Explaining Knowledge

Explaining Knowledge

We have many kinds of knowledge: perceptual, introspective, scientific, linguistic, ethical, logical, mathematical, aesthetic, historical, and others. Epistemologists have asked which of these is best justified, and which least justified. We should be able to rank them for degree of justification: introspective and mathematical knowledge might get high marks for justification, perceptual and historical knowledge relatively low (skepticism will feature in determining this ranking). But I want to ask a different question: which kind of knowledge is the most easily explained, and which the least easily? The question concerns explicability not justifiability; it belongs more to natural science (in a broad sense) than to normative epistemology. Which kind of knowledge is the most easily understood and which the least easily understood? The various kinds of knowledge have different kinds of subject matter and different procedures of verification, so the explanation might not be the same in each case; some may be easily explained, some not explained at all, some inexplicable. A Cartesian might suppose that knowledge of oneself is most easily explained; an empiricist might contend that knowledge of perceived objects is most easily explained; a rationalist might think that logical and mathematical knowledge is most easily explained; a post-modernist might believe that knowledge of language (“texts”) is most easily explained. Any explanation must accept that it has to account for how the knowledge in question is knowledge, not merely belief or a certain kind of brain state or a disposition to verbal behavior. The explanandum is knowledge as such (under that description); it can’t be something merely correlated with knowledge. So, we will need to operate with a general conception of what knowledge is—something along the lines of true justified belief. The explanation will then take the form of specifying what accounts for the existence of true justified belief in a given area. How is such a state arrived at, by what mechanism or procedure?

I will begin by stating, intuitively and dogmatically, what I take the correct ordering to be. Then I will enquire into the principles governing it. None of this will be obvious or easily demonstrated; we are in uncharted and muddy waters. Still, some suggestive results may emerge. Here goes then, from easiest to hardest: proprioceptive knowledge, introspective knowledge, linguistic knowledge, perceptual knowledge, ethical knowledge, logical knowledge, and mathematical knowledge (I will eventually get to philosophical knowledge). I am pretty firm in this ranking, down to the precise ordering of subject areas, arbitrary as it may seem. Perhaps astute readers have a sense of the naturalness of the ordering already, but we will need to elucidate that sense; it doesn’t correspond to any existing ordering in epistemological studies. The claim, then, is that knowledge of one’s own bodily position and motion is the most easily explained type of knowledge, with knowledge of one’s own mind the second easiest; knowledge of one’s native language is the third easiest; then knowledge of objects in the external perceivable world; followed by knowledge of right and wrong; then logical knowledge; and finally mathematical knowledge (philosophical knowledge will be left dangling for now). The easiest to explain is proprioceptive knowledge; the most difficult to explain is mathematical knowledge—does any hypothesis spring to mind? Is there a detectable pattern here?

I think a chord will be struck if I say that knowledge is most easily explained when the object of the knowledge (its subject matter) “comes before the mind”. When the object rears up in front of the mind, presenting itself to the mind, then the knowledge arises explicably (relatively speaking). For then, and only then, is the thing known clearly accessible to the mind in its cognitive endeavors—right before its gaze, so to speak, begging to be known. Let’s scrutinize this phrase “before the mind” more analytically; so far it has only been an intuitively natural idiom or image or picture. There are two sides to it: the object is before the mind spatially and immediately. It is removed from the mind, separate from it, and it is also directly apprehended by the mind. It is outside the mind and also inside it—proximate to it, touching it. It is without and within. It needs to be outside the mind because knowledge is a relation between the mind and the world, but it needs to be in close proximately to the mind in order to be apprehended directly. Then, and only then, do we have knowledge. If the object were identical to the state of knowing, then we would not have the relation of knowledge; but unless it is directly apprehended, we cannot easily understand how it can be genuinely known. The object needs to be right there but it can’t be just the knowing mind itself. Perhaps it is now clear why I said that proprioceptive knowledge is the easiest to explain: the body is not identical to the mind, coinciding with it, but it is immediately present to the mind—we can therefore know things that are thus given to the mind. The body is spatially removed but directly apprehended: it is before the mind spatially (it is an object in space), and it is before the mind epistemically (fully given). It thus fulfills the two conditions for knowledge to be explicable, given what the concept requires—relational immediacy. It is intelligible to us that we know our own bodily position; nothing in the concept of knowledge is violated by this type of knowledge. It makes sense that we have knowledge in this case: the body comes before the mind, and hence we know about it. It is up close and yet at some distance.

Next in the ordering comes introspective knowledge. Here we notice that one condition is amply fulfilled but the other violated: the mental state known is immediately present to the mind, but it is not spatially remote from the mind and not apprehended as such. It threatens to collapse into the knowing mind. Thus, there is pressure to deny that the concept of knowledge really applies in this case; one of its necessary conditions is unfulfilled (cf. Wittgenstein). It is difficult to explain why there is knowledge in this case—more difficult than in the case of proprioceptive knowledge. And yet we do have a clear case of epistemic immediacy, even certainty, so we can always cite this in defense of a claim of knowledge. There is some difficulty explaining the existence of introspective knowledge, but it is not (felt as) decisive. Linguistic knowledge comes next—knowledge of meaning, grammar, and use. Here also we find a dilution of the constitutive conditions for knowledge, though not an abrogation: we have a fair degree of immediacy, but the spatial condition is again found wanting, since language is in the mind, not in the world outside (I am speaking of the idiolect primarily). Do I really explicably know the meaning of my words? Well, I am not often wrong about such things, but my meanings are not in space removed from my mind, unlike my body. My meanings are not before my mind in any kind of space, though I apprehend them directly (not by inference or conjecture). So, there is some difficulty explaining how I have knowledge in this case (and of course this has been denied). Meanings are not as transparent as pains, so the immediacy condition is less clearly met, but it is met well enough to allow knowledge of language to be so described. Perceptual knowledge of the external world satisfies the spatial condition perfectly, but it stumbles with the immediacy condition: we can easily be wrong about what we are seeing and hearing (etc.), so justification is wanting in this case. External objects don’t come before mind like pains and meanings; they are more susceptible to illusion and error (hence the possibility of skepticism). Still, we can allow that we have knowledge of this kind because we have evidence concerning external objects, and they clearly meet the relational requirement. Here knowledge is arguably explicable following the paradigm already laid down, though imperfectly. What about ethical knowledge? Obviously, we are now moving into the normative domain, but the same principles apply in this case as hitherto. Values such as right and wrong can be apprehended by the mind, but they are not as clearly separate from the mind as one would like; they seem to hover somewhere between mind and world. The externality condition is not clearly met, so it is less explicable that we have a case of knowledge here. The initial paradigm for adequate explanation is breaking down, though remnants of it remain; we feel uncertain about the “before the mind” condition. The possibility of error and the appearance of internality start to undermine the propriety of applying the concept of knowledge in the ethical case. It becomes harder to explain why there is such knowledge (though it is evident that there is). Logical knowledge is not dissimilar: it is normative and non-spatial (in reality and phenomenologically), yet we do have reasons in good standing for our logical beliefs. We don’t experience logical truths as laid out before us, spatially or even quasi-spatially, so the relational character of knowledge is not clearly respected. We thus have no model for how such knowledge is to be explained, or only part of a model. The case of mathematical knowledge pushes the problem even further: numbers are not spatially related to us, or apprehended as occupying space, so we can’t explain the application of the concept of knowledge in the usual way. The presence of the infinite only exacerbates the problem: infinite totalities don’t “come before the mind”. It is very hard to explain how mathematical knowledge arises. One is tempted to describe it as a complete mystery. It is even hard to understand how numbers could be immediately given to the mind, apprehended as the objects they are; the phenomenology is obscure. It is nothing like the experience of awareness of one’s own body, an object in space whose properties are immediately presented. Thus, I anoint mathematical knowledge as the most difficult type of knowledge to explain.[1]

I have said nothing about explanation and causality. Is it that knowledge gets harder to explain the less causally explicable it is? No doubt causality is involved in questions of explanation, but I don’t think it goes to the heart of the matter. It is a blunt tool for current purposes. First, it is too poorly understood itself (as Hume pointed out). Second, it is too various and vague. Third, it is not specific to the case of knowledge. Fourth, the condition captured by the “before the mind”” formulation works better in describing particular cases. If we relied on causality to make the relevant distinctions, we couldn’t distinguish proprioceptive knowledge from introspective or perceptual knowledge in point of explanatory difficulty. We need something more fine-grained, more supple, less binary. The concept of causality has proven less useful in epistemology than we thought in the heyday of causal theories.

I have stressed the duality inherent in the concept of knowledge, which I described as separation and immediacy. But I suppressed the tension that exists between these two requirements, and hence afflicts the concept of knowledge itself. For the more separation there is, the less immediacy there is; and the more immediacy, the less separation. Knowledge requires distinctness from the mind but also proximity to it. Ideal knowledge would consist in identity with the knowing mind, but then it wouldn’t be knowledge at all, which requires relationality. Knowledge calls for a balance between these two forces—separate but not too separate, immediate but not too immediate. This is why proprioceptive knowledge is the paradigm: the body is not the mind, but it is close enough to afford epistemic immediacy. Introspection gives us plenty of immediacy but no separation. Mathematics gives us neither. It seems conjured from nowhere. Certain knowledge is not ipso facto explicable knowledge, because we need an explanatory framework that renders it intelligible. The further we move away from this framework the less intelligible the putative knowledge becomes. This is why knowledge has been disputed in many areas (but not on skeptical grounds): either lack of immediacy (perceptual knowledge, historical knowledge, and scientific knowledge) or lack of separation (introspective knowledge, ethical knowledge, logical knowledge, mathematical knowledge). But we don’t find a denial of knowledge in the case of proprioception (on non-skeptical grounds anyway): here we have the perfect blend of distance and immediacy. I really do know that I am in a seated posture now.

Finally, philosophy: how hard is it to explain philosophical knowledge? Does it combine separation and immediacy in the right proportions to count as full-blown knowledge? That depends on its subject matter. If it concerns platonic essences, it will approximate to mathematics. If it resembles empirical science, it will conform to that branch of knowledge. If it is about concepts, construed as psychological entities, then it will belong with introspective knowledge. I can’t think of any meta-philosophy that assimilates it to proprioceptive knowledge, so it will not fit that paradigm. If it combines all these forms of knowledge, then it will vary in its difficulty of explanation. You choose.[2]

[1] Astronomical knowledge is much easier to explain than mathematical knowledge, despite the distances involved and the absence of certainty. Mathematical entities are not even in space. Nor do they emit light.

[2] This paper flies at a very high altitude. It would be understandable if the reader felt oxygen-deprived.

Share
11 replies
  1. Giulio Katis
    Giulio Katis says:

    A very interesting approach to getting a foot in the door on this problem. I might have put interoception on par, if not before, proprioception.

    Knowledge is a relation between the mind and the world: am I right to assume this implies knowledge is not simply a derivative of the mind, or does not reside entirely within the mind, but spans mind and matter (or mind and mind)? Is there a distinction between the activity of knowing (a spanning activity, an interaction of sorts) and the mental artefact (e.g. a concept, reification) that may be the output of this activity? With the latter being wholly within mind, and the former not. If so, would the concept of mind be the reification of the activity of knowing the knowing process?

    Reply
    • Colin McGinn
      Colin McGinn says:

      I was using proprioception to include interoception and kinesthesia.

      Knowledge is factive unlike belief, i.e. it implies truths about the world. It spans mind and world.

      Reply
  2. Giulio Katis
    Giulio Katis says:

    Numbers may not be in space, but they do have a relationship to external space, as an abstraction of certain properties space has, or, more accurately, structures space gives rise to. If we stick with discrete whole numbers (counting), they can arise in (at least) two ways. One is associated with the granular nature of stuff, where space provides separation between cohesive blobs (components). Numbers are an abstraction of such blobs, such as a bag of stones. – the number of stones is a form of knowledge of that bag of stones. (Once we had numbers, shepherds could keep track of their flocks without having to construct bijections between the flock and a bag of stones.) The other is to do with knowledge of the circle – specifically, the windings of the circle around itself, which has to do with standing waves and harmonics (Pythagoras). Interestingly, though people often talk about Quantum Mechanics in terms of the former, it actually arises out of the latter.

    Reply
      • Giulio Katis
        Giulio Katis says:

        I mentioned two. The first (which you refer to) is Cantor’s (lauter Ensein) which led to the great mathematical abstraction of set (as mengen) from which cardinality is an abstraction. In modern mathematics (algebraic topology). this is encoded as the connected components functor from a category of spaces to a category of sets. It is the ground zero of understanding the mathematical structure of space (its coarsest aspect). It is interesting that Cantor’s work came so late in mathematics (eg after the insights on the mathematical nature of space of Euler, Gauss, Riemann…)

        Reply
  3. Giulio Katis
    Giulio Katis says:

    Your post causes me to reflect that though mathematics and philosophy came after medicine, sports and warcraft, agriculture, astronomy, they were very early (perhaps the first?) areas of formal knowledge, paving the way for the development of sciences of the body, mind, and world around us. This suggests their subject matter is about something deeper structurally than the subjects of the sciences that needed to be surfaced for the other sciences to develop. (I believe this can be stated without recourse to Platonism.)

    Reply
      • Giulio Katis
        Giulio Katis says:

        I was thinking about, say, knowledge of meter. Which seems to span inner and outer, a syncing of an internal (bodily, perhaps interoceptive?) with an external pattern.

        Reply

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.