Australasian Journal of Philosophy

Vol. 81, No. 2, pp. 229245; June 2003

SCOTTISH COMMONSENSE ABOUT MEMORY
A DEFENCE OF THOMAS REIDS DIRECT KNOWLEDGE ACCOUNT
Andy Hamilton
Reid rejects the image theorythe representative or indirect realist positionthat
memory-judgements are inferred from or otherwise justified by a present image or
introspectible state. He also rejects the trace theory, which regards memories as
essentially traces in the brain. In contrast he argues for a direct knowledge account
in which personal memory yields unmediated knowledge of the past. He asserts the
reliability of memory, not in currently fashionable terms as a reliable belief-forming
process, but more elusively as a principle of Commonsense. There remains a
contemporary consensus against Reids position. I argue that Reids critique is
essentially sound, and that the consensus is mistaken; personal memory judgements
are spontaneous and non-inferential in the same way as perceptual judgements. But
I question Reids account of the connection between personal memory and personal
identity. My primary concern is rationally reconstructive rather than scholarly, and
downplays recent interpretations of Reids faculty psychology as a precursor of
functionalism and other scientific philosophies of mind.
Thomas Reids defence of memory as direct knowledge of the past is an original though
rather neglected aspect of his Philosophy of Commonsense. In the course of his critique of
the Ideal System or Theory of Ideas, he argues against two accounts of memory which
treat it as indirect access to the past. The representative or indirect realist position regards
remembering as essentially a present experience from which one infers about past events.
It says that memory-judgements are inferred from or otherwise justified by a present
image, experience or introspectible state known with certainty. I will term this view the
image theory. The theory was attractive to empiricists because of its starting-point in
present experience; contemporary psychologists, in this respect still surprisingly indebted
to empiricism, also frequently assume it. There is a powerful impulse towards the image
theory; for instance, we often call the memory-images themselves, and the feelings which
accompany them, our memories. It is not a soft target.
1
Reid also rejects the trace theory,
which regards memories as essentially traces in the brainan apparently immovable
assumption of contemporary scientific psychology. The two theories co-exist uneasily in
Locke, and they remain most influential.
What is distinctive and original about Reids discussion is that, against the Theory of
Ideas, he argues for a direct knowledge account in which personal memory yields
229
1
It is defended by Pollock [1987] and Campbell [1994]; though Gareth Evans seems to disavow it, it
still guides his thinking in Varieties of Reference [1982], and that of John McDowell at least in
Reductionism [1997]. The issue is discussed further in Hamilton [1998b].
unmediated knowledge of the past. He asserts the reliability of memory, not in currently
fashionable terms as a reliable belief-forming process, but more elusively as a principle of
Commonsense. My conclusion is that Reids critique is essentially sound, and that the
contemporary consensus is wrong. The dialectic of the article runs as follows. Reids
arguments against the image theory, which concerns the justification of individual
memory-judgements, are claimed to be decisive (Section I). It may still be argued that the
justification of individual judgements at least presupposes the justification of memory per
se as reliable, but Reid rightly rejects this idea also (Section II). For him, memory is one
of the distinct and original faculties of the mind, whose conceptual and doxastic
operations are unique to the power of the faculty. It is an unaccountable feature of our
constitution, in two respects: memory is always accompanied by belief, and when distinct,
it gives rise to true belief. I argue that the first of these is a conceptual truth, but that the
second may be regarded as a principle of Commonsense and thus in Reids sense
unaccountable. His position further involves the rejection of the trace theory (Section III).
Finally, there is one important respect in which Reids treatment of memory goes
wrongin relation to personal identity, where his subscription to substance dualism
proves damaging (Section IV). Throughout the article, my primary concern is rationally
reconstructive rather than scholarly. The aim is to expand on what I see as the truth in
Reids account, downplaying recent interpretations of his faculty psychology as a
precursor of functionalism and scientific philosophies of mind.
2
It is essential at the outset to emphasize the distinction between personal and merely
factual memory of ones past. When Reid discusses distinct remembering, and says that
the evidence we have of our own identity . . . is grounded on memory, and gives absolute
certainty, he is referring to personal or autobiographical memory. This is memory of
events which I have witnessed or experienced, where my judgements are not made purely
as a result of receiving indications after the eventwhether someone elses testimony, or
something I have readthat they happened. I can be reminded or prompted, but to
describe what happens in this way, is to imply that my judgement is not made simply as a
result of being told that something happened. If it was made as a result of being told, this
would be purely factual memory of my pastwhat is also termed habitual memory of
information, that which has been learned and not forgotten [Anscombe 1981].
3
Although I
may on occasion be unsure whether my knowledge of the past comprises personal or
merely factual memory, the distinction remains a vital one.
Personal memory-judgements often take the form of ordinary past-tense claims not
preceded by the words I remember . . .; for instance, I fell down the steps. But for such
a claim to be considered a personal memory-judgement, the subject must be willing to
present it in the form I remember falling down the steps . . . or I remember so-and-so
falling down the steps . . .. The hallmark shows that personal remembering is not straight-
forwardly a propositional attitude, and failure to recognize it has resulted in inadequate
accounts of both personal memory and personal identity. I remember that . . . is normally
elliptical for I remember reading that . . ., I remember being told that . . . and so on
230 Scottish Commonsense About Memory
2
For instance Keith Lehrers claim that Reid is a modern cognitive scientist as well as an
eighteenth-century metaphysician [1989: 7].
3
The distinction between personal and factual memory, and other aspects of the present treatment,
are developed further in Hamilton. forthcoming.
the varieties of factual memory. The continuous-verb hallmark implies the possibility of a
spontaneous manifestation or willed rehearsal of the remembered events in the form of
memory images or experience, often accompanied by feelings of nostalgia, remorse,
pleasure or so on. For many writers, such images imply an image theory; I will argue in
contrast that they may be the form which my personal memory-knowledge takes.
I. Memory [Has] Things Which Are Past . . . For Its Objects
A. The Direct Knowledge Account
Reid argued, against the Ideal System, that memory is a kind of mental act which cannot
be analysed simply in terms of the having of ideas or impressions; it involves judgement
that directly concerns a past object. Thus he claims in the Inquiry that memory appears
to me to have things that are past, and not present ideas, for its objects . . . I beg leave to
think, with the vulgar, that, when I remember the smell of the tuberose, that very sensation
which I had yesterday, and which has now no more any existence, is the immediate object
of my memory [Reid 1997: II.iii, 28]. In the Essays he argues that It is by memory that
we have an immediate knowledge of the pastjust as perception offers immediate
knowledge of the presentand not indirect knowledge inferred from some present
experience [2002: III.i, 253].
4
Reid holds that the remembrance . . . is a particular act of
the mind which now exists, and of which we are conscious, one which it would be absurd
to confoundas do proponents of the Ideal Systemwith the thing remembered [ibid].
The immediate objects of memory are past things and not present ideasintentional
objects, objects of thought or judgement, not objects of awareness. Just as fictional entities
such as unicorns can be objects of thought, so can past objects. Hence memory as direct
knowledge of the past must involve judgement. Reids recognition of this explains the
great superiority of his account over empiricist rivals, and he argues that distinct
memory is always accompanied by a belief in the past existence of that which we remem-
ber [III.i, 340]. Talk of objects of memory may remain unclear, however. Another way
of saying that memory has past things for its objectsthat the smell of the tuberose is the
immediate object of my memoryis that my memory-belief is not the result of a process
of inference, nor otherwise justified, from a present mental image or introspectible state.
Direct realismsupplanting the earlier natural realismhas become accepted as
the label for Reids view, a label applied also to the positions of philosophers as diverse
as Samuel Alexander, G. F. Stout, and Wittgenstein. But a better description is direct
knowledge account. Reid gives short shrift to scepticism, but realism is not necessarily
a good label for its antidote. The term direct knowledge also helps to counter a miscon-
ception which sees Reid as postulating direct awareness of the past. This misconception
renders Reids vital analogy between memory and perception quite incredible, since it
regards him as claiming an awareness of objects that no longer existwhat one might call
the telescope into the past view of memory. If there were such a faculty, memory-
experience would be akin to hallucination, the seeing of the past as contemporaneous. But
when someone is engrossed in their memories, and exclaims I can see him now, as if it
Andy Hamilton 231
4
All otherwise unqualified Reid references are to Reid 2002, the Edinburgh University Press edition
of Essays On The Intellectual Powers, indicating book and chapter as well as page number.
were yesterday, they do at least think that it is yesterday!
5
Sir William Hamilton
comments in one of the many intrusive footnotes in his Reid edition that An immediate
knowledge of a past thing is a contradiction [Reid 1967: 339 (EIP III.i)]. But it is
immediate awareness that is the contradiction.
I referred to the analogy between memory and perception as vital to Reids account. It
may be developed in the following way. There is a human capacity spontaneously and
reliably to report both past and present events. If asked to describe the view from the
window of this room, I will mostly do so spontaneously, though partly via inference. (That
square over there must be Red Lion Square, because I remember passing a street-sign with
that name on my way here, and so on.) In normal circumstances it would be absurd for
someone else to question whether this really is what I am seeingmy judgements in this
respect are authoritative, and indeed constitute avowals in the Wittgensteinian sense. It is
acceptable for me to misidentify the person walking on the other side of the road, but
failure to describe a tree that blocks out much of the view indicates a perceptual or
cognitive dysfunction. The reliability of perception is paralleled by that of memory. A
spontaneous past-tense claim is one not made simply as a result of receiving indications
after the event, whether statements or evidence, that the thing happened, and not based on
an inference. The connection with perception is not just an analogy, however. Though
with memory there is no possibility of re-focusing in response to questionsat least not in
the sense of looking again at the event in questionthe object of personal remembering
is, I would argue, identical to that of the earlier perception.
The elements of the sequence of impression, sensation, and belief involved in the
process of perception are easily confounded, Reid believes; in particular, the Ideal System
disastrously confuses the conceptual and doxastic operations of the faculty for the
impressions and sensations that occasion them. But a recurrent issue of interpretation has
been the extent to which Reid really does repudiate the Ideal System. In the case of
memory, some commentators have suggested that implicit in his account is a role for the
awareness, if not of past objects, then of present memory-impressions. In his treatment of
perception Reid has a place for sensations, and a common criticism from Thomas Brown
onwards has been that these just are ideasthat Reids diatribe against the Ideal System
results only in a new nomenclature. I would argue that for Reid, sensations must be given
an adverbial analysis as mental acts. They represent no object, and cannot form the basis
for an inference.
6
However, Reid does not specify the operations of memory in the same
detail as those of perception, and does not refer to sensations as such. He has been
interpreted as saying that I have an experience or perceive an object by the senses, then
later experience a present state of mind which naturally, by an original instinct, suggests
or prompts a memory belief about that object. (This is Skorupskis view [1993: 7].) But
this interpretation may be incorrect, for as we will now see, Reid holds that remembering
consists at least partly in a memory belief.
232 Scottish Commonsense About Memory
5
Compare Wittgenstein: Remembering: a seeing into the past. Dreaming might be called that, when
it presents the past to us. But not remembering; for, even if it showed scenes with hallucinatory
clarity, still it takes remembering to tell us that this is past [Wittgenstein 1980: 592].
6
Reids adverbial account is presented for instance in the Inquiry [1997: VI.xx, 1823]. Chappell
argues that Reid should have dispensed with talk of sensations altogether [1989: 53, 61]; his
interpretation is the obverse of those which treat Reid as an indirect realist in disguise. J.-C. Smith
in the introduction [2000] convincingly defends a middle way.
B. The Act of Remembrance and Seeming to Remember
The operations of perception are easily distinguishd from all other acts of the mind, as
are those of memory, Reid believes. But what is the present act of mind that I
experience, which prompts the memory belief, and in what sense is it easily
distinguishd from other mental acts? Some writers have argued that the act of mind is
not remembering, but seeming to remember. So when Reid refers to what I distinctly
remember he really means what I distinctly seem (or seem distinctly) to remember. (See
for instance Skorupski [1993: 7].) This interpretation is prompted by two considerations.
First, on the assumption that remember is a factive, the conditional If I distinctly
remember o-ing, then I o-ed becomes analytic, and so could hardly be a principle of
Commonsense as Reid intends. Second, that unless the act of mind is one of seeming to
remember it could not be easily distinguishd from other mental acts.
The second consideration is mistaken, I will argue. The first has some justification, but
its conclusion is mistaken. It is not clear that If I distinctly remember o-ing, then I o-ed
is analytic. This is because although remember is a factive, distinctly remembers o-ing
seems not to be. I distinctly remember locking the front door, but I cant have done is
paradoxicalan expression of puzzlementbut not unintelligible. I distinctly remember
. . . functions both as a claim of certainty and as a justification that may, on rare
occasions, be defeated.
7
The puzzlement expressed by I distinctly remember . . . but I
cant have done arises in the unusual situation where a distinct memory seems to be
mistaken. It would be wrong to formalize this puzzlement by means of the expression
I seem distinctly to remember, let alone by I distinctly seem to remember, which is not
intelligible at all. On Reids account it is remembering which is basic, not seeming to
remember. Here we see the close connection between Reids account of the act of
remembering, and his assertion of its reliability, of which more below.
The second motive for claiming that seems to remember is the act of mind in
question follows directly from the image theory. Unabashed proponents of the theory such
as Pollock argue that only putative memories are introspectively distinguishable.
Acknowledging that many philosophers have denied that there is a state of seeming to
remember, he argues that memory must provide us with beliefs about what we seem
to remember and then we infer the truth of what are ordinarily regarded as memory
beliefs from these apparent memories. Pollock maintains that we can distinguish, by
introspection, beliefs that we have on the basis of memory from those that we have on the
basis of perception, or for no reason at all. Or rather, we can distinguish putative
memory-beliefs from other kinds: I do not mean that we can tell introspectively whether
we are correctly remembering what we take ourselves to be remembering [Pollock 1987:
512]. The idea seems to be that there is an occurrent state of mindwhat Reid calls
remembrancewhich is introspectibly tagged as such.
Pollocks qualification is clearly necessaryimage theorists must avoid the absurd
claim that memory yields infallible knowledge of the past. They must rather say:
Although I cannot be sure whether the event happened or not, I can at least be sure that if
Andy Hamilton 233
7
The way in which personal memory may be overridden by other sources of knowledge is
misunderstood by the narrative conception of memory popular among psychologists, an issue
discussed in Hamilton [1998b].
it happened, I know that it did on the basis of memory rather than because I received
indications after the event. Clearly this is not the case, however. I can certainly wonder
I o-ed, but Im not sure whether I genuinely remember o-ing or was told that I had o-ed.
Only with regard to recent events does this doubt make no sense. If someone says, about
having a headache earlier the same day, I dont know if I remember this, or if I was told
it, extraordinary circumstances would be needed by way of explanation, but in other
cases it may be perfectly intelligible. Thus there is no introspectible state of seeming to
remember. There are two kinds of apparent memory: Im not sure whether I o-ed
maybe nothing of the sort happenedand I o-ed, but Im not sure whether I genuinely
remember o-ing or was told that I had o-ed. Does the existence of the second doubt
undermine my earlier claim that I seem to remember . . . usually functions as a tentative
assertion about the past, and never as a report of an introspectible state? I do not think
so. Certainly I am unlikely to be confident that I know that something happened if I am
not sure whether I witnessed it, or learned of it in some other way. I seem to remember
o-ingthe second kind of caseis a claim that is tentative about the source of evidence,
and again not a report of an introspectible state. I seem to remember . . . is always
motivated by a doubtand the kind of doubt must be specifiable.
If Reids claim that remembering is easily distinguishd from other mental acts
amounts to the claim that I know for sure that I at least seem to remember, then it would
be mistaken in the same way as the image theory. Thomas Browns criticism mentioned
earlier, that Reids sensations just are ideas, would then receive some vindication. That
criticism does indeed seem valid in the context of Reids discussion of personal identity,
as I will argue below. But in the present discussion he seems to be making a different
claim: that personal remembering is primitive and irreducible, and so is easily distin-
guished in our philosophical reflections from other mental operations. This interpretation
is supported by his contrast in the same paragraph with the operations of perception,
which are hard to disjoin in our conceptionsthat is, in our philosophical reflections.
C. Distinct Memory and Belief
To understand the state of mind which Reid postulates, it is necessary to look more
closely at his discussion of distinct memory and belief. Reid is explicit that memory
involves judgement and thus belief. In the Inquiry he comments that the sensation
compels my belief of the present existence of the smell [of the rose], and memory my
belief of its past existence [Reid 1997: II.iii, 29]. In the Essays he writes that in mature
years, and in a sound state of mind, every man feels that he must believe what he
distinctly remembers, though he can give no other reason of his belief, but that he remem-
bers the thing distinctly; whereas, when he merely imagines a thing ever so distinctly, he
has no belief of it upon that account. Reid then contrasts distinct memory with cases
where memory is less distinct and determinate, and where [the subject] is ready to allow
it may have failed him [III.i, 254]. He gives no explicit gloss of distinctly remembers,
but when discussing perception, he says that we see things more distinctly when they are
close, in good light, and so on [II.v, 967]. Certainly it makes sense to say that a distinct
original perception is required for distinct memory. Thus to allay neurotic anxieties when
going on holiday, the present writer makes sure to get a distinct perception of locking the
front door, to fix it in his memory that he has done so. Perhaps a distinct original
234 Scottish Commonsense About Memory
perception is all that is required for distinct remembering. I distinctly saw it, and what is
more I now distinctly remember what I distinctly saw seems pleonastic. It is true that one
can distinctly see something, and in the course of time forget about it. But nothing is
added to the force of I distinctly saw by saying and I distinctly remember.
Reid comes close to asserting a conceptual connection between rememberingor
distinct rememberingand belief. But it might be argued that to establish this connection,
he would have done better to avoid talking of an act of mind of which one is conscious.
He could then have glossed distinctly remembers as: is inclined confidently and sponta-
neously to assert a past-tense claim, in that grammatical form distinctive of personal
memory, and is able to give supporting detail in that form. Spontaneous means: not
made simply as a result of receiving indications after the event, whether statements or
evidence, that the thing happened, and not based on an inference. On this account, the
connection between distinctly remembers o-ing and believes that she o-edbetween
distinct memory and beliefis clearly a conceptual one. This is less clear in Reids
discussion, when he says that it is an unaccountable feature of our constitution, that
distinct memory is always accompanied by belief. (Or perhaps one should say almost
always; as noted earlier, the co-existence of distinct memory with grounds for doubt
causes puzzlement in the subject.)
The suggested gloss on distinctly remembers, and the rejection of an introspectible
state of seeming to remember, should not be equated with a rejection of memory-
experience. Although Reid himself has little to say on the matter, direct knowledge
theorists have generally been hostile to memory-experience. The view of Stout,
Wittgenstein, Anscombe, and others seems to be that ones images are themselves
assessed for correctness by memory, not vice versa.
8
These writers fail to recognize that
when people refer to their memories of someone, or of some event, a memory-image
may be the form that their knowledge of the past takesa formulation favoured by
Mounce [1994]. This claim is not inconsistent with the Reidian idea that judgement is
central to memory, and offers a middle way between an image theory and Rylean
reduction of image-talk. (A more recent representative of the latter is Dennett [1981].) A
spontaneous image implies the possibility of a judgement. But this does not mean that, in
the case of distinct memories, images are accompanied by the conviction that the event
happened. Rather, the claim is negative: It does not occur to the subject to doubt the
correctness of such images. In the case of flashbacks had by trauma victims, or other less
dramatic but still spontaneous memory-images, there is no need for the subject
additionally to judge: And this is how it happened. Where personal memory can be said
to have a phenomenology, it does not admit of analysis into two separable components,
image and judgement.
It may be argued that to have a memory-image is simply to conceive of the past object,
and that any stronger claim involves postulating static objects of internal perception or
introspectiona profoundly unReidian position.
9
In response, I would argue that when I
have memory-images of my latest minor car accident, there is an experiential quality
Andy Hamilton 235
8
For instance Stout [1930]. Arthur Collins writes that if there is a memory-image, its features count
as things that I remember only if I judge that the past I experienced was as the image is [Collins
1997: 79].
9
This objection was made to me by J.-C. Smith.
absent from simply thinking about the crash; though such images should not be regarded
as objects of internal perception. But the issue is a deeply problematic one on which much
more needs to be said. However, it should be understood that an excessively rationalistic
approach in philosophical discussion of memory results from failure properly to
acknowledge the role of images and feelings. If someone says My memories of him are a
comfort, they are not referring to the truths which they know on the basis of witnessing,
etc. Fundamentally, personal memory is retained knowledge of the past. But as narrative
models emphasizeor over-emphasizememories are not a static body of information,
since their meaning changes according to ones present situation or knowledge. Personal
memory also generates or sustains self-identitythe subjects conception of the person
they are. Perhaps because of this, misrememberings may still be regarded as memories
as when we say your memories of these events are getting hazy. But none of these
considerations implies an image theory.
Reid is concerned to undermine the image theorys account of the justification of
individual memory-reports. How then are such reports justified? I was there, I witnessed
it makes explicit the basis on which the past-tense judgement was made. This is not a
justification independent of memory, of course, since it is itself a personal memory-claim.
Once it is accepted that the past-tense claim is a putative memory-report, it may be
justified in many different waysby appeal to the reports of other witnesses, and to
material evidence. (There was a small earthquake here last night How do you know? I
was here, I felt it. But how do you know it was an earthquake? . . .) Where the subject
claims a distinct memory, however, these will be justifications for the audience, not for
the subject. To use currently prevalent but unhelpful terminology, such justifications are
externalist in the sense of not appealing to an introspectible or inner state, but internalist
in the sense of within the sphere of reasons. But surely memory itself can be justified as
reliable? An externalist account of the justification of memory-reports claims that memory
is a reliable belief-forming process. There are many things wrong with this claim.
Memory might be better-described as a belief-retaining than as a belief-forming
process. Most fundamentally, however, the attempt to justify memory as reliable is
misconceived, as Reid well understood. It is to his treatment of this elusive issue that I
now turn.
II. Those Things Really Did Happen Which I Distinctly Remember
Reid postulated as one of his twelve principles of contingent but self-evident truth that
Those things really did happen which I distinctly remember, arguing that [the] belief,
which we have from distinct memory, we account real knowledge, no less certain than if it
was grounded on demonstration [2002: VI.v, 474; III.i, 254]. Such beliefs, Reid
maintains, are irresistible, and our confidence in them cannot be explained as based on
experience. His formulation of principles of Commonsense may be regarded as a legiti-
mation of everyday beliefs in the face of Humean scepticism. But in the case of
perception, memory, and so on, the belief is not legitimated by showing that the faculty in
question is reliable.
If there are principles of normative epistemology, something like Reids memory-
principle must be one of them. Those who claim that he greatly overstates the reliability of
memory are mistaken. I have a distinct memory of moving to my present home on an
236 Scottish Commonsense About Memory
exceptionally mild winters day, and arriving at my house to find the removal van already
there. These are distinct memories, which I could elaborate with further detail, as well as
information based on inferencefor instance the year in question. Some of that detail
the colour of the removal van, how sunny it was that dayis less distinct and may be
inaccurate, but I find it inconceivable that the distinct memories as stated could turn out
incorrect. As it stands, however, Reids principle is too strong. It should be re-formulated
as a denial of completely-false memories: Those things which I distinctly remember are at
worst false-in-detail. That, I would argue, is what the reliability of memory amounts to.
Completely-false implies that nothing like the events reported ever happened.
(Completely-false memory is short for Sincere, attentive but completely-false memory-
claim.) Such memories are necessarily rare. Where memories are mistaken, they are
almost invariably false-in-detailin ordinary language, mis-rememberings. The contrast
between false-in-detail and completely false memories has been neglected, and it helps to
clarify recent debate over False Memory Syndrome.
In his discussion of False Memory, Ian Hacking contrasts an alleged mis-
rememberinga patient recalls being abused by her uncle when in fact it was her father
who was the abuserwith a completely-false memory of the same event, where no
abuse of any kind happened [Hacking 1995: 2589]. (He refers to the two cases as
contrary and merely false.) But misidentification of the perpetrator is surely too
significant to count as a mis-remembering. Even if it is imprecise and context-dependent,
however, the contrast between false-in-detail and completely-false is a vital oneas it is
also in the case of perception. The claim that mistaken memory-claims are normally
merely false-in-detail is not one to be asserted on the basis of psychological research.
Rather, this is how claims such as Your memory is mistaken or Your memory deceives
you are understood. If the subject is assumed to be sincere, we are inclined to think that
their memory-report is at worst false-in-detail. That, to reiterate, is what the reliability of
memory amounts to. Certainly it can happen that what I think I remember, turns out to be
something I dreamt vividly, or saw in a film, or was told. This would be a case of
completely-false memory, but it is rare, and special circumstances are needed to explain
it. Someone who regularly confuses what they have seen in a film with what they have
witnessed in real life is displaying not forgetfulness, but some kind of cognitive disorder.
Earlier I contrasted memory, apparent memory in its two senses, and imagination.
The varieties of past-tense statements purporting to concern a witnessed event may now
be demarcated more precisely. If true, the statement may be a personal memory-
judgement, or a judgement based on testimony or some other way of learning about the
event after it occurred. If false, it may be based on testimony and completely false, or
false-in-detail; or it may be a mis-remembering or false-in-detail memory; or it may, on
rare occasions, be a completely false memory. This completes the gloss on the re-
formulated principle of Commonsense, Those things which I distinctly remember are at
worst false-in-detail.
Now for the question of whether such a principle can be justified. Many psychologists,
and some philosophers, seem to believe that an empirical justification is possible. Alvin
Goldman writes: Memory can certainly yield mistakes. The question is: How pervasive is
the unreliability of memory, and what specific mechanisms account for such unreli-
ability? Psychologist James McClelland argues that memory is constructivethat is,
reconstructiveand that this has profound implications for the question of the
Andy Hamilton 237
veridicality of memory and the extent to which it may be influenced by suggestion,
preexisting knowledge, and other related experiences [Goldman 1986: 208, 183;
McClelland 1995: 69]. In contrast to these writers, Reid would maintainrightly in my
viewthat the reliability of memory could not be justified empirically. He regards the so-
called track-record argument for the reliability of memory as viciously circular, since it
depends on that reliability in its very construction: Every kind of reasoning for the
veracity of our faculties, amounts to no more than taking their own testimony for their
veracity [VI.v: 481]. Proponents of the track-record argument may urge that, when using
memory to arrive at the conclusion that memory is reliable, one does not make claims
about the reliability of memory; but there remains an effective circularity. (As illustrated
for instance by Audi [1998: 2249], and Pollock [1987: 11422].)
Reid had a deeper reason for rejecting an empirical justification of memorys
reliability, however, for he maintains that the question How reliable is memory? cannot
sensibly be posed. This, I would argue, brings him into conflict with reliabilism, a position
with which some writers have wanted to associate him.
10
But contrasting Reids view with
reliabilism requires some care. The latter position is externalist, defining knowledge as
true belief acquired by a reliable method, whether or not that method can be known to be
reliable; hence the claim is a conditional one, viz. if it constitutes a reliable belief-forming
process, then memory yields knowledge. Reids rejection of an empirical justification of
the reliability of memory therefore cuts no ice with reliabilists. However, they must at
least maintain that Memory is a reliable belief-forming process is an empirical claim,
and here Reids elusive remarks offer the beginnings of a critique. Reid is not a reliabilist,
I will argue, because he does not believe that the reliability of memory has empirical
status. This is shown by his separate claim that the track-record argument is not only
circular, but also inefficacious.
Reids claim occurs in the context of his discussion of memory as an original faculty
whose reliability is unaccountable. He continues:
When I believe that I washed my hands and face this morning, there appears no
necessity in the truth of the proposition . . . How then do I come to believe it? I
remember it distinctly. This is all I can say. This remembrance is an act of my mind. Is
it possible that this act should be, if the event had not happened? . . . [Unless a
necessary connection between them can be shown,] that belief is unaccountable, and
we can say no more but that it is the result of our constitution.
[2002: III.ii, 256]
If there is no necessary connection between the distinct remembering and the event, could
there be an empirical justification of the reliability of memory?
Perhaps it may be said, that the experience we have had of the fidelity of memory is a
good reason for relying upon its testimony. I deny not that this may be a reason to
those who have had this experience, and who reflect upon it. But . . . [it] must be some
very rare occasion that leads a man to have recourse to it; and in those who have done
238 Scottish Commonsense About Memory
10
For instance Woudenberg [1999], Baumann [1999], and De Bary [2002], which appeared too
recently for justice to be done to its arguments here. Van Cleve [1999] is, I think, mistaken in
claiming that Reids comments here leave space for a track-record argument.
so, the testimony of memory was believed before the experience of its fidelity, and that
belief could not be caused by the experience which came after it.
[Ibid.]
Reids claim here is not that the reliability of memory cannot be justified empirically
without circularity; rather, that our trust in memory could not be caused by experience of
its reliability.
It may be felt that Reids claim differs only in emphasis from Humes, in that he
simply describes our original instincts; note his statement that the belief which we have
of what we remember [is] unaccountable, and we can say no more but that it is the result
of our constitution, which is the will of our Maker. Thus we are psychologically impelled
to accept the deliverances of memory. Thomas Brown indeed suggests that Reid and
Hume disagree only in the emphasis which they put on the respective unjustifiability and
unavoidability of the principles of Commonsense: the sceptic pronounces the first
[unjustifiability] in a loud tone of voice, and the second [unavoidability] in a whisper,
while his supposed antagonist passes rapidly over the first and dwells on the second with a
tone of confidence [Brown 1846: II, 89]. Browns comment shows, I think, why Reid is
sometimes held to be a reliabilistbecause it is felt that otherwise, his position cannot be
distinguished from Humes, who is clearly not a reliabilist. At the root of Reids
reliabilism, on this view, is his Providential theism.
Clearly Brown is right to discern a structural similarity between the positions of Hume
and Reid. But they nonetheless inhabit strikingly different philosophical worlds, in a way
to which the reliabilist interpretation of Reid does insufficient justice. Indeed one should
resist Alstons view that, for Reid, the only (noncircular) basis we have for trusting
[memory and other sources of belief] is that they are firmly established doxastic
processes, so firmly established that we cannot help [trusting them] [Alston 1993: 127].
This is not a basis.
11
The reliabilist interpretation makes Reids position comparatively
uninteresting, and misses the subtletyor at least the elusivenessof the normative
dimension to his discussion. Reid maintains that, since the reliability of memory is never
seriously doubted, any purported empirical justification is a charadea symptom of the
insanity of the Ideal System. That the reliability of memory is never seriously doubted is,
moreover, a normative claimmeaning is never seriously doubted by a well-ordered
practical intelligence. (The phrase is from Gordon Graham [1998: 206]; see also
Skorupski [1993: 1114].) Reid here anticipates Wittgenstein in refusing to engage with
the sceptic. To say that a question makes no sense in ordinary discourse is not just to
describe original instincts or linguistic practices. It is to say that the sceptical question has
not been given a context in which it is really intelligible; hence its denial cannot take the
form of a straightforward empirical claim as reliabilists assume.
Henry Sidgwicks treatment shows how these claims may be developed into a critique
of reliabilism. Sidgwick, like Reid, suggests that the philosophical doubt is inconsistent
with other, ordinary, beliefs held by its proponents [Sidgwick 1905: 415]. But although
much-influenced by Scottish Commonsense, there is also a Kantian strain in Sidgwicks
work, shown by his explicit claim that the reliability of memory is a presupposition of
Andy Hamilton 239
11
Sosa is right to reply that such an argument would be just as circular as a track-record argument
[1994: 2789].
human enquiry. To obtain scientific or historical knowledge, he argued, we must assume
the general trustworthiness of memory, and the general trustworthiness of testimony under
proper limitations and conditions [Sidgwick 1882: 5423]. Sidgwicks tone is more
transcendentalist, in its claim of necessary conditions of knowledge; Reid for instance
says that we might have been so constituted that we had immediate knowledge of the
future rather than the past, and that this would have been no more unaccountable,
although much more inconvenient [2002: III.ii, 258]though this may be an example of
his dry humour. But like Sidgwick, Reid is not a reliabilist because he does not believe
that it could be a matter of empirical fact that memory is a reliable belief-forming process.
The pursuit of these issues takes us to the core of the interpretation of Reids Philosophy
of Common Sense, and suggest that its principles should be treated as basic presumptions
which we must take for granted, rather than as self-evident principles which form a basis
for our beliefs.
12
III. Memory Is An Original Faculty . . . Of Which We Can Give No Account,
But That We Are So Made
Reid claims two principles as unaccountable: that memory, or strictly, conception in
memory, is always accompanied by belief; and that, when distinct, it gives rise to true
belief. (A point made by Woudenberg [1999].) The first claim appears in the Inquiry:
Why sensation should compel our belief of the present existence of a thing, memory a
belief of its past existence, and imagination no belief at all, I believe no philosopher can
give a shadow of reason, but that such is the nature of these operations; they are all simple
and original, and therefore inexplicable acts of the mind [Reid 1997: 2.iii, 28]. I have
argued that it is a conceptual truth, rather than an unaccountable feature of our consti-
tution, that distinct memory is always accompanied by belief. In contrast, it is plausible
to regard the second principle, that distinct memories are true, as a principle of
Commonsense and thus unaccountable. For Reid, there is no ultimate explanation of
such facts or alleged facts, other than that God or Nature has so arranged our faculties. His
teleological conception of the constitutive properties of mind contrasts with Humes
causal conception; the substance dualism that underlies it is now deeply unpopular. Reid
has a more extensive category of the unambiguously mental, with a more consistent
demarcation from the physical, than Descartes. In the Meditations Descartes rejected the
view that I am lodged in my body like a pilot in a vessel. For Reid however, that
description is correct; the body is the instrument of the mind. Despite attempts to portray
Reid as a precursor of cognitive science, he rejected a science of the mind, denying the
possibility of an explanation of the relations between matter and mind.
13
In most of these general respects Reids account is clearly no longer viablealthough
for some contemporary sympathizers of Reid, evolutionary theory, though not an absolute
240 Scottish Commonsense About Memory
12
The contrast is found in Wolterstorff [2001], and commented on persuasively in Brun-Rovet [2002]
and Pakaluk [2002].
13
Edward Reed [1997], a humanistic critique of scientific psychology, argues that Reid advocated
a descriptive psychology, to elucidate the modes of this adaptation of the self to world [contrived
by the Deity], but denied even the possibility of an experimental, causally based psychological
science [1997: 27; also 13, 236, 28].
stopping-place, may take the place of Divine Providence. But his claim that the ability to
remember is primitive and unaccountable remains defensible, and is not in itself anti-
scientific. It means various things: Memory is not analysable in terms of the more basic
capacities of imaging and inductive inference, as the image theory says; its reliability is
not susceptible to empirical justification, as reliabilists claim; and memory-judgements are
not a causal consequence of neural processes or states which constitute the essence of
memory, as the trace theory maintains. This last issue I will now examine.
Reids rejection of the memory-trace model or trace theory is, I would argue,
consistent with the preservation of scientific psychology. Thus I disagree with the recent
claim that Reids opposition to memory-traces was achieved at the cost of dropping the
idea of explanation [of memory] and buying into non-physical souls and truly originary
free will.
14
There are many important questions about memory which psychologists could
continue to investigate without presupposing a trace theory, such as: optimum conditions
of recall, and how subsequent suggestion causes inaccuracies; learning tasks that can be
performed by amnesiacs; localization of brain function governing different recognitional
capacities. The trace-concept is, I would argue, an illegitimate extension of the concept of
localized brain-function which has been central to the development of the brain sciences
and remains almost an article of faith within them. Localization concerns particular
capacities such as recognition or memory. Because these clearly are capacities, there is
less likely to be a misunderstanding about what imputing localized function means, viz.
simply this: that if there is damage to a certain area of the brain, that function will be
impaired. Neural processes are not the essence of facial recognition, for instance, but its
causal condition. In fact, localized brain-function is not as well-founded as many psychol-
ogists supposeonce plasticity is admitted, claims of localization look emptybut the
idea of a localized trace is even more dubious.
The beginnings of a critique of the trace theory are found in Reids discussion, though
this is laconic even by his own standards. His main target is Locke, whose account has
elements of both image and trace theories, with some resulting incoherence. Locke
postulated a faculty that discerns similarity between past and present ideas, and Reid
responds that a memory faculty which had separate access to past ideas would make the
present idea redundant [III.vii, 353, 3556]. Reid is scathing about images in the brain,
regarding them as part of the Ideal Systems attempt to account for the unaccountable. He
argues that although the brain has been dissected times innumerable by the nicest
anatomists it has proved impossible to find any vestige of an image of any external
object. This is not surprising, for the brain is the most improper substance that can be
imagined for receiving or retaining images, being a soft, moist, medullary substance
[II.iv, 93]. If this is sarcasm, there is a serious philosophical point behind it. Traces surely
have to be structural analoguesin some sense, imagesof events remembered, and
Reid is right in claiming that the brain is not the right kind of thing to receive such
impressions.
15
It may be objected that such arguments against traces are now redundant, since
psychologists reject localized in favour of distributed traces. Proponents of connectionism
Andy Hamilton 241
14
Modern critics must either take on [the latter package], or find an alternative theory to do the jobs
which God, soul, and free will once did [Sutton 1998: 261].
15
On the neglected issue of why traces have to be structural analogues, see Squires [1969: 1934].
espouse a holistic position which maintains that information is evoked or stored
everywhere in the brain. Although such concepts are nebulous, they still imply a structural
analogue, otherwise the connectionist claim would be as anodyne as Reids acknowl-
edgment that memory is dependent on some proper state or temperament of the brain
[III.vii, 282]. Thus Reids view that memory is unaccountable forms the basis for a
continuing critique of the latest trace theories. Such a critique would insist that
information is essentially a personal, not a sub-personal or impersonal concept, and so
total neural activity cannot constitute memory-information as proponents of distributed
traces envisage. Relatedly, the psychological model of encoding, storage, and retrieval is
non-explanatory, since information-retention and remembering are interdefined concepts.
The basic sense of information-retention is retention by a person, not by a brainthat is,
remembering. Connectionism may suggest ways of theorising about the mechanisms of
memory, but not of capturing its essence.
16
IV. The Evidence We Have of Our Own Identity . . . Is Grounded on Memory,
And Gives Absolute Certainty
In the period between the decline of the Scottish Commonsense school and his recent
induction into the canon of major philosophers, Reid was probably best-known for his
objections to Lockes account of personal identity. His deepest criticism, shared with
Butler, is that memory presupposes personal identity and so cannot ground it; on Lockes
criterion, he argues, personal identity is confounded with the evidence which we have of
our personal identity [III.vi, 277]. I have argued elsewhere that this criticism ignores the
possibility that memory and personal identity form a conceptual holisma possibility
which J. S. Mill recognized when he wrote that the phenomenon of Self and that of
Memory are merely two sides of the same fact [Mill 1989: 21213]. (Despite Mills
hostility to the school of intuition which Reid represented, to a surprising extent he
followed a Reidian agenda.)
17
Here I wish to examine Reids central positive claim that the proper evidence [of a
permanent self] is remembrance . . . The evidence we have of our own identity, as far back
as we remember, is totally of a different kind from the evidence we have of the identity of
other persons, or of objects of sense. The first is grounded on memory, and gives absolute
certainty [III.iv, 2646]. This claim becomes one of the twelve first principles of
contingent truth: we know immediately our own personal identity and continued
existence, as far back as we remember anything distinctly [VI.v, 476]. Reids claim
expresses an interesting measure of agreement with Locke. It is, however, an incautious
one. As argued earlier, the subject does not have infallible knowledge of their identity
over time, since there is no introspectible state known with certainty and common to all
cases of at least apparent memory. The claim of authoritative access to ones past actions
242 Scottish Commonsense About Memory
16
Space does not permit development of these claims here; the issue is pursued at greater length in
Hamilton forthcoming.
17
Mills view was discussed further by Hamilton [1998a]. The interpretation of Locke as offering a
memory criterion was greatly influenced by Reids presentation of his account; that interpretation is
persuasively criticized in Southgate [2000].
and experiences is not without foundation, however. Earlier it was claimed that personal
memory is a way of knowing about others, in that its characteristic forms are I remember
X o-ing as well as I remember o-ing. But only in the latter casewhere it concerns
oneselfdoes personal memory involve a distinctively self-conscious way of knowing.
That is, it is not a way in which I could know about someone else.
This distinctively self-conscious way of knowing is manifested in a phenomenon quite
close to Reids certain knowledge of our identity through memory, viz. immunity to error
through misidentification (IEM). Consider the claim I remember swallowing a ball-
bearing when I was a young child. If, for whatever reason, I come to doubt that this did
happen to me, it will make no sense for me to continue to maintain, using the same justifi-
cation, that nonetheless, someone swallowed a ball-bearing. That is, I cannot say Well, I
distinctly remember someone swallowing it. (Note the importance of the continuous-verb
formulation.) When the claim is based on testimony, howeverwhen it constitutes merely
factual memory of my pastthe corresponding retreat claim does make sense.
18
Thus although Reid was wrong to hold that personal memory gives infallible
knowledge of ones identity, it does yield unique knowledge. Personal memory is a
distinctively self-conscious way of knowing about oneself, and is integral to self-
consciousness. IEM thus constitutes a special feature of personal as opposed to factual
memory-judgements. Its existence shows that sympathizers of Reid are misguided when,
out of a concern that it sustains the image theory, they seek to undermine the distinction
between personal and factual memory. One example is Roger Squires, who rejects the
idea that personal memory is a special or strict sort of memory, claiming instead that all
remembering is essentially the retention of knowledge or information. He is right that
personal remembering is not introspectively distinguishable as the image theory
maintains, but wrong to elide it so closely with factual memory.
19
Others have argued that
the continuous-verb locution fits cases of recollection or recall, but not other examples of
personal memory, while Malcolm defines perceptual memoryassociated with images
as a sub-category of personal memory, which does not necessarily involve mental
imagery. But the lack of a continuous-verb description, or of any supporting detail, must
surely make the subject wonder whether they really remember the event at all, or were
simply told that it had happened.
20
Although in respect of its relation to personal identity, Reids account of memory is
flawed, it remains original and historically important. In emphasising that memory
involves belief and hence judgement, he was the first philosopher to transcend the impov-
erished account offered by the theory of ideas, in this respect there are parallels with Kant.
Although his discussion is compressed, Reids caution on psychological theorising about
Andy Hamilton 243
18
These claims, and the conceptual holism of memory and personal identity, are defended in
Hamilton [1995; Hamilton forthcoming].
19
He claims that what is special is the witnessing rather than the type of rememberingthe detail of
the witnesss description is more open-ended, while non-witnesses can give only circumscribed
reports dependent on information imparted to them [Squires 1969].
20
Malcolm thinks it makes sense to say: I remember [my grandfather] well. He always wore his
Confederate uniform, chewed tobacco, and told jokes; but I cant any longer see him in my mind as
I formerly could. But then, surely, I do not remember him wellmy memory is more factual than
personal [Malcolm 1963: 221].
the reliability and operations of memory is of enduring value. He thought more deeply
about memory than many of his successors who have been more widely discussed.
21
Durham University
Received: March 2002
REFERENCES
Alexander, S. 1920. Space, Time and Deity: Gifford Lectures, 191618, London: Macmillan.
Alston, W. 1993. The Reliability of Sense Perception, Ithaca: Cornell University Press.
Anscombe, G. E. M. 1981. Memory, Experience and Causation, in Collected Philosophical Papers,
Vol. II, Oxford: Blackwell.
Audi, R. 1998. Epistemology, London: Routledge.
Baumann, P. 1999. The Scottish Pragmatist? The Dilemma of Common Sense and the Pragmatist Way
Out, Reid Studies 2: 4758.
Block, N., ed. 1981. Imagery, Cambridge MA: MIT Press.
Brown, T. 1846. Lectures on the Philosophy of Mind, 4 vols., Edinburgh: William Tait,.
Brun-Rovet, E. 2002. Reid, Kant and the Philosophy of Mind, Philosophical Quarterly 52: 495510.
Campbell, J. 1994. Past, Space and Self, Cambridge MA: MIT Press.
Chappell, V. 1989. The Theory of Sensations, in The Philosophy of Thomas Reid, eds. M. Dalgarno
and E. Matthews, Dordrecht: Kluwer: 24056.
Collins, A. 1997. Personal Identity and the Coherence of Q-Memory, Philosophical Quarterly 47:
7380.
Dalgarno, M. and E. Matthews, eds. 1989. The Philosophy of Thomas Reid, Dordrecht: Kluwer.
De Bary, P. 2002. Thomas Reid and Scepticism: His Reliabilist Response, London: Routledge.
Dennett, D. 1981. The Nature of Images and the Introspective Trap, in Imagery, ed. N. Block,
Cambridge MA: MIT Press: 15071.
Evans, G. 1982. The Varieties of Reference, Oxford: Clarendon Press.
Goldman, A. 1986. Epistemology and Cognition, Cambridge MA: Harvard University Press.
Graham, G. 1998. Lukacs and Realism After Marx, British Journal of Aesthetics 38: 198208.
Hacking, I. 1995. Rewriting the Soul: Multiple Personality and the Sciences of Memory, Princeton:
Princeton University Press.
Hamilton, A. 1995. A New Look at Personal Identity, Philosophical Quarterly 45: 33249.
Hamilton, A. 1998a. Phenomenalism and the Self, Cambridge Companion to Mill, ed. J. Skorupski,
Cambridge: Cambridge University Press.
Hamilton A. 1998b. False Memory Syndrome and the Authority of Personal Memory-Claims, and
Reply to Commentators, Philosophy, Psychiatry, and Psychology 4: 28399.
Hamilton, A. forthcoming. Memory and the Body: A Study of Self-Consciousness.
Lehrer, K. 1989. Thomas Reid, London: Routledge.
McClelland, J. 1995. Constructive Memory and Memory Distortions: A Parallel-Distributed
Processing Approach, in Memory Distortion: How Minds, Brains and Societies Reconstruct the
Past, ed. D. Schacter, Cambridge MA: Harvard University Press: 190204.
Macdonald, G., ed. 1979. Perception and Identity: Essays Presented to A. J. Ayer, London: Macmillan.
McDowell, J. 1997. Reductionism and the First Person, in Parfit and His Critics, ed. J. Dancy, Oxford:
Blackwell: 23050.
Malcolm, N. 1963. Knowledge and Certainty, Eaglewood Cliffs NJ: Prentice-Hall.
Martin, C.B. and M. Deutscher 1966. Remembering, Philosophical Review 75: 16196.
Mill, J.S. 1989. Collected Works Volume 31: Miscellaneous Writings, Toronto and London: University
of Toronto Press and Routledge.
Mounce, H. 1994. The Philosophy of the Conditioned, Philosophical Quarterly 44: 17490.
Pakaluk, M. 2002. A Defence of Scottish Common Sense, Philosophical Quarterly 52: 56481.
244 Scottish Commonsense About Memory
21
I am grateful for comments from Peter Baumann, Robin Hendry, E. J. Lowe, David Papineau,
Roger Squires, John Skorupski, and especially John-Christian Smith, whose patient responses have
saved me from several errors of interpretation; and from the anonymous referees of this journal.
Pollock, J. 1987. Contemporary Theories of Knowledge, London: Hutchinson.
Reed, E. 1997. From Soul To Mind: The Emergence of Psychology from Erasmus Darwin to William
James, New Haven: Yale University Press.
Reid, T. 1967. Philosophical Works, Hildesheim: Georg Olms.
Reid, T. 1997. An Inquiry Into The Human Mind, ed. D. Brookes, Edinburgh: Edinburgh University
Press.
Reid, T. 2002. Essays On The Intellectual Powers Of Man, ed. D. Brookes, Edinburgh: Edinburgh
University Press.
Schacter, D., ed. 1995. Memory Distortion: How Minds, Brains and Societies Reconstruct the Past,
Cambridge MA: Harvard University Press.
Sidgwick, H. 1882. Incoherence of Empirical Philosophy, Mind 7: 53048.
Sidgwick, H. 1905. The Philosophy of Common Sense, in The Philosophy of Kant and Other Lectures,
London: Macmillan.
Skorupski, J. 1993. English-Language Philosophy, 17501945, Oxford: Oxford University Press.
Smith, J.-C. 2000. Companion To The Works Of Philosopher Thomas Reid, 17101796, Lewiston NY:
Edwin Mellen Press.
Sosa, E. 1994. Philosophical Scepticism and Epistemic Circularity, Proceedings of the Aristotelian
Society Supp. 68: 26390.
Southgate, N. 2000. Personal Identity and Human Animals: A New History and Theory, PhD: Durham
University.
Squires, R. 1969. Memory Unchained, Philosophical Review 77: 17897.
Stout, G. F. 1930. In What Way is Memory-Knowledge Immediate, Studies in Philosophy and
Psychology, London: Macmillan: 6680.
Sutton, J. 1998. Philosophy and Memory Traces, Cambridge: Cambridge University Press.
Van Cleve, J. 1999. Reid on the First Principles of Contingent Truths, Reid Studies 3: 330.
Wittgenstein, L. 1958. Philosophical Investigations, Oxford: Blackwell.
Wittgenstein L. 1980. Remarks on the Philosophy of Psychology, Volume 1, Oxford: Blackwell.
Wittgenstein, L. 1982. Last Writings on the Philosophy of Psychology Volume I, Oxford: Blackwell.
Wolterstorff, N. 2001. Thomas Reid and the Story of Epistemology, Cambridge: Cambridge University
Press.
Woudenberg, R. 1999. Thomas Reid on Memory, Journal of the History of Philosophy 38: 13048.
Andy Hamilton 245