guide him.
This is an illustration of how the normal phenomena of
memory can strike us as mysterious. Between the man's seeing the
green cloth and his confidently selecting the green bolt, there is a
gap that is puzzling to us until we fill it with some sort of mental
mechanism, such as an image (accompanied, perhaps, by feelings
of familiarity, pastness, and warmth and intimacy).
The most damaging criticism of this insistence on a mental
guidance mechanism is the following: How does the man know

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
64 NOitS

that if he has an image of green color he should pick out a bolt of

green color? Why not a yellow bolt, or a blue one, or one that is
both yellow and blue? Doesn't he really need a second mental
intermediary to guide his use of the first one? And a third one to
help him with the second? And so on? In order to avoid the threat
of an infinite regress of guidance systems (which would make it
impossible for the man to act at all) we must assume that at some
stage he can take a step without being guided. To paraphrase the
Wittgenstein of the Investigations: Guidance soon gives out, and
then one acts without guidance (see, e.g., ?? 211, 217, 326, 433).
Yet if the man can do this at some stage then why not right at the
beginning? Why cannot he have been able to know what the right
color was, without guidance-that is, without making use of a
representation? The notion that there must be a mental inter-
mediary that shows us the past color, tune or scene, is a self-
defeating movement of thought. Its purpose is to remove a mystery:
but it merely transfers a mystery from one place to another.

6. The insistence that in remembering there must be a copy or

picture of what is remembered has a deeper source, however, than
the feeling that we need to have guidance. The deeper notion is
that remembering is a form of tlhinking, and in all thinking there
must be something (variously called an "idea", a "picture", a "prop-
osition") that is the content of the thinking. Stout says the follow-
ing about the role of the image in memory:
The image is the specifying content of the thought, determining it
as the idea of a specific impression.Apart from the image or some-
thing discharging an equivalent function, this thought would be
empty. . . (Studies in Philosophyand Psychology,p. 367).
The image fills the thought with content.
Russell had the same notion. In his paper "On Propositions:
What They Are and How They Mean" he imagines that a child is
relating an occurrence from memory. Russell makes this comment:
It is clear that, in so far as the child is genuinely remembering,he
has a picture of the past occurrence, and his words are chosen so
as to describe the picture; and in so far as the hearer is genuinely
apprehendingwhat is said, the hearer is acquiring a picture more
or less like that of the child. (Logic and Knowledge, ed. R. C.
Marsh, p. 502)
Russell was sure of this because he held the view that thinking is
picturing. A thought has to contain a picture. He says: "To 'think'

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
MEMORY AND REPRESENTATION 65

of the meaning of a word is to call up images of what it means"

(ibid., p. 300). Remembering and apprehending are forms of think-
ing-that is, forms of picturing.
Thus we see that Russell's assumption that memory requires
an image was an off-shoot of a picture-theory of thinking. The same
may have been true of Aristotle. In De Anima he says that "the soul
never thinks without an image" (431a). In De Memoria he refers
to this doctrine when he asserts that there must be an image in
memory (450a).
I have offered the examples of Stout, Aristotle and Russell in
support of the suggestion that the nearly universal insistence by
memory theorists on the necessity for an image in memory, is
primarily a consequence of an equally widespread view about the
nature of thinking. This view, which receives its most perfect ex-
pression in the Tractatus, is that a thought-picture is a fact com-
posed of elements that are correlated with elements of reality.
When the reality-elements are structured in the same way as are
the picture-elements, the thought-picture is true.

7. Let us try to break out of this maze of preconceptions by

describing a few examples of memory. Suppose you called on a
friend to take him for a drive and when the two of you left his
house you saw him lock the front door. Subsequently you remem-
bered his locking it. A prominent question in t-he philosophy of
memory has always been, "Just exactly what happens when a person
remembers something?". Now a variety of different things could
have happened. Consider the following cases:
1) Your friend asks you, "Did I lock the front door?". You
promptly reply, "Yes", without any effort to remember or any
imagery. Your attention is concentrated on driving your car over
an icy road.
2) He says nervously, "I wonder if I locked the front door".
You reply, sternly and emphatically, "Youlocked the front door!";
and you think to yourself, "Are we going to have another outbreak
of neurosis about locking the door?"
3) He asks, "Do you recall whether I locked the door?". You
say, "Now let me think", and you feel a twinge of anxiety. You try
to picture his exit from the house. An image of him bending over
the doorknob comes to you. You say, with a feeling of relief and
satisfaction, "You locked it".
4) There is a police investigation, and a police officer says

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
 A
66 NOUS

to you, "I want you to act out for me the movementsyour friend
made in leaving the house".You comply by walking to your friend's
study and picking up his briefcase, then walking to the hall closet
and taking his coat. You go out the front door and lock it. You
walk down the porch steps to the garage. While carryingout this
performanceyou feel nervousand apprehensive.You think to your-
self, "Willhe be cleared?"In your mind'seye you see your friend's
pale face behind bars.
5) Another example of a police investigation.But instead of
asking you to imitate your friend'smovements,as in the preceding
case, the police officer says: "To speed things up I am going to
describethe sequence of your friend'smovementsas I believe they
occurred.If I mention something you do not remember,just raise
your hand".The officerthen relates a sequence of events, including
the locking of the door. You listen closely, with a feeling of tension,
watching the officer'sface. You do not see your friend'smovements
in your mind's eye; nor do you say to yourself, after each event
related by the officer,"Yes,that occurred".When the officersays,
"He stopped on the porch to put on his coat,"you raise your hand.
All of these were cases in which you rememberedyour friend's
locking the door. In only one of them was there any relevant
imagery;and in that one the image was only of his bending over
the doorknob,not of his locking the door. In which of these cases
did you represent his locking the door, or produce a representation
of it? This cannot be said of the one in which you merely said
"Yes",while your attentionwas fixed on the icy road. Nor can it be
said of the one in which you did not raise your hand when the
officersaid, "Thenhe locked the door";yet the officercould rightly
state in his report that you rememberedyour friend's locking the
door. The only one of thfesefive cases in which there would be any
naturalnessin saying that there was a representationof his locking
the door would be the one where you acted out his movementsfor
t-hebenefit of the policeman.
The claim of Martinand Deutscher that "nobodyactually re-
members anything until he comes to the point of representingin
some way what he has observed or experienced",is far from true.
The words "represent"and "representation",as we actually use
them, are not bound up in that way with the word "remember". The
ordinaryconcept of rememberingdoes not demand a representation
any more than it demands an image or a feeling of pastness. A
philosopher'snotion that there has to be a representationhas the

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
MEMORY AND REPRESENTATION 67

same source, or sources, as does the notion that there has to be

an image. Indeed, the words "image", "picture", "idea", "copy",
"representation", all tend to be equivalent terms in the philosophy
of memory. "Representation" is, however, a more satisfactory term
than "image"for a memory theorist, since its greater vagueness gives
his thesis better protection against refutation.
Whichever term is used, the basic notion is that in order to
remember a past occurrence your thought of it must, as it were,
duplicate it. The "idear' state of affairs would be to have the past
event over again, just as it was. This being impossible, the next
best thing is to have the structure of the past event duplicated in
the remembering. This is the way in which there is a "copy" of
the past event in memory. The remembering must contain a "struc-
tural analogue" of what is remembered.

8. Let us turn our attention to this doctrine that "the memory

event" (the "act" or "experience" of remembering) must contain
the structure of what is remembered. In the first place, what is "the
memory event"? I described several examples of your remembering
your friend's locking the door. Different things happened in these
examples. In one case you said, "Yes"; in another you said, "You
locked it"; in another you locked the door yourself; in still another
you did not raise your hand. In one case you tried to remember;
but not in the rest. In one case you had a mental image of your
friend bending over the knob; but not in the others. In some of the
cases you had various feelings and thoughts; but none of these were
present in all of the cases.
Which of the various utterances, thoughts, feelings, or actions
that I mentioned was the remembering, or the memory event? Or
did I leave out something? If so, what? Or is it something unsay-
able?
It is true that neither your saying, "Yes", nor your saying,
"You locked the door", nor your having an image, nor your acting
out the locking of the door, was the remembering. But neither was
the remembering some other event I failed to mention. It is a bad
mistake to think that in those examples of memory, there was, in
addition to the mentioned utterances, actions, thoughts and feelings,
some other occurrence which was the remembering itself. If we
avoid this mistake we shall have no inclination to think that the
remembering ("the memory event") contains the structure of
what is remembered. We shall no longer feel that we know what

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
68 NOIS

it is that is supposed to correspond in structure to the remembered

occurrence.
In the second place, is there such a thing as the structure of
a fact or event? What was the structure of your friend's locking
the door? Did your saying the word, "Yes",have the same structure;
or a different one? What is the structure of the ceiling of this room;
and does my hand have the same structure as the ceiling?
Obviously, we don't know what we are talking about. If some-
thing has a structure then it is composed of elements, parts or units.
What are the parts of my hand? Different decisions are possible.
We could decide that fingers, knuckles, palm and back, are the
parts. Or that the parts are skin, bone, flesh and fingernails. It is
meaningless to ask whether A and B have the same structure until
we are prepared to specify the elements of A and of B; and are
ready to specify a definite relation as holding between the elements
of A, and a definite relation as holding between the elements of B;
and have a method of correlating the two groups of elements and
relations. The specifications and correlations could be made differ-
ently. According to one set of specifications my hand and the
ceiling will have the same structure; according to another, not.
What would be the elements, units or parts of your friend's
locking the door? Probably none of us are ready to say. Given time
for thought, and ingenuity, we might come up with some decisions
-different ones for different purposes. One decision might be
better than another, in relation to a given purpose. No decision
would be, in some absolute sense, the correct one. Therefore, there
can be no such thing as the structure, in an absolute sense, of the
locking of the door-or of any other event or fact. It would be
possible to think of a way of specifying a structure for a remem-
bered event, and a structure for a memory response, such that the
two had the same structure. But it would be just as possible to
obtain the opposite result. The insistence by memory theorists on
a structural correspondence between what is remembered and the
memory-representation, assumes that events and facts have struc-
ture in an absolute sense. Otherwise, this demand would be point-
less, since structure can be assigned in such a way that any two
things have the same structure.
Thus the ancient philosophical conception that the event or
state of remembering must contain a reprdsentation which has the
same structure as what is remembered, commits several profound
errors, which I summarize as follows:

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
MEMORY AND 1REPRESENTATION 69

1) Nothing that occurs when a person remembers something,

can be picked out as the remembering ("the memory event").
2) There is no requirement of an image, copy, picture, pattern
or representation in remembering.
3) It is meaningless to speak of the structure of anything, in
an absolute sense.

9. I began this paper with the intention of devoting the second

half of it to the idea of a memory trace; but I have already exceeded
the limit. Let me briefly indicate what I believe to be the thinking
that lies behind the current assumption of traces by psychologists,
neuropsychologists and philosophers. It goes like this: You have an
experience. The experience itself cannot survive the passage of
time; but the structure of the experience can be stored in the
brain. The neural mechanism that performs this function is the
trace. As long as the trace persists you have a "dispositional" mem-
ory of the experience. The experience is "coded" into the trace.
If one knew the code (the law of projection) one could "read off"
an experience from its trace. When the trace is subjected to the
right sort of stimulus an "occurrent" memory results, which con-
tains an active representation of the original experience.
Most of these notions are exhibited in the following quotation
from a recent, comprehensive survey of neurological speculations
about memory:
There seem to be four fundamentalfunctions that a memory mech-
anism must perforrn: (1) the configurationof external and internal
stimuli impinging upon an organism, which constitute an experi-
ence, must somehow be coded into a neural representation; (2)
the neural representation of that experience (coded information
about the set of stimuli) must be stored; (3) it must be possible to
gain access to the coded information in order to retrieve specific
experiences from storage; and (4) the retrieved data must again
be decoded into neural activity, which somehow recreates the sensa-
tions and qualities of the original experience and thus constitutes
a "memory". (E. R. John, Mechanisms of Memory (New York,
1967), pp. 2-3.)
The conception is basically the same as Aristotle's. The structure
of an experience is stamped into the body, where it persists in a
dormant state, occasionally emerging as an active representation.
This is how past experience is preserved in the human organism.
What is here called a "neural representation" might be thought
of as a propositional-picture, in the sense of the Tractatus, en-

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions
70 NOUS

fleshed, so to speak, in a neural mechanism. In its neural form it

is a potential memory-thought. One and the same structure is pres-
ent in the original experience, in the neural representation, and in
the memory event.
From what we have said about "structure", "representation",
and "the memory event", it should be evident that this conception
of a memory trace, despite its aura of scientific respectibility, is
brimming with confusion.

ABSTRACTS OF COMMENTS

A. by STANLEY MUNSAT

I am in agreement with what Malcolm has said in criticism of the

role of images or some such "representation"of the past in theories of
memory. A slightly different perspective on the motivations behind the
requirement for (something like) an image in memory is this-we
recognize that the behavior (such as a nod of the head) by which one
manifests his rememberingsomething contains nothing on the fact of it
to connect it with the remembereditem. By bringing in a memory image
as the medium for the memory content and seeing the behavior as some-
how issuing from the image this connection appears to be established.
The crucial issue is not merely (as one might suppose from
Malcolm's discussion) that the image theory is factually mistaken in
large numbers of instances. Rather, in being forced to admit that the
theory is factually mistaken we are then in a position to recognize that
the image theory had all along only relocated the real problem. For
what has a piece of behavior like nodding one's head or pointing got to
do with the image one has in his head of the remembered item? Once
that question is answered, we will find that we no longer need the image
to connect the memory behavior with the remembered item. For the
answer to that question will enable us to establish the connection
directly.
Then what does make e.g. nodding one's head a "case" of "re-
membering"that such and such happened? The answer (as we now
recognize) is to be found in the "situation"or the "context"in which
the behavior takes place, though that involves many different sorts of
things.

This content downloaded from 109.135.15.50 on Mon, 12 May 2014 04:45:08 AM

All use subject to JSTOR Terms and Conditions