This paper offers a reading of Memento as a therapeutic dialogue, one purpose of which is to loosen the grip of both dualism (the ‘Cartesian’[1] picture of mind) and behaviourism. We do this by illuminating aspects of the film with a reading of the opening of Wittgenstein’s Philosophical Investigations (Hereafter, PI).
PI opens with the following (we quote at length and in full):
“When they (my elders) named some object and accordingly moved towards
something, I saw this and grasped that the thing was called by the sound they
uttered when they meant to point it out. Their intention was shewn by their
bodily movements, as it were the natural language of all peoples: the
expression of the face, the play of the eyes, the movement of other parts of
the body, and the tone of voice which expresses our state of mind in seeking,
having, rejecting or avoiding something. Thus, as I heard words repeatedly used
in their proper places in various sentences, I gradually learnt to understand
what objects they signified; and after I had trained my mouth to form these
signs, I used them to express my own desires.” (Augustine, Confessions,
These words, it seems to me, give us a particular picture of the essence of human language. It is this: the individual words in language name objects—sentences are combinations of such names.—In this picture of language we find the roots of the following idea: Every word has a meaning. The meaning is correlated with the word. It is the object for which the word stands.
Augustine does not speak of there being any difference between kinds of word. If you describe the learning of language in this way you are, I believe, thinking primarily of nouns like “table”, “chair”, “bread”, and of people’s names, and only secondarily of the names of certain actions and properties; and of the remaining kinds of word as something that will take care of itself.
Now think of the following use of language: I send someone shopping. I give him a slip marked “five red apples”. He takes the slip to the shopkeeper, who opens the drawer marked “apples”; then he looks up the word “red” in a table and finds a colour sample opposite it; then he says a series of cardinal numbers—I assume that he knows them by heart—up to the word “five” and for each number he takes an apple of the same colour as the sample out of the drawer.—It is in this and similar ways that one operates with words.—“But how does he know where and how he is to look up the word “red” and what he is to do with the word “five”?”—Well I assume that he acts as I have described. Explanations come to an end somewhere.—But what is the meaning of the word “five”?—No such thing was in question here, only how the word “five” is used.
(PI: §1)
Now, the standard[2] reading of §1 of the Investigations reads it purely as an attack on the Augustinian picture of language. This is “Augustine’s pre-theoretical, pre-philosophical picture of the working of language which informs Augustine’s own remarks on language as well as a multitude of sophisticated philosophical analyses of meaning” (ACPI i: p. 61).[3] This picture of language, we are told, provides the paradigm within which Frege (FA), Russell (PoM) and Wittgenstein (TL-P)[4] were alleged to operate (see Baker and Hacker 1980: pp. 45-59).[5] Baker and Hacker claim that what is of interest to Wittgenstein in PI §1 is not an ‘inner’ and ‘outer’ theory of mind, and other concerns that may be implicit in the passage from Augustine, but merely a number of related issues relating to word meaning. They write:
[Wittgenstein] is concerned only with the points explicit in the quotation in (a).[6] (iv) Words signify or name objects. (v) Sentences are combinations of words. (vi) that a word signifies a given object consists in the intention with which the word is used. (vii) The intention with which a word is used (i.e. the intention to mean that object) can be seen in behaviour, bodily movement, facial expression, tone of voice, etc.
(ibid: 61)
The trip to the grocer in paragraph (d) is taken, by Baker and Hacker, to illustrate different types of words (ibid: 63). They write, “The example is designed to stress the fact that the contention that the three words are of different types rests on the differences in the operations carried out in each case, and on the ordering of the operations” (ibid).
We think this vastly underplays both the subtlety and the significance of PI §1:d. In what follows, we first outline a reading of (d) that we think captures the nuance—not to mention the philosophical import—of the example. We do this taking our inspiration from Stephen Mulhall’s reading of the passage.[7] We then turn to a reading of Memento to show that this stands as a feature length version of Wittgenstein’s ‘short’, and in doing so explores the issues in play in a manner rich enough to provide one with further philosophical (therapeutic) insight of which Wittgenstein would have been proud.
- - - - - - - - - -
There is something of a conundrum for all who pick up Wittgenstein’s PI, and particularly for those who read the opening as Baker and Hacker do. Why did Wittgenstein choose a passage from Augustine’s Confessions and not one from a recognised work in the philosophical canon? Indeed, after choosing to cite Augustine’s Confessions he then chooses to cite in particular a passage from the autobiographical sections of the text rather than from Augustine’s more overtly (and sometimes, indeed, explicitly) metaphysical writing. Furthermore, why did he choose to illustrate the limitations of the picture he identifies at play in the quote from Augustine with what is, on reflection, a decidedly eccentric depiction of a trip to the grocer?—the shopper appears to be dumb, the (rather mechanical) grocer keeps apples in drawers and counts them out individually after matching the colour with a colour chart.
Is it really a satisfying conclusion to write that the grocer example is an “illustration of different types of words.” That it shows simply that
‘Five’, ‘red’, and ‘apple’ are words each one of which belongs to a type the use of which is fundamentally different from the use of words of the other types. To say that ‘apple’ is the name of a fruit, ‘red’ the name of a colour, and five the name of a number would mask deep differences beneath superficial similarities. Again, one might think ‘apple’ involves correlation with an object, ‘red’ with a colour, and ‘five’ with counting objects of a type, so all words involve correlation with something. The web of deception is readily woven. (ACPI i: p. 63)
Well we do not find it so–or at least, the “web” is not quite what Baker and Hacker say it is, and the difference is important. If there is a ‘web of deception’ here, it is a more subtle web. It is a web which Wittgenstein wants you to fall into and wants you to then struggle free from entanglement therein. A crucial part of the therapeutic work–that one does on oneself via the reading of the opening of PI—comes through the gradual realisation of just how bizarre the scenario(s) depicted by Wittgenstein is (are). It is one’s temptation to (initially) see the shopkeeper scenario as a plausible enough scenario with philosophical morals following from it that is subject to a deconstruction that is illuminating–a ‘deconstruction’ that one must perform for oneself. What we consider the ‘overture’ to PI, the first 36 sections or so, is in effect Wittgenstein asking the reader over and over again, ‘Will this satisfy you as being language? Perhaps not? Then will this be enough for you to regard what is happening here as language, with all the consequential results you think will accrue from such a designation? …And are you not worried by your tendency to gloss over how peculiar the scenarios are that I have asked you to ‘conceive’ as language?’
By contrast, the Baker and Hacker reading is not satisfactory because:
a. it leaves so many questions and niggles hanging in the air, either unanswered or simply unasked and
b. in saying that each word belongs to a different type of use, it implies there is a ‘type of use’ which can be definitively associated with each word in the language.
We think that responding efficaciously to both a. and b. is achievable, through a more subtle reading of the passage, expanding on the series of implicit questions that we just attributed to Wittgenstein as his intention in posing the ramifying set of ‘examples’ that the reader is confronted with, early in PI. We begin with b.
How does claiming that each word belongs to a type of use get us further than appealing to correlations with things, or words as names of things? All one has in fact done is exchange ‘things’ for a ‘type of use’. To come at this from one side, consider that both ‘five’ and ‘red’ can name things when embedded in certain sets of practices. When playing football I can readily name my mate Jim, the left back, ‘Five’, because that is his number in the team, the number written on his shirt; and, furthermore, it makes sense to do so because there are two players named Jim on our team. Similarly, Red Adair’s friends were quite in order referring to him (calling him by name) as ‘Red’. It was his name. The suggestion of an appeal to use that Wittgenstein makes in his later work is not to show that words have different types of use with which we can classify them accordingly; replacing our crude grammatical terms such as noun, verb, adjective, etc. But rather to show that words might play many different sorts of roles, and which one they do play depends ultimately on the usage in a context and on an occasion. Words do not ‘belong to types of use’.
The mistake here then is Baker and Hacker’s thought that what is problematic for Wittgenstein—what he wants to critique in these opening remarks—is that words name things or correspond to objects, with the emphasis laid on the nature of what is on the other side of the word-Ω relationship. Rather, we contend that what is problematic in this picture is that words must be relational at all—whether as names to the named, words to objects, or ‘words’ belonging to a ‘type of use.’[8] It is the necessarily relational character of ‘the Augustinian picture’ which is apt to lead one astray; Baker and Hacker, in missing this, ultimately replace it with a picture that retains the relational character, only recast.[9] There is no such thing as a word outside of some particular use; but that is a different claim from saying, with Baker and Hacker, that words belong to a type of use. For a word to be is for a word to be used. Language does not exist external to its use by us in the world. Language cannot, in John McDowell’s phrase, be viewed from sideways on. This is a thought that animates Wittgenstein’s thinking from TL-P onwards.[10]
The key to understanding PI §1, we think, is in reading the passage as a whole, but with particular attention to the remark, toward the end, of Wittgenstein’s interlocutor, and to the way Wittgenstein then responds. Recall that in response to Wittgenstein’s ‘short story’ the interlocutor says, and Wittgenstein responds, as follows:
“But how does he [the grocer] know where and how he is to look up the word “red” and what he is to do with the word “five”?”—Well I assume that he acts as I have described. Explanations come to an end somewhere.—But what is the meaning of the word “five”?—No such thing was in question here, only how the word “five” is used.
The question to ask, therefore, is why is Wittgenstein’s interlocutor not satisfied with the scenario? What is it that she yearns for in asking her question? To take Baker and Hacker’s line on this is to give no thought to the purpose of the interlocutor’s remark but rather to interpret Wittgenstein’s response to it as indicating that questions of the genesis of ‘meaning’ are not philosophical questions as they are contingent and thus of no philosophical import (cf. ACPI i: 64). But Wittgenstein is the author of the interlocutor’s question (it is his question, he has this yearning or he at least sees it as a significant, appropriate, yearning); why author a question only to dismiss it as insignificant and inappropriate in the following sentence? A more satisfactory (to our eyes) interpretation of the purpose of the interlocutor’s question is Stephen Mulhall’s. Recent Mulhall argues that the question invokes the notion of ‘meaning’ coming from an inner mental process. The interlocutor is not satisfied with the explanation only being given with reference to outward criteria, or behaviour; the use of colour samples and the counting out of the apples one-by-one leaves her still wanting to know more about from where meaning might come. Mulhall writes, “the cast of her [the interlocutor’s] questions rather takes it for granted that nothing behavioural can settle the issue of understanding even in principle; only a transition to the entirely separate realm of the inner can give her the reassurance she craves” (Mulhall, 2001: 44).
It is our contention, that the grocer example in PI §1 is analogous to (is a more profound precursor of) John Searle’s famous ‘Chinese Room’ example. However, rather than serving as an argument against strong AI, it serves more generally to bring to light the underlying prejudices that lead one to behaviourism and dualism and so forth. Searle’s Chinese Room example is designed to prompt one to question whether the processing of symbols could ever be a sufficient condition for the attribution of ‘understanding’ to the processor. Wittgenstein sets up the grocer example in such a way that it leads one to crave something that will satisfy one to the extent that one would be happy to say of the grocer, “He understands”.
Wittgenstein’s example is designed to tempt us into positing inner mental processes of some sort. This, Wittgenstein’s interlocutor does. However, this is only to begin to understand the reach of the example. For when we reflect upon what such a (inner) process might be we find that we want to describe something very similar to the (‘external’) behaviour that Wittgenstein’s grocer does exhibit. Consider: A note is passed to him—data is entered. The words on the note are related to objects—the input data is related to inner mental items (samples). More precisely: the word ‘apple’ is matched to the object ‘apple’—the apple-data is related to a mental image of an apple; the word ‘red’ is matched to a colour sample—the colour-data is related to a mental image of red. Then, having ascertained what ‘a red apple’ ‘means’—having related the data with the correct mental image of what we call ‘a red apple’—we count five of them—we mentally mark-off the lines in the five bar gate, or mentally slide the beads of the abacus across.[11] The data is thus processed. The grocer retrieves five red apples and hands them to the note-bearer—he ‘understands’ the request.
Here is the point of the ‘eccentricity’ of the example. The example is structured to mirror the (alleged) form of inner mental processes. In tempting the interlocutor to ask for more so that she might be satisfied in predicating of the grocer understanding Wittgenstein tempts the interlocutor into undermining her own prejudices. Mulhall writes: “…if the public, externalised versions of such procedures were not in themselves enough to establish the presence of understanding to the interlocutor’s satisfaction, why should their inner counterparts?” (Mulhall 2001: 45) Is it because they are inner? This is surely not enough.
However, the subtlety of the Wittgenstein’s example does not stop there. For, as Mulhall notes (46),
If Wittgenstein’s shopkeeper’s way with words strikes us as surreal and oddly mechanical, to the point at which we want to question the nature and even the reality of his inner life, and yet his public behaviour amounts to an externalised replica of the way we imagine the inner life of all ordinary, comprehending language-users, then our picture of the inner must be as surreal, as oddly mechanical, as Wittgenstein’s depiction of the outer.
This brings us back to Baker and Hacker, because now we can gain a fuller understanding of what is misleading—and missing—in their account.[12] The purpose of PI §1 is not in the end that of replacing Augustine’s picture with another—that of words “belong[ing] to a type of use”, but of facilitating one’s realisation that Augustine’s picture amounts to nothing one wishes to hold on to. It is the thought of words as essentially relational that is holding one captive here.[13] It leads one to yearn for inner mental processes when ultimately these can never be more satisfying than external processes. It leads one to continue one’s search for something to which one’s words might relate.[14] Appealing to words belonging to ‘type of use’ does little to wean one off this because one can just as well appeal to the ‘words’ of (say) mentalese belonging to a ‘type of use’, while arguing (comforting ourselves) that (all) we are (doing is) taking the analysis to a more fundamental level. It is the relational view of words that leads to both dualism (in all its varieties) and behaviourism, or (most generally) to one or another variety of unnecessarily theoretical account of the mind.
Memento
Memento is the story of Leonard Shelby. On Leonard’s own account he has an inability to form memories. This, he claims is the result of neurological damage sustained during an attack on he and his wife; an attack in which his wife was raped and killed. In order that he can function and pursue his wife’s killers he calls on a number of external resources (sometimes Byzantine in their strangeness or complexity) to ‘replace’ his memory. He has a file, similar to a police case-file, in which he accumulates evidence pertaining to ‘John G’ (his wife’s ‘murderer’); in addition he has a large wall-chart which he carries in the file which is designed like a flow chart or spider-gram employing notes and photographs. He has a Polaroid camera with which he takes photographs of people he meets, new belongings such as cars, and places such as the hotel in which he is staying; he then writes notes to himself on the back of these photographs. These notes might, initially, be just the person’s name, but over time Leonard will add details such as “Don’t trust him” or “She is your friend”. Leonard carries these annotated photographs about his person, constantly flicking through them as someone approaches who appears to know him. The final way in which Leonard compensates for his inability to make new memories is by having crucial information tattooed about his person. The tattoos are comprised of those that serve as reminders of his task—“kill him”—and crucial pieces of evidence which will bring him closer to achieving his task—such as names or partial names of chief suspects: John G..
This is Leonard’s ‘system’. He remarks repeatedly, “You’ve gotta have a system.”
So far, we have presented this account straightforwardly. As anyone who has seen the film will know all too well, however, the film itself is not straightforward in its mode of presentation. The film opens with a scene in which one gradually comes to realise, that, somewhat bizarrely, the film-spool is running backwards. After a couple of minutes, it suddenly reverses and runs forward for a few seconds. One’s sense of relief at this, however, lasts a very short time. For one is thrown off one’s guard again, as there is a cut to a completely different scene—this time in black and white—where the protagonist who one has just seen (who one comes to learn is called “Leonard”) begins to tell, from media res, of his confusion at the situation he is in. Soon the film switches to a colour scene, again unconnected, in which what one will come to see as the ‘action’ of the film begins. In the black and white scenes, which from then on are interspersed with the main action of the film, Leonard attempts to understand what is happening to him, and to explain to an unseen listener the nature of his condition, a condition that keeps him in a state of perpetual demi-confusion. Meanwhile, the ‘main action’ of the film proceeds in a sequence of scenes that run forward, normally—but one comes to realize that each scene is temporally prior to the action of the next. I.e. we are witnessing a story whose main action goes back in time as the film progresses.
One gradually comes to work all this out as one watches the film. The film is a puzzle that one has to work hard to try to unravel. Starting from a position of confusion, one tries to puzzle out one’s ‘condition’, as the viewer of this film. Is the film’s obscurity warranted? Or is it rather just a grand and clever trick, as some other recent ‘cult’ films such as ‘12 Monkeys’ and ‘The Usual Suspects’ arguably are?
The film sets out to induce in us an experience; an experience as of the very protagonist whose experience we are seeking to understand. One comes vicariously to inhabit Leonard’s peculiar, confusing, tragic condition, by being placed in this position of oneself having to play the detective, as he perpetually does. One comes to understand, to know, for the first time, what it would be like to be like someone like him. There is no counterpart to this in most other ‘trick’ films. The reflexivity ‘Memento’ requires of an attentive audience is something of intellectual substance.
So: we inhabit Leonard’s position, vicariously. The film forces us to confront his deep difference from us. It enables us as viewers to progress in understanding what a mind might be that was unlike our mind, and what philosophical problems might be real to such a mind that are generally unreal to any of our minds.[15] ‘Memento’ involves the attentive viewer in a prolonged, dialogical inquiry, in which the viewer must be active–and part of that activity consists sooner or later in trying to answer questions such as ‘Just how similar to or different from Leonard are you? You want to identify with him; does he have what you are unreservedly prepared to call an ordinary human mind or personhood or a moral code when you see him doing this? Or apparently imagining this? And are you not increasingly worried by your tendency to gloss over some of the ways in which he is so strikingly different from you/us that perhaps you should have more reservations? (And reservations that extend to philosophical conclusions you may be inclined to draw concerning what must be in the head of someone, that is perhaps lacking inside his head?)’ [16]
We suggested that Wittgenstein encourages us to inhabit the attractions of ‘Augustinianism’, far more than Baker and Hacker acknowledge. But will one have really gained anything, if one then simply lives in one of these apparent alternatives, as Baker and Hacker do, and dismisses all the rest? Philosophy for Wittgenstein is about liberation from the compulsion of any picture(s).[17] ‘Memento’ involves a viewer who is actually watching it in philosophical work that involves a deeply similar struggle for liberation. The film’s ‘obscurity’ [18] is warranted; and indeed, we suggest, intrinsic. It the film’s central brilliance.
Just what, then, are we to make of Leonard? Given the earlier discussion of Wittgenstein’s grocer, we might make the following observations. One is tempted in watching the film to see Leonard’s ‘condition’ as depriving him of the ability to form normal human relationships—he does not recognise the woman he slept with on waking the following morning; he only knows who are friends, who are trust-worthy, who are enemies etc. by having written as much on the back of a Polaroid of them. One is also made aware that his ‘condition’ makes him vulnerable to the manipulation of others—the Carrie-Ann Moss character (Natalie) provokes him into punching her, then leaves him alone in her house after removing all writing materials. She returns after a few minutes (when the thought has left him) and bloodied and bruised tells him that someone else had beaten her. Leonard is then sent off by her to get revenge on the person who she says (lying) ‘beat’ her, a person she wants run out of town or done in. Indeed the dénouement[19] of the whole film rests on Teddy (a corrupt police officer who has ‘befriended’ Leonard) confessing to using Leonard as a way of killing off drug dealers and pocketing the cash—moving from town to town and feeding Leonard enough information so that he will identify the dealer as his wife’s murderer. The two points are related: it is Leonard’s inability to form ‘normal’ human relationships which allows for his vulnerability to the manipulation of others. In addition, we might say that in being so open to the manipulation of others Leonard could be said to be deprived of something further—his moral responsibility. Leonard, it would seem, cannot be held responsible for his actions because of his condition.[20]
Where does this lead. Well, we—initially—find ourselves wanting to agree, that it is his inability to form memories that precludes him from forming normal human relationships; this is also what makes him vulnerable to the manipulation of others and results in his lack of moral responsibility. But were we to find a cure for Leonard; were we able to operate and repair those misfiring synapses, what then? Would we not, as did Wittgenstein’s interlocutor, merely be implying that the internalising of those things that Leonard already does externally is enough to solve our problem? If the file and tattoos become memories and the Polaroid photographs become mental pictures associated with certain memories what have we (and Leonard) gained? The very fact of interiority brings us nothing over what we (Leonard) already had.[21] And if you still incline toward thinking that interiority must be better, then try the following remark of Wittgenstein’s:
If I give someone the order, “fetch me a red flower from that meadow”, how is he to know what sort of flower to bring, as I have only given him a word?
Now the answer one might suggest first is that he went to look for a red flower carrying a red image in his mind, and comparing it with the flowers to see which of them had the colour of the image. Now there is such a way of searching, and it is not at all essential that the image should be a mental one. In fact, the process may be this: I carry a chart co-ordinating names and coloured squares. When I hear the order “fetch me [etc.]” I draw my finger across the chart from the word “red” to a certain square, and I go and look for a flower which has the same colour as the square. But this is not the only way of searching and it isn’t the usual way [!!]. We go, look about us, walk up to a flower and pick it, without comparing it to anything. To see that the process of obeying the order can be of this kind, consider the order “imagine a red patch”. You are not tempted in this case to think that before obeying you must have imagined a red patch to serve you as a pattern for the red patch which you were ordered to imagine.[22]
Wittgenstein here establishes a possibility that frees us from the mental cramp of imagining that things have to be as Cognitivists et al imagine them to be. And let us now compare in more detail the second point Mulhall made regarding Wittgenstein’s grocer. If Leonard’s life with his ‘aides memoirs’ (and in Leonard’s case they are just memories period) strikes us as surreal and mechanical why does our interiorising such things make them less surreal and less mechanical? It does not—it only makes them seem less so, because of the psychological or cultural roots of philosophical delusion. A conception of mind as an inner realm populated by mental representations (whether pictorial or syntactically structured) which we access on the input of sensory data is precisely that which is being represented externally in ‘Memento’. ‘Memento’ then dissolves our appeal to an inner realm by showing—through Leonard’s character—how such an appeal falls short of giving us what we want: a person.
What was it that we found in Leonard that made us think his ‘condition’ made him abnormal? We said he could not form normal relationships. We said he could be easily manipulated by those who chose to do so. We said this susceptibility to manipulation abdicated him of the moral responsibility that we might accord a ‘normal’ person. But what is abnormal about the relationships Leonard forms? Well it is surely abnormal to wake-up naked in someone’s bed and then have to then look through your selection of Polaroid photographs and match one with her, then turn it over to read her name and your comments on what you ‘think’ of her based on previous encounters. But interiorise that scene. Leonard wakes-up naked in someone’s bed, he sees a woman laying beside him, he looks through his selection of Polaroid photographs and matches one with her—his senses send an image of her to his mind, the image is then matched with a mental representation of her; Leonard turns the photograph over reads the name and comments he has written on the back which tell him what he ‘thinks’ of her based on previous encounters—the mental image is associated with some memories of their previous meetings this tells him what he ‘thinks’ of her.
The first move in the viewer’s diagnosis of Leonard then is not what one most likely thought it was. Leonard’s inability to form ‘normal’ relationships is, we submit, not born of his condition preventing him from being able to form memories (new mental representations) but of others’ awareness and exploitation of the fallibility of his system of forming memories. The awareness is informed by Leonard’s own propensity to tell everyone he meets about his ‘condition’[23] and outwardly manifest his ‘condition’. However, we suspect that, more importantly, Leonard has submitted to a pathological pull toward being consumed by the desire to avenge the murder of his wife. Leonard’s ‘abnormality’—his seeming inability to form relationships and his heightened susceptibility to manipulation—do not result in his being devoid of moral responsibility but result from his abdication of moral responsibility. Leonard abdicates his moral responsibility and his inability to form normal relationships and his susceptibility to manipulation by others results from that abdication.
This becomes more apparent at the ‘end’ of the film, when we see Leonard’s understandable, yet repellently nihilistic decision to go after Teddy, the corrupt policeman who has been manipulating Leonard for his own ends. Leonard chooses to go after Teddy—to the point that he will finally kill him (at the ‘start’ of the film)—to give himself a reason to live. Leonard was our moral guide through the noir morass of ‘humanity’ around him in ‘Memento’. One inhabited (t)his world with him—only now, with this moral (and not merely cognitive) shock, does one realize what a revolting monster one has been party to.[24] This exercise in unreliable (and potentially therapeutic) narration and ‘dialogue’ parallels Wittgenstein’s. ‘Memento’ is merely more clearly moral in purpose than PI.
Memento: A philosophical investigation
To recap: ‘Memento’ tempts one, as does Wittgenstein’s grocer example, into the thought that the problems—of meaning, understanding or, in the case of Leonard, a fully human life where he can form friendships, be able to trust others and be responsible for his actions—can be solved by ‘going inner’. That is to say, one is tempted to the thought that if one posits an inner realm then one will have provided sufficient conditions for the grocer understanding the note; and if one gives Leonard back his ability to form memories then one has sufficient conditions for Leonard to live a fully human life, forming friendships, trusting others and being responsible for his actions as a person. The way Wittgenstein’s ‘short’ and Nolan’s ‘feature’ are played out undercuts this move. All you do in making the move ‘inner’ is predicate of those same practices that they are inner; nothing is gained from the move, though much is lost. The move inner provides neither the necessary conditions for understanding nor the necessary conditions for living a fully human life.
Think again for a moment of the ‘Chinese Room’. This is a technique of ‘externalisation’. It is Searle’s rendition of this technique of Wittgenstein’s. That technique is, we hope, now manifest in the opening of PI (and in ‘Memento’).[25] But the way it works contains more complexity—and (temporary) complicity—than Baker and Hacker assume. One needs to become aware of the subtle, even devious self-deconstruction[26] of many of Wittgenstein’s ‘thought-experiments’, if one is not to miss his thought entirely. Take the ‘builders’, for instance, or take the ‘wood sellers’.[27] Wittgenstein’s ‘webs of deception’ precisely fool you into thinking that he is not giving you webs of deception. Baker and Hacker swallow the bait, but have failed to spot the hidden hook. (We return shortly to the question of just how fully ‘Memento’ too self-deconstructs.)
Words do not necessarily play a relational role of any kind. One need not relate the word ‘memory’ with some inner process any more than one need relate the word ‘red’ to a mental representation of ‘redness’. Our words having meaning does not require such a move. What does this mean for our thoughts on ‘Memento’? Well it is not that Leonard cannot relate the woman next to him in bed with a mental representation of her, nor that he has no memories to relate to Teddy that will tell him whether he is trustworthy or not. It is not this inability to relate sensory experience to mental representations and memories that is the problem. It is rather, we wish to posit, that the narrative of Leonard’s life has been radically disrupted, or even broken. The new narrative he is trying to construct leaves him vulnerable—recall when we as people normally set out on forming the narrative for our lives we are children and are cared for by guardians (parents, teachers, older siblings), indeed we are not generally considered responsible for our actions—the break and the resultant re-establishing of a narrative has for Leonard to be done without the benefit of childhood and the security that comes with it. Leonard is a child in an adult world (because not recognised as a child by others) but a child in the most adult of worlds because of his desire to avenge that which led him to his second childhood.[28]
As we have already hinted, Augustine is quoted, and a passage from childhood recalled in the autobiographical sections of his book for good reason. We have noted that the shopping trip example is bizarre and somewhat eccentric. Why do we, as readers, accept it as a valid example at all? In other words, what makes Wittgenstein’s example of the shopping trip appear, at least initially, plausible? Following Mulhall we take it that the example is in a sense mimicking children playing at shopping.[29] Think of an example of Cavell’s. You visit the newsagent with a young child—let us say your 5 year old daughter. You take your newspaper and give sixty pence to the child saying “Would you like to pay the man?” You hand sixty pence to the girl. She takes it in her hand, and then tentatively and somewhat shyly pushes her upturned fist forward opening it to reveal the sixty pence in front of the shopkeeper. As the shopkeeper takes the money the child looks up at you and, somewhat proudly, smiles; the shopkeeper looks at the child, smiles and says in a self-consciously friendly manner “Thank you very much”.—Has the child entered the shop and purchased the newspaper? What condition do they have to fulfil to be said to have done so? Do they need to ‘understand’?
We think they do need to understand. But an appeal to understanding here is not, as should now be clear, an appeal to an (a conception of ‘understanding’ as an) inner mental process. An appeal to understanding here is an appeal to the child’s enculturation, their full participation in a form of life, their ability to be-meaningfully-in-the-world-with-others. What being able to understand means here, is, learning to be, being guided in one’s attempts at being, and to be, with others, being trained to be, mimicking being. To understand, having understanding, is to be able to fully participate in the world with others. Of course there are a number of worldly as well as personal prerequisites for such a state. Leonard’s moral pathology has returned him to a stage where he needs to re-establish those personal prerequisites. He needs to regain his understanding. Only it is much more difficult because his lack of understanding is not induced by his status as a child—he is not recognisably a child—though it is the guidance, training, learning that is accorded to children that Leonard requires. He actually gets the opposite from those adults that surround him in the film.
The narrative of Leonard’s life has been subject to a radical break, and he is ‘forced’ into a second childhood, into the kind of position analogous to that gently and harshly characterised by Wittgenstein when he returns briefly to discussing Augustine in PI §32. But what, on our account, is the nature of the break? Has Leonard’s wife been murdered? Or is he a moral nihilist, because of the shock caused to his system (sic.) by his wife leaving him after the attack? Or is it all a result of guilt for the death of Sammy Jenkiss’s wife, which he was arguably directly responsible for (in that he told her that Sammy was a faker of memory loss, when in fact he wasn’t, and she killed herself at Sammy’s hands, as a result.). Is it even of any importance to decide?
It is interesting to note that the vertiginous expansion of possibilities as to what is really going on in ‘Memento’ comes at the very moment, almost at the end of the film (i.e. at the start of the action), when one finally felt close to closure. One has come to understand the nature of Leonard’s ‘condition’, one thinks. One has at last mastered what he has mastered intellectually, if not experientially. One’s complacency is undermined by Teddy’s ‘denouement’ speech; and one begins trying to figure out all over again ‘what the story’ is.
So: which story is correct? We don’t know. The film is puzzling, though less ‘systemically’ so than (say) ‘Marienbad’. But the counter-picture—to Leonard’s own—that we think makes the best sense of the film draws on Teddy’s speech, on what Leonard sees in his mind’s eye during their dialogue, and expands on what we have already written concerning how Leonard’s difference from us ‘normals’ may be less neuro-cognitive and more moral-psychopathological.
It doesn’t matter if you find our story of the story less than fully convincing. Indeed, it is arguably even better that way! For it is enough that there is a real possibility that Leonard is not neurologically abnormal after all. Against those who want to ‘go inner’ to prove that Leonard is abnormal, it is enough to show that this ‘move’ actually takes us nowhere. The heart of the film’s self-deconstruction, is this, alluded to though not fully explicitated by Teddy in his coruscating speech challenging Leonard: how has Leonard mastered his condition? How can he know, as a matter of intellectual and historical fact, about his own condition, given that he actually does have such a condition?[30]
Teddy suggests to Leonard that his wife survived the attack. That Sammy Jenkiss was a faker, not someone with an incurable neurological condition. That he, Leonard was the one who had ended up in an asylum for a while. That Leonard killed John G. years ago, after Teddy helped him find him. And that Teddy has been using Leonard to kill people for his, Teddy’s, corrupt ends, ever since. Teddy more or less insinuates then an identification or confusion on Leonard’s part between himself and Sammy, and more or less insinuates that Leonard’s condition may not be at all what he (Leonard) endlessly tells people it is.
Leonard is unsure what to think, now. Even his memories of his wife, seemingly so secure, start to disintegrate under the strain. (He learns the hard way that memory-scepticism cannot be confined.) The possibility that we think requires serious consideration at this point is that Leonard doesn’t actually suffer from an incurable neurological problem at all. Rather, as his memory gradually returned or threatened to return after the trauma of the attack, he fought it off: Because of the unbearable psychological pain that was caused him not so much by what had happened in the attack but by his wife being unwilling to bear him after the attack, and leaving him. He reconstructed his past such that his wife had indeed been not only raped but killed, and that he, the perfect caretaker and lover, was on an eternal quest to avenge her, a quest that in fantasy would then end with him sitting smiling in bed with her, with a tattoo saying “I did it” (i.e. killed John G.) emblazoned on his chest.
His constant mental work – his distinctive innerness, his interminable deeply anxious doubts and his bombastic or dogmatic counter-assertions to those doubts – then stands in the way of facing the facts. He thinks; therefore he is not aware of something that he finds unbearably painful, something that must always be slightly outside of the frame of this film that we journey through with and through him. The deceptive aspect of the denouement to the story we are now presenting, that comes at the end as we watch the film, but (as in PI) at the beginning of the ‘narrative’, is perhaps the story behind Teddy’s story, a story that (re-)structures everything that ‘follows’ it. One is forced to rethink (through) the story/stories that Leonard has constructed (perhaps, temporally, in response to this story of Teddy’s). (Compare PI: as the text proceeds, one is forced—at any rate, if one is attentive to the way the text is, so carefully, constructed, then one is forced—to reassess the thought-experiments and apparent-statements by Wittgenstein that one had hoped to rely on. This happens in the Tractatus in one gigantic explicit manoeuvre. In PI, it happens through continual piecemeal reflection: on the ‘grocer’s shop’, on the ‘builders’, and so on.) Leonard’s stories, built in reaction to what (at least according to Teddy) really happened to him, start to fall apart on him, and on one, all over again. The final beauty (and tragedy) of ‘Memento’ is watching this happen—and then watching how Leonard stops the falling apart, and restarts himself on a freshly destructive and denying path.
Wittgenstein’s pseudo-stories, and Wittgenstein’s pseudo-theses, are not meant to hold up. Those who think that they do, those who like for example to extract ‘grammatical truths’ from Wittgenstein’s work, have negated the spirit of his philosophy. ‘Wittgenstein’ is an unreliable narrator. You learn from him by bringing to consciousness what you are inclined to think about ‘philosophical matters’, by considering alternatives to those inclinations, by seeing those alternatives in turn collapse when you try to make them bear weight, and by as a result being compelled neither by the picture you started with nor by seemingly better, smaller etc. alternative pictures.
The denouement
to ‘Memento’ gives one the surprising outlines, outlines we have now filled in
a little, of an alternative possible plot-picture to that which one had been
assured of by one’s narrator. His
philosophy, which is dogmatically ‘anti-Cartesian’,[31]much
like Ryle and most ‘
Whether the trauma that ‘broke’ him was his wife leaving him or his ‘killing’ Sammy’s wife, the upshot is the same: our tentative diagnosis of Leonard is that he is a ‘victim’ of a hysterical amnesia resulting from the need to suppress the memory of a trauma and resulting in a continually-maintained bad faith. Thus a closing reason why ‘Memento’ is film as philosophy is because it ‘ends’ by presenting one with an intellectual possibility not familiar from the Diagnostic and Statistical Manual nor from the likes of Karl Jaspers: namely, a form of bad faith undreampt of in Psychiatry’s and Phenomenology’s and Existentialism’s philosophy. It presents a new possibility in (the philosophy of) psychopathology. Leonard has caught himself in a huge, awful and useful web of deception. His ‘condition’ leaves him terribly vulnerable to others’ manipulations, and he experiences almost nothing but such manipulations in the course of the action of ‘Memento’. But in effect he takes this to be worth it. His tragic position is less bad than the alternative: facing up to what has happened to him, to what he has done, and to what he has made himself into.
Rather than maintain himself in such a constant state of anxious and delusional denial, Leonard could take a different route. The therapy offered by the film itself to its viewer, which is somewhat like the therapy offered by the entirety of Wittgenstein’s text to his reader, is we think encapsulated in the beautiful motion of the camera as Leonard’s car swings into and out of the yard of the abandoned building where the film starts and ends, and by its curve as he drives away ‘into the future’. Living in the present, which requires acknowledging but not being possessed by the past (as those who obsessively remember it or obsessively ‘forget’ it are), would involve appreciating that beauty, which is visible to us as viewers, but which Leonard cannot see, locked as he is into an obsessive inwardness of thought and a bizarre externalisation of memory. As the car curves away down the road, we are given a point-of-view shot, for perhaps the first time in the film. We see the scene as Leonard sees it. It is beautiful; but he can see none of that. He has trapped himself in the present, which is the very contrary of living in the present.
‘Memento’, more than the fabulous short story ‘Memento Mori’ from which it was adapted, and far more than Philip Kerr’s interesting ‘Wittgensteinian’ detective-novel, ‘A philosophical investigation’, offers we think a philosophical investigation in the spirit of ‘Philosophical Investigations’. We have investigated ‘Memento’ philosophically, and found that it not only shows but also is a philosophical investigation; one that depicts the soul of someone whose problem, perhaps, is actually that they cannot bear to remember, not that they cannot remember to forget.
[1] We make no claim as to what type of dualism Descartes held. Much recent scholarship (see for instance the work of Gordon Baker and Katharine Morris) has done much to disabuse us of the ritualistic dismissal of Descartes as a crude substance dualist. When we use the term Cartesianism here we do so to indicate a pervasive conception of the mind, taken as distinct from the person, and animating that person (body). This therefore includes not only traditional self-proclaimed dualisms such as substance, attribute and property but also many ‘materialist’ conceptions of mind which deny their dualistic nature; we have in mind here for instance Jerry Fodor’s psychosemantics and Ruth Millikan’s biosemantics, as well as any theories which draw an analogy between the mind and computer software.
[2] When we talk of the ‘standard reading’ we have in mind, primarily, that offered in the four volumes of the Analytical Commentary on the Philosophical Investigations (hereafter ACPI i-iv). Similar readings of PI are offered by Hanjo Glock, Paul Johnston, Anthony Kenny and Severin Schroeder. It is important to note that Gordon Baker was co-author of the first two volumes of ACPI only. After this time he went through a radical change of mind. His later work, published between 1991 and his death in 2002, advances what he terms a ‘radically therapeutic’ reading of Wittgenstein’s PI and, on his own account, has strong affinities with the readings advanced by Cavell, Conant, Diamond and Dreben. See Baker (2004) Wittgenstein’s Method: Neglected Aspects. It may strike the reader as strange to therefore refer to ‘Baker and Hacker’ critically throughout. However, we follow Gordon Baker who in his later work referred to his own early work with Hacker in this way and used it as a stalking horse for his own new reading.
[3] So far as this matter can be specified, we employ throughout the following usage of diacritics. Double quotation marks are used for quotations. Single quotation marks are used as ‘scare quotes’, when mentioning words and for quotes within quotes unless the original employed double quotes here also. Italics are employed by us for emphasis, except for within quotations; here, in order that we do not confuse our emphasis with that in the original, we embolden for emphasis. Any instance of single or double quotation marks or italics within a quote is from the original.
[4] Gottloeb Frege Foundations of Arithmetic (FA), Betrand Russell & Alfred North Whitehead Principles of Mathematics (PoM), and Ludwig Wittgenstein Tractatus Logico-Philosophicus (TL-P).
[5] It is of considerably more than mere passing note (given recent exegetical disputes) to observe that while Baker and Hacker argue the case for, and take themselves explicitly to have established the case for, Frege and Russell operating within this Augustinian paradigm, they—equally explicitly—stop short of making the same claims regarding Wittgenstein’s TL-P. They write, “it would be absurd even to try to give here a definitive proof that the Tractatus conforms to an Augustinian picture of language. Instead, we shall simply show that this is a plausible view of the book. […] The exclusion of all matters of ‘psychology’ differentiates Wittgenstein’s logical atomism from Russell’s. It also makes it pointless to search in the Tractatus for many of the theses characteristic of the Augustinian picture” (ACPI i pp. 58-59) Nevertheless, they do conclude by asserting that Wittgenstein was working within the Augustinian paradigm, following Frege and Russell.
[6] Paragraph (a) of PI §1.
[7] Mulhall’s (2001) reading of PI 1 is what first brought to our attention the possibility of the topic of this paper. The critique of Baker and Hacker is ours as is the identification of the relational nature of language being, at root, what Wittgenstein identifies as holding us captive.
[8] It is important to note the “must” here. Of course words often refer to things in a trivial non-controversial sense; it is just that this is not a condition of their having meaning.
Also, we take it as unproblematic that “belonging” is a form of relationship. Belonging denotes a relationship holding between the possessor and possessed. Certain words are possessed by certain types of use.
(For more on such relationality, and for our analysis of a mistaken version of it, mistaken in a way that closely parallels Baker and Hacker’s mistakes, see our Review of two books by Joseph Margolis, forthcoming in Philosophical Quarterly).
[9] Thus our central criticism of Baker and Hacker could helpfully be phrased thus: they reify ‘the Augustinian picture’ as an ogre to be combatted and avoided at all costs. Unfortunately in mis-diagnosing what is at fault in that picture, they themselves recommend what is merely a variant of it!
[10] Note Wittgenstein’s remark in the Preface to the Tractatus: “This book will, therefore, draw a limit to thinking, or rather -- not to thinking, but to the expression of thoughts; for, in order to draw a limit to thinking we should have to be able to think both sides of this limit (we should therefore have to be able to think what cannot be thought). The limit can, therefore, only be drawn in language and what lies on the other side of the limit will be simply nonsense.” Read and Crary’s The New Wittgenstein (Hacker’s rejoinder excepted, of course) could fruitfully be viewed as an elaboration of the ‘thought’ of McDowell’s just cited alongside this prefatory remark that serves as a very fruitful guide to interpreting Wittgenstein’s work as a whole.
[11] Of course we do not wish to restrict ourselves to pictorial mental representations here. The syntactically structured ‘mentalese’ of Fodor’s psychosemantics and Millikan’s biosemantics will do just as well.
[12] Here our account departs from Mulhall; as far as we are aware he does not make this observation and, by extension, critique of Baker & Hacker. The point regarding the relational view of language is our own.
[13] It is this thought that informs many (would-be) Wittgensteinians’ appeals to a somewhat reified conception of grammar and grammatical rules as a way of settling philosophical disagreement and dissolving philosophical problems. It is this easy brand of ‘Wittgensteinianism’ (Rylensteinianism) that has done much to marginalise Wittgenstein’s work in contemporary philosophy.
[14] When we talk of ‘relating’ and ‘relational’, we are talking of external relations, i.e., of (genuine) relations. Of course if we were to say “all words relate either internally or externally” then there is no problem, at least in the sense that we state nothing. There’s a sense in which to identify a relationship as internal (such as between a ‘mental illness’ and its central symptoms) is to say that there is no relationship at all, that they are (part of) the same thing. (For further explication, see Denis McManus’s forthcoming book on the Tractatus, The Enchantment of words).
[15] For example, ‘Memento’ does not play the relatively cheap philosophical trick of ‘The Matrix’, which simply tries to enforce Cartesian doubt (updated through roughly Putnam’s ‘brain-in-a-vat’ thought experiment) on its audience. ‘The Matrix’ first attempts to insist that you are just like Neo, the film’s ‘meditator’, and then after a while, in an imperial gesture that attempts to negate its initial paranoid hypothesis, reveals the true reality behind appearances. ‘Memento’ has no such imperial project. While one tries to identify with a film’s narrator, ‘Memento’ starts from the presumption that we are learning about someone’s state of being who is very different from normal. (This presumption later gets questioned, such that the film gradually becomes a kind of dialogic experiment in the viewer’s trying to figure out where they stand in relation to ‘Memento’’s narrator. See also n.16, below)
[16] As we shall see, this set of questions too eventually gets trumped, as one starts to wonder whether Lennie isn’t a lot more like you and I, cognitively or epistemically speaking, than he lets on or (alternatively) than he realises.
[17]
See Baker (2004) Chs. 1,3,8,9. Also Hutchinson and Read (Forthcoming) ‘The
Fundamental Significance of PI §122.
in Moyal-Sharrock (ed.) Perspicuous
Presentations: Wittgenstein’s Philosophy of Psychology.
[18]
It is worth recalling that Wittgenstein is frequently criticised, even by some
of his ‘admirers’ such as ‘
[19] Though in this case the end is just the beginning, as it were.
[20] He is like a child. Here, we think that Cavell’s (and Mulhall’s) remarks on the crucialness of seeing that the person who ‘shops’ in PI 1, like many of the people in PI, is implicitly (like) a child are very salient. We expand on this point below.
[21] Furthermore, isn’t it true that, as Leonard says, memory is unreliable? Leonard can helpfully be seen as Empiricism, practiced, writ large. Everything is externalised, everything verified. ‘Just the facts, sir’ could be his motto. It seems absurd to suppose that this would be better, that it is better not to be ‘encumbered’ by memory, as Leonard at times claims. Yes, it is absurd. But, as we saw Mulhall in effect remarking, above, it is no more absurd than Cognitivism.
[22] The Blue and Brown Books, p.3.
[23] We place ‘condition’ in scare quotes now as it is not really a condition but merely an external representation of the internalist picture of mind. As will become apparent neither do we think it is a condition in the way in which Leonard says it is in the film. There are many clues in the film to point in the direction of another conclusion. [We discuss this in more detail below.]
It is worth adding that there is one other thing Leonard tends to tell everyone about: Sammy Jenkiss. We deal with the significance of this toward our conclusion, below.
[24] As we elaborate below, Leonard, on our interpretation of ‘Memento’, is normal, except for his pathological moral difference from us. As one is sequentially sympathetic to and then appalled by all the other characters in ‘Memento’, so one’s identification with Leonard, even through one’s sense,
In other words: one identifies morally but (unlike in for instance ‘The Matrix’) not cognitively-epistemologically with the film’s main protagonist, during the film. But our suggestion that ultimately Leonard is to be distinguished from you or I morally but not cognitively or epistemologically brings with it something potentially uncomfortable: our long journey with him forces us to confront our own possible ‘shadow’ immorality or psychopathy.
[25] Another
splendid instance/discussion of it can be found on p.26 of The Voices of
Wittgenstein (by Wittgenstein and Waismann, edited by Gordon Baker (
[26] We do not wish to imply any particular identification with Derrida (or his followers) in our employment of this term. We employ it in accordance with ordinary usage.
[27]
[28] In this regard, Lennie parallels the replicants of Blade Runner. See Mulhall’s exemplary reading for detail.
[29] This is also a prominent theme in Stanley Cavell’s readings of Wittgenstein, something he returns to again and again.
[30] It is worth bearing in mind that real individuals who have suffered complete long-term memory loss from after a certain point in their life and inability to ‘make new memories’ do not come to know about their condition. They are always at best confused, and tend to deal with this confusion either through a continual desperate effort to understand their condition that makes no progress or through a lack of interest in anything except the present and the distant past (including a lack of interest in their condition).
[31] Near the end of the movie, Leonard remarks, in a moment of self-doubt when he will not self-doubt, in a moment when the unpalatable reality of his existence threatens to break through all his resistances to it, that “The world doesn’t disappear when you close your eyes.” He then opens his eyes, confirms that the world is still there, and triumphantly drives on – toward murder.
[32] Teddy’s role here, somewhat like that of Hannibal Lecter in another fine philosophical film, ‘Silence of the Lambs’, is that of an unreliable but nevertheless in important part true educator; a kind of morally-repellent Socrates.