Issue #18 November 2018

Merleau-Ponty, Education, and the Meno Paradox

James Paulius (@jamespaulius)

— Plato, Meno, 80d

— Merleau-Ponty, 2005, p.33

(1) The Problem

The Meno Paradox looks like a simplistic sophism: You can’t come to know what you don’t know because you need to know something about a thing in order to come to know it (we can also substitute ‘understand’ here for ‘know’, as well as ‘comprehend’, etc). But what could be more natural and mundane than learning something? One issue that drives Merleau-Ponty’s phenomenology is that previous approaches to the mind-world relation are embarrassingly unable to give an account of what is wrong with the Meno paradox. That is, both the empiricist and rationalist projects have not given a sufficient account of the dynamic relationship between mind and world such that developmental or learning processes can be understood.

The problem of the Meno paradox seemingly emerges from all-or-nothing thinking. We know ‘all’ of very little, but a ‘little’ of many, many things. The issue arises when we try to account for this ubiquitous ‘partial’ character of knowledge that stands witness to, and is the condition of possibility of, learning as such. What must the relation of the mind and the world be, how must an ‘idea’, or concept, be constituted and structured such that, the world can present something initially unfathomable to the mind, that the mind then comes to ‘fathom’ in degrees?

The intuitive response is that which is unfathomable comes to be interpolated into that which has already been fathomed. Learning is to reduce the unknown to the terms of the known. But here Meno’s paradox lies in waiting: if that’s the process, then where does the ‘that which has already been’ come from? Furthermore, even if it is granted that there is ‘has already been’ understandings (via innate structures or prior experiences), how are these to get a purchase upon something not understood? If this former understanding can ‘grasp’ the edges of the unknown phenomenon, then, why is it unknown? If it can be subsumed by the understanding, then why hasn’t it been? We seem to be missing a middle and mediating term, or stage.

For the constructivist pedagogy of Piaget the issue is in terms of how a cognitive, conceptual structure is ‘upgraded’ to incorporate greater complexity than it is initially capable of. The issue is one of ‘bootstrapping’. As Fodor puts it:

On the basis of this, Fodor, with Chomsky, saw the need to hypothesize innate structures that are then ‘activated’ or ‘articulated’ by experience, rather than acquired through experience. In form, this is the Kantian move to the Empiricist’s deadlock concerning a priori knowledge. In fact, this drama from the history of philosophy still animates the discussion.

The scope of the paradox

It’s important to determine the scope here. Much of our day to day learning does not exhibit the Meno Paradox because we already have the conceptual structure in place to ‘subsume’ new elements. Bereiter (1985) and Daniels (2001) give the example of learning to use a new washing machine. Even though the particular process is unknown, as regularly functioning adults we possess the conceptual schema such that the particulars of the process ‘fall into’ a pre-existing structure, as though we are just filling out a form. We also already possess certain knowledge that can act as an interpretive heuristic (the numbers around the dial, being in increments of 10 initially, but then switching to 15, are no doubt wash time durations expressed in minutes, and not prices, or people’s ages).

But then think of the child learning to read for the first time. If, as literate adults, we encounter some foreign script, we understand it generally within the schema of writing, there is just the ‘form filling’ component of the meaning of this or that symbol (as vast as that component is). But an illiterate child needs to come to comprehend this schema without a model having been given in advance. They come to read without knowing at first what reading is like and what it cognitively involves. This requires the bootstrapping of cognitive abilities radically different to the ones they bring to the table.

Bereiter (1985) gives an example from observations of children independently working out a more efficient arithmetic algorithm. Initially children approach an addition problem like ‘4+3=?’ by counting out 4, then counting out 3, then counting the combined set, thus using the number line three times to get the result. However, eventually most children independently ‘realize’ that they needn’t count to 4 initially, as 4 is already given, so they begin the number line at 4, and just continue counting the integers for the addendum. Of this development he writes:

He continues:

If something as sophistic sounding as the Meno paradox still haunts us, and generates discussion, then something has gone terribly wrong with either our conception of the mind’s relation to the world, or knowledge, or its acquisition, or some combination of some or all of these.

A consideration from the phenomenology

All of this would be much more easily solvable if the phenomenology revealed great voids of knowledge in our experience of the world to us, with spinning, whirling, perplexities twirling around everywhere, but unfortunately the precise opposite seems true. The world appears for the most part to be just obstinately and uninterestingly there. We don’t walk around immersed in confusion. This is despite the fact that we, today, can intimate the general degree to which we don’t understand what is in front of us — we are aware that there is much understood by specialist others, but we don’t know what that understanding consists in.

Take, for example, mirrors. Now most of us probably remember some diagram or explanation giving us some fuzzy idea that mirrors work by refracting light in some super-efficient way. But the point is, what is going on with the physics of mirrors for the most part doesn’t trouble us at all. For as long as I can remember I have stood in front of them in the morning and watched myself brush my teeth, with a bored expression on my face, never finding it astounding that here before me was a surface that replicated the entire world in pristine, exact detail, yet completely reversed. Mirrors are just there.

Thus, from the phenomenological perspective, the issue in learning is not that the learner is ignorant, that there are great swathes of unfinished regions and gaps in the maps of their understanding, but, rather, that the world, for the most part, appears as quite comfortably simple, easily resolved, and perfectly mundane. We very rarely encounter complexities that outstrip our ability to digest them, in the terms with which we are familiar.

We see this when learning the phonetics of a new language. One of the issues for learners is when the target language makes a phonemic distinction that the learner’s first language does not. There’s always challenges in grasping new pronunciations, but the issue of learning a new phonemic distinction is compounded by the fact that the learner cannot hear the distinction (Strange & Dittman, 1984). The phonemic structure of our language imprints upon us certain distinctions that are crucial, for example English’s [l]-[n] distinction (million-minion), whereas others are left unarticulated, left as vague regions. Crucially, this is not just a question of pronunciation, but a question of differentiating in perception as well. When we attempt to learn a language that does make crucial distinctions of sound clusters our native language leaves unarticulated, we are met with the slightly Monty Python-esque scene of the student feeling as though they are hearing the teacher say the same word twice, they then repeat it twice, with the teacher sometimes saying ‘correct’ and sometimes ‘incorrect’.

In this case, the process of learning a more complex structure (a phonemic distinction on a previously homogenous ‘sound space’), is tied to an experiential change in the way the world is perceived. It’s not the case that some previously concealed, or internal, area has opened itself up for inspection, but that the very thing perceived has separated itself into two things. And here the paradox lies waiting: how can something experienced as homogenous come to contrast with itself to create two distinct regions that are then experienced?

James Paulius (@jamespaulius)

(2) Traditional Solutions and Their Shortcomings

Though Merleau-Ponty does not mention the Meno paradox directly, his work in the expansive, multi-part introduction to The Phenomenology of Perception, can be seen as developing a directly analogous issue, as well as diagnosing its source. The introduction offers an expansive critique of rationalist (or ‘intellectualist’) and empiricist approaches to perception. Here I trace that critique as it is relevant to initial, intuitive responses to the Meno paradox. These responses fall with the projects they are based on. What Merleau-Ponty offers us, therefore, is a clear elucidation of just how unequipped the major streams of philosophy are for dealing with the paradox.


The Fodor-Chomsky solution to the paradox, as hypothesizing innate cognitive structures that outstrip any particular instance or instantiation, can be seen as the rationalist one: the world is carved into its structures along the lines of our innate cognitive faculties. This was also the route that Plato took in the Meno itself — learning as a kind of ‘recollection’ (what is learned is native to the subject, not received from the world). Perhaps the most elegant and ‘baroque’ use of this manoeuvre can be found in Kant’s transcendental arguments. For example, Kant argues that we could not derive (learn) the concept of space from experiencing proximities and distances, because taking these as ‘spatial’ relationships presupposes an understanding of space. Just like with the children and arithmetic, where the notions of cardinality and combining sets are presupposed by, yet more complex than, the development of the more efficient addition algorithm, the concept of space in general is presupposed by, and more complex (general, abstract, ‘total’) than, the recognition of particular spatial relationships given in experience. Kant’s philosophy moves by identifying these ‘Meno’ like structures and then resolving them by positing that, though the content of our experience is provided through sensation, the form our experience takes is the form of our cognition. Just like with learning to use a new washing machine, we bring a form (in the sense of a survey) to be filled out, Kant argues when we approach the world we do so with forms that condition the way that sensation will be given to us. So, the children grasping cardinality and sets have just realized something about the way they structure their experience that was always there — the world was given to them in cardinal and set theoretic terms already. Thus, the improved algorithm can be seen as an analytic simplification rather than a synthetic increase in complexity (cardinality being a dimension of the concept of an object in general, not something ‘added’ to a deficient concept to make it more complete).

Learning becomes, then, a process of paying attention. When one directs their attention to an object, one begins unravelling the intellectual structures through which we, in consciousness, constitute the object.

To go back to our language learning example, if learning a particular language is an ‘activation’ of a segment or possible structure within an innate, universal ur-language, through paying careful attention to the structures of the language, the phonemic distinction will be discovered to be necessarily entailed by the structures themselves. You just cannot have a language that makes a distinction here, here and here, and equivocates this, that, and that, without it, thereby, requiring a distinction between these two sounds. That is not to say we always have enough knowledge of actually existing languages, and the ur-language from which they are supposedly derived, to be able to make such predictions, only that languages are structured with necessary limitations and entailments that fall into the possibilities demarcated by the ur-language. The cognitive power of ‘judgement’ is then brought into the experience to explain the change in the phenomenon as it is experienced (the phonemic distinction emerging for perception). For the rationalist account, judgement “is often introduced as what sensation lacks to make perception possible.” (Merleau-Ponty, 2005, p.37)


The other, contrary approach, is the empiricist one. Here, learning occurs through making associations through repeated observations, and the projection of memories (past observations). The source of knowledge here is very much the world, and the stable patterns that exist within it, rather than in the mind. For example, Bereiter (1985) has argued for a number of possibilities that could allow the kind of ‘bootstrapping’ required to resolve the Meno paradox. Many of these involve feedback loops; autocatalytic processes of experience, action, and result. One partial solution of his is to think in terms of initially semi-random attempts and mistakes hitting upon efficient solutions by accident, which then, through repeated experimentation, become self-enforcing.

So, in the example of the children adding, the child might realize they have miscounted the addendum (or dropped one of the blocks they were counting, etc) and start again by going back to the beginning of the second counting procedure — that is, starting from 4. Over time, the intermittent step may become contracted: “4+” changes in meaning from “count to 4 then-“ to “start from 4 then count-“, because every time a mistake or accident has created the opportunity for it be interpreted in the latter way, the result has come good (positive feedback).

Now, Bereiter is neither advocating for a pure empiricist solution, nor is he putting forward the above as the only process that may be going on, but we can see the general form of this partial solution, and how it differs from the rationalist one. The key thing is the developing associations between experiences. Where previously the operation of addition was associated with a procedure of counting three individual sets, a fortuitous experience allows a ‘better’ association with a new procedure. Prior to any experience at all that could be called on to associate with the present one, experience must, therefore, be a chaotic sensory flux (the mind as tabula rasa). At the root of this is not rationalist’s innate structures of cognition, but a chaotic maelstrom of sense data, from which patterns are gradually isolated by repetition, forming ideas of greater and greater abstraction (by the ‘extraction’ of ever more minute repeating patterns). So, to learn is to associate patterns of sensation, with each other and with memory. For Merleau-Ponty, these two assumed cognitive powers (association and the projection of memory) form for the empiricist what judgement and attention do for the rationalist: assumed structures that ultimately do the heavy lifting when it comes to explaining the mind’s relation to the world, and, by association, the question of knowledge, and thereby, ultimately, of knowledge acquisition.

Thus, from the history of philosophy we have two ways of solving Meno’s paradox.

James Paulius (@jamespaulius)


Merleau-Ponty is categorical that neither of these approaches can even explain what they have been developed to explain. Both of them assume the very relation they are supposed to articulate. Despite their opposition, where both agree is that sensation is atomic, and devoid of ‘sense’. If the world is to appear to us in consciousness as being ordered, structured and related as such to be knowable (and learnable), then, it is assumed, that it is in virtue of some power of our cognition, be that association, memory, attention, or judgement.

However, the argument that Merleau-Ponty endlessly deploys is that if this flux or input of sensation is to be meaningless, by what miracle can the mind associate, categorize, or stamp meaningful connections and concepts onto/within it without this process being arbitrary (which it demonstrably is not)?

Let’s assume the empiricist approach. We say that babies don’t ‘know’ anything in lieu of the experience they have not had yet — so the world to them is a kind of chaotic noise of sensations, not a world of a stable, recognizable objects. Now, we here as adults also have our nervous systems bombarded with the same signals, sensations, but these sensations are of a recognized sort, so we associate them with past sensations via memory: the result being that the world is not chaotic noise, but stable and comfortably arranged into classes and familiar objects. Here a table, there a pot-plant. But, did you see the sleight of hand there? “These sensations are of a recognizable sort”! It’s in virtue of their being recognizable that I can make associations, such that here is a table, not a kaleidoscope of possible objects being tested by my consciousness for congruence. But if the sensations are recognized, what is the appeal to memory and the further power of association adding? It has been snuck back into the beginning of the process, and needlessly redoubled.

Furthermore, for association to function in its role of uncovering identities among heterogenous sensations, that is, forming classes of things for knowledge, then it is unclear how this is to come about without assuming that the association works by connecting ‘what is significant’ about the class of things in each of their encounters. But association is what is being used to develop our ability to recognize what is significant. Merleau-Ponty is clearer about this difficulty in The Structures of Behaviour (1963) when he is critiquing the behaviourist model of learning. It’s an experimental fact that subjects in conditioning experiments (be they rats, monkeys, or people) very quickly ‘extract’ from their behaviour just those salient actions that result in the reward response. But if all that is occurring is a series of actions linked to the positive outcome, such that the positive stimulus associates the actions together in a chain (as the behaviourist model suggests) then how is the subject to ‘extract’ only certain relevant actions to be enforced, and discard others? If the rat turns and walks around its cage, sniffing the corners before stumbling across the button which it sniffs, then nudges, then presses, receiving food which then conditions the prior behaviour, explaining repeated re-pressings of the button, why don’t we also see repeated pacing, sniffing, etc also being conditioned? The simple answer, the rat works out that only a small set of the previous actions are relevant to getting the food. But how does it do that if all there is to learning is the association of one path of experience, procedure, etc, with a result? It would seem we would need to grant an interpretive power that far outstrips what is being explained with conditioning, and largely makes it irrelevant. If the rat is extracting meaningful elements of its experience, and just those particular movements needed to generate a desired result, all the talk of conditioning is superfluous: the rat has all the behavioural resources it needs to reliably get food. It does not need to be additionally ‘conditioned’.

(Truthfully, Pavlov was aware of this issue, and hypothesized a suppression mechanism, but Merleau-Ponty demonstrates at length just how ad hoc this addition is, see the section Pavlov’s Reflexology and its Postulates in The Structures of Behaviour for the full discussion).

Another way to put it is that if we form knowledge by associating repeating sensations, we need another faculty that could limit this associative power within ‘bounds’ circumscribed by an extra logic of significance, otherwise all sensation would wind up associated with itself in a giant ‘block’. Once we have posited such a faculty, that could attend to the significance of objects, we have presupposed what association hoped to explain (the significance of certain objects and behaviours), thus demonstrating that association in fact can’t explain what it hopes to.

The same difficulty awaits the rationalist approach when we ask where error comes from. Obviously, it’s some miscarriage of judgement, or deficiency of attention (these being the only resources available). But if we were to ask what is it about normal, clear, distinct, and ‘right’ judgement and attention, and the mad, hallucinatory, judgements of the insane, we fall into a similar question begging that awaited the empiricists:

That is to say that if the sane and the mad both make judgements, the only thing to separate these judgements into the true and the false is to appeal to something beyond cognition, already structured (as if by a judgement). And since we are forced to make this appeal anyway, that the sensory given is given as always-already structured, then what exactly is an additional faculty of judgement in the rationalist model adding?

It’s here where Merelau-Ponty offers the source of both the empiricist and the rationalist error: they both assume that ‘the world’s’ contribution to perception is atoms of sensation that are not sufficient to our experience, and thus require some subject-side organization, be that the stamp of judgement or the work of association. However, the initial assumption being made, it is contradicted by the supposed work upon it. We find the work needing to be already done so that it may begin! In modern parlance, this issue persists in talk of top-down and bottom-up processing, where the ‘bottom-up’ activity of nerves is assumed insufficient to the task of constructing experience, so needs to be processed additionally form ‘up-high’. But, the two domains, thus separated by a methodological assumption, need to be remarried in a unified experience — that is, the methodology needs to stick back together what it is cleaving in two.

What Merleau-Ponty’s critique suggests is that this question-begging that goes on in the rationalist and empiricist accounts of perception, and ultimately knowledge, based upon the assumption that the world is given as an inherently meaningless material in need of cognitive structuring, is perhaps the very source of the Meno paradox. What is his way of avoiding this?

James Paulius (@jamespaulius)

(3) Merleau-Ponty’s Solution; Habit and the Body

Development of habit

Let’s return to the phenomenological setting of the Meno paradox mentioned in section (1): the world appears to us as a mostly simple, already interpreted kind of affair. We do know that the explanations and discussions and problems of the things before us can fill volumes upon volumes, but this ‘possible knowledge’ doesn’t appear to us as a void, like a locked box in a movie surrounded by questions as to its contents, or missing or gappy sections in our ‘world projection’, or a ‘nausea’ of the concatenation of Being . The world is just there, and is mostly familiar, mundane, and seemingly ‘known enough’, despite us knowing very little about it.

Should Merleau-Ponty hear us speaking in this way his response would perhaps be to pose the question: who or what ‘knows enough’ of the world before us such that it can feel, for the most part, effortlessly familiar and mundane? His answer: the body. The room around me does not present itself as a series of geometrical, historical, scientific, philosophical, geographic and economic questions that I am unequipped to answer, but first and foremost as a thing to be moved through. It appears before me with this ready-made significance. And insofar as moving through it presents no issues to the silent intelligence of my body, that somehow coordinates itself and transports me around, it appears to me, to this very same body with eyes that see it, as perfectly ‘settled’ for its purposes. This ‘settling’ for bodily proficiency Merleau-Ponty identifies as “habit”. As Talero (2006) writes:

But the room is not only a ‘thing to be walked through’. Here is a chair to be sat on and a desk to sit at and a computer on which to type. We’re used to thinking of the skill of typing as something physical and habitual. One practices and then one’s hands just know how they need to move in order for the desired words to appear on the screen. I am not astounded by my hand’s quick movements, because they are not astounded by them. Coiled into them is the heavy sediment of habit.

It would seem that my hands, through habit, know where the letters of the keyboard are. But can we think of this ‘knowledge’ in terms of being propositional, or rationally justifiable?

If it could be formulated in detachment of the actual effort of bringing it about, then a number of things should hold true that don’t: I should be able to spell “empiricism” correctly with a pen, like I can when I type. I should be able to report which finger I use for each letter of the keyboard. I should be able to list off the bottom row of letters of a keyboard. The fact that I can’t seemingly do any of these things, but none of that gets in the way of my typing, points us to a kind of knowledge that is not remotely anything like ‘justified true belief’, or the subsumption under a concept.

However, this is where our appeal to habit usually ends — in the movements of the hands across the keys forming the desired words. But is this really what it is like to type? To desire a word and then engage my hands in creating it? As someone who is typing at this very minute, I can tell you the work of ‘automatic habit’ coiled up inside my hands seems to be driving the process much more than a ‘loyal tool doing the bidding of the mind’. It is hard for me to determine at what point of the process, if I want to think of my hands as habituated-through-practice tools, this habituation cuts off. It would seem to me, observing them now as I type, that my hands are also habituated in the use of grammar insofar as they happily bounce around contributing the correct prepositions and tenses (the linguistic rules of which I can tell you very little about). And where does this end? It seems my hands are also habituated into the act of writing an essay on the Meno paradox. I may have ‘deliberated’ over an outline, but the writing itself is carried along by their inner and silent intelligence, even when that entails observations, interpretations, examples and arguments that ‘emerge’ as opposed to being pre-planned, such as this lengthy paragraph you are reading now, which my hands seem anxious to type out at length (perhaps so they can get the recognition they finally deserve).

Or think of encountering a friend on the road. You raise your hand in a wave. A habit, you admit. As if to announce the talking distance has been reached, out comes the “Hi!”, perhaps another habit. You’re asked how you are. Out comes whatever habitual response you usually give, as if by itself. Then some standard back and forths. Maybe you’re still willing to grant these remarks on the weather as the product of habit too, or maybe you think something else has now taken over. But I would ask: what has taken over? At what point in the ensuing conversation does something like ‘conscious mediation’ take over? At what point are ‘the desired words being selected’? Or rather are the words just there, discovered as you say them, responding, and nodding and replying. The familiarity of the friend, and the familiarity of this conversation, that’s different every time it is conducted, but neither of you needing to spare a thought or deliberation on it at all. The complexities driven by a silent intelligence that perfectly understands what has been said, what is to be said next, and what is happening around it. The habits of the body.

The idea of a body ‘understanding’ through habit, each habit being such an understanding, requires us to revise what we mean by the term ‘understanding’:

Or as O’Loughlin (1995) puts it:

Already, this drawing of our attention to the knowledge of the body, contained in habit, is making progress in the problem concerning the relation between the mind and the world such that knowledge, and its acquisition, could be possible. The relationship of interest is now the seemingly more manageable one of the relationship of the body to the world. For Merleau-Ponty, for the most part, Heidegger’s Being-in-the-World is largely thought in terms of our bodies and their habits. If the world is to present itself as something like a ‘field of equipment’ as Heidegger understood it, then it appears that way to and for a body that is habituated into its use. But, furthermore, and this is key, our body’s habituated capabilities directly emboss, or ‘fold’ our experience. The room around me appears mostly ‘resolved’ and mundane, not exploding in mysteries and perplexities, because it is addressed to a body that is perfectly capable of moving through it without the least difficulty or consideration. Those habituated movements that would have me walk to the kitchen and back ‘carve out’ or ‘fold’ my fundamental and primordial understanding of my phenomenal situation. My body sits in and moves through its ruts quite comfortably.

This brings us to the important temporal aspect of habit, its acquisition and development. It’s not the case that I instruct my body with some architectural schema when I first move in to a house. The smoothness with which my body moves about it is the condensation and sedimentation of thousands of walks I have taken through it. Our body’s relationship to the past is deeply intimate; we sit and complain about the vagaries of memory, all the while utilizing effortlessly and exquisitely thousands upon thousands of unique vocal articulations called ‘words’ — our larynx, lips, and tongue dancing around ‘remembering’ volumes of the past while we struggle to bring to mind some vague image from a few days ago.

Insofar as the world appears before us as it does, curved and ‘unthought’ because of the knowledge contained in our body’s habituated capabilities, our body brings our full past to bear upon the present moment. This is not in the sense of a ‘projection of memories’, where, for example, a visual experience is supplemented, collage like, by fragments of previous visual experiences, but, rather, the very significance of things, the particular way the world folds itself in my attention, is arranged by what my body, through habit, can do. For most of us the street appears as a space to be walked through, but for the skateboarder, parkour runner, graffiti artist, or homeless person sadly accustomed to sleeping on the streets, entire new cartographies open up. This, then, brings habit’s condensation of the past into a projection upon the future — those possibilities that become present, in the present, speak to the ruts and furrows carved out through the body’s habituated understanding.

Now, finally, we can return to the process of learning. It is certainly possible to learn how to skateboard, and thus to ‘open up’ the space of the street to different possibilities and significances (Richards, 2017). By teaching the body new ways of movement, the very world is rearranged, or ‘re-prioritized’ in regards to its significance and how it is to be encountered. However, we seem a far cry from the children learning mathematical operations. If that seems to be the case, however, then it means we are holding on to the old categories. If we think of Merleau-Ponty’s notion of ‘habit’ as a mere ‘programming of the body’ the way we might program a stop watch, granting that the body as a semi-autonomous object might be ‘primed’ in this or that way, we have missed how radical his solution is.

If Merleau-Ponty is right about this, the arithmetic learning that the children undergo flies under a cognition that would associate or analyse the mathematical relationships involved, straight into a silent intelligence of the body that has sedimented all of the past instances of beginning the addition algorithm by counting out the first term. The body now perceives the first term as already counted, the enumeration compressed into a result that the child just sees directly there in front of them, carving out the initial rivulet of an entire field of mathematical perception. This perception is not there waiting in a priori concepts in the mind, nor is it the association between sense data amassed in fragments in some storehouse of memory, but is habituated into the body at the same time the significance of cardinality emerges in the world, as a new past-present-future path of what the body can do.

James Paulius (@jamespaulius)

Back to the paradox, Indeterminacy

It should be obvious that habit is doing the heavy lifting in Merleau-Ponty’s rephrasing of the relationship between ‘mind’ and ‘world’ (though these terms should hopefully now be seen as poorly applied to his view) such that knowledge and its acquisition could be possible. But does this notion manage to provide the mediation between the two seemingly irreconcilable sides of the Meno paradox?

Ultimately, the paradox is solved by this connection of habit and perception. Even though my body can negotiate my house all on its own, it does not do so as if the necessary movements were programmed into it, step by step in some cartesian geometrical space projection. This is to miss the close connection between this habit and how the world is presented in virtue of this habit. The body is not blind; it is the very condition of our being-in-the-world. Thus, the understanding, now defined as the “the harmony between what we aim at and what is given, between the intention and the performance” (Merleau-Ponty, 2005, p.167) is given over to degrees. When habit enacts itself in the present there is an endless novelty introducing ‘halos’ of indeterminacy, standing witness to the fact that my body can never be so adept that the entire world situation is absolutely resolved.

The temptation, elicited by the deficient notion of ‘sense-data’, is that there can be no indeterminacy in ‘pure’ perception. The interpretation of countless optical illusions, in this reading, is that the ‘eye’ sees correctly, it is just the mind makes an erroneous association with a memory, or a judgement applying the wrong concept. However, Merleau-Ponty affirms the ambiguous character of experience — that character pursued by the Impressionist painters. In order to ‘demonstrate’ the error involved in an optical illusion, the terms of the illusion need to be changed. One uses a ruler to measure the lines, or stares at the moon on the horizon through a tube. We are then told that in this new, modified situation, things are as they always were all along: now we are really seeing what we were always seeing, without the cognitive distortion. But what of the experience prior to this rearrangement, when the parts formed into a gestalt generating the ‘illusion’? It’s not that the moon on the horizon, when viewed through a tube, has reduced in size by 1.7 cm, or some determinate amount like this, but:

That is to say that it is perfectly natural for perception to consist of indeterminate sizes. This is a contradiction in pure geometry, but not perception. Not only this, but furthermore our world is populated by indeterminate forms: the ‘back’ of every object present to me now is indeterminate (insofar as it is unseen), but the capabilities coiled within my body through habit resolve this indeterminacy mostly — rarely am I surprised to reach out and pick up an object only to realize that my hand has severely misjudged its grasp because of the form of the unseen back. My hand is mostly ‘on top of’ this kind of indeterminacy, so the objects around me don’t pose incessant sceptical questions about the existence of their reversed sides.

It’s precisely this indeterminacy in perception, that the silent knowledge of habit is always more or less capable of ‘handling’, that opens up cracks of novelty and the possibility of development. It’s possible for some object to appear ‘half-known’, not in the determinate sense of posing a specific request for a proposition that would complete the knowledge, but in its appearing ambiguous right where it is. All that is required is for the body to be directed towards this ambiguity, investing it with a significance that can become an object of habit, as when the student of language begins to ‘perceive’ the phonemic distinction around the time they learn how to articulate it in their own pronunciation.


The Meno paradox shares with Zeno’s paradoxes of motion the incredulous response of the reader. We obviously learn things, and walk past tortoises, so obviously some trick is being played. However, like the Zeno paradoxes, the Meno paradox is not simply resolved by untangling the terms. These paradoxes stand as the excesses of fundamental errors that run much deeper than the terms of the paradoxes suggest. In regards to the Meno paradox I’ve hoped to show how the dominant philosophical views critiqued by Merleau-Ponty prefigure us being caught in its trap. With the rationalist’s and empiricist’s complex ways of conceiving the relationship between the mind and the world, and their assumptions, we are beholden to the absurdity that learning is impossible. This is quite the embarrassment. The fact that Merleau-Ponty’s model identifies the source of this predicament, the precise moment that empiricists and rationalists become fated to encounter the Meno paradox, as well as offering us a way out of it, should be more than enough to recommend it.

John C. Brady is a student of philosophy and educator situated in Beijing. He gets most of his reading done in traffic jams. He is also a co-editor of this magazine, by way of full disclosure.

Works Cited

Basut, L. (2017). Meno’s “Paradox”: An Analysis of the Eristic Argument. Filozofski Vestnik, 7(21).

Bereiter, C. (1985). Toward a Solution of the Learning Paradox. Review Of Educational Research, 55(2), 201. doi: 10.2307/1170190

Merleau-Ponty, M., & Fisher, A. (1963). The Structure of Behavior. Boston: Beacon Press.

Merleau-Ponty, M., & Smith, C. (2005). Phenomenology of perception. London: Routledge.

O’Loughlin, M. (1995). Intelligent Bodies and Ecological Subjectivities: Merleau-Ponty’s Corrective to Postmodernism’s “Subjects” of Education. In A. Neiman, Philosophy of Education Yearbook. Chicago: The Philosophy of Education Society. Retrieved from

Plato, & Waterfield, R. (2005). Meno and Other Dialogues. Suffolk: Oxford University Press.

Richards, J. (2017). Enframing, Inhabitation, Skateboarding — Epoché Magazine, Issue 02 . Retrieved from

Scott, D. (2006). Plato’s Meno. Cambridge, UK: Cambridge University Press.

Strange, W., & Dittmann, S. (1984). Effects of discrimination training on the perception of /r-l/ by Japanese adults learning English. Perception & Psychophysics, 36(2), 131–145. doi: 10.3758/bf03202673

Talero, M. (2006). Merleau-Ponty and the Bodily Subject of Learning. International Philosophical Quarterly, 46(2), 191–203. doi: 10.5840/ipq20064622


November 2018


A Few Notes on Identity Politics

by Timofei Gerber

Merleau-Ponty, Education, and the Meno Paradox

by John C. Brady

How to end every war, beyond the Derridean horizon, towards Outer Heaven, or a very partial response to “Violence and Metaphysics”

by Raphael Chim