In ‘Edmund L. Gettier’s “Is Justified True Belief Knowledge?”’, I indirectly defended the traditional analysis of knowledge as justified true belief by applying Kant’s definition of sufficient justification as including both subjective and objective factors. I interpreted this, without claiming that this be what Kant had in mind, as internal and external justification.1 Hence, I am inclined to take – in the broadest sense of the word – an externalist position of justification for now. I write ‘for now’ because positions may, and often do, change over time.
I shall begin with the significance of external factors for justification (1). Subsequently, I shall argue that, at least on my account, knowledge does not necessarily require knowing that one know (2). Finally, I shall try to show the sceptic’s claim that if we take justification as a criterion for the analysis of knowledge, we necessarily start an infinite regress to be wrong. Against this, I shall offer a sketch of a theory of knowledge consisting of several layers in analogy to micro- and macrophysics (3).
1. Justification and the Import of External Factors
My most important reason for this view is that almost every justified true belief concerning the external world depends upon external factors. For, first, what makes a belief, consisting of a proposition pn, true, is not whatever else a subject Sn may believe but that whatever pn may state, in fact is the case. Thus, for instance, what makes the proposition pe ‘Event en is happening’ true is not anyone’s believing that en be happening, or that whatever else be connected to en be the case, but en’s actual happening, and this clearly is a factor external to Sn’s psychological and perceptual states. Of course, I am not denying the possibility that Sn may hallucinate or dream that pe. Again, this poses no problem to my analysis because in hallucinating or dreaming that pe, Sn may be subjectively (internally) justified in believing that pe, yet not objectively (externally), since outside Sn’s hallucination or dream pe is not the case.
Second, the very being of something (whatever it be) external to Sn’s cognitive processes which imports on his belief-forming in the first place renders impossible the claim that only internal factors could count as justifying grounds for holding a belief.
As some readers, in particular philosophers, may deem the notion of truth opaque, we can at this point of discussion reformulate ‘“pn” is true if and only if pn’ [‘pn’ ↔ pn] as ‘pn is satisfied if and only if Δpn’ [pn ↔ Δpn], where ‘Δ’ represents the actuality of pn’s content. This may have the advantage – that is, if it be an advantage, which I shall not discuss here – that we rid ourselves of quotation marks. Be that as it may, I shall deal with the notion of truth in detail in a separate future article.
Resuming our present concern, notice that, on my interpretation, justification differs from explanation, such that we need to distinguish justified true belief from true opinion with an explanation, the latter of which Plato considers as a third possibility of what knowledge may be in his Theatetos.2 Gettier, on the other hand, by hypothesis credits, with reference to the same passage, Plato with (at least contemplating) the view that knowledge be justified true belief.3 I regard this as implausible, since an explanation does not justify a belief, let alone in the manner Plato considers. The notion of explanation is, admittedly, ambiguous, for it depends on what you really try to explain. Explicating, for example, how one came to believe that pn does not necessarily equal both subjective and objective justification for holding the belief. To illustrate, one may well cite one’s own perceptions that pn, one’s being told that pn, and so on, while, nonetheless, pn is false, and so one indeed is not (fully, because only subjectively, but not objectively) justified in holding the belief that pn. This is the interpretation of justification Gettier contemplates which I rejected on the ground that I think it too narrow, as it ostracizes objective factors as bearing on the notion and, most importantly, criterion of justification within the tentative analysis of knowledge as justified true belief.
2. Knowing Does Not Logically Entail Knowing That One Knows
On my account, it is possible but not necessary that a subject Sn know that they know that pn. As usual, let us call this a propositional attitude of secondary order, that is, a propositional attitude ranging over, or directed at, a propositional attitude of first order, which in its turn ranges over, or is directed at, a proposition pn. For if there can be external factors significant for the justification of Sn’s holding a belief that pn of which, however, Sn can be unaware, then it is not necessary that Sn know that they know that pn in order to be justified in holding the belief. Just as in order to know that a knife’s blade will, under normal circumstances (the blade has not gone blunt by overuse, it is a real knife, not a toy, and so forth) cut my bear flesh if I rub its cutting edge against my skin, I need not be aware of the macrophysical object’s underlying microphysical structure, I can be unaware of some knowledge which I have just the same. In other words, I can know that pn without knowing that I know that pn. Besides the problem of an infinite regress we would face if we made it a criterion of knowledge that one know that one know – for this would require that one know that one know that one know, which, in turn, would require that one know that one know that one know that one know, and so on without end –, I deem this criterion too restrictive, even if we ignore the infinite-regress problem, since it would rule out many beliefs we usually would count knowledge. Moreover, this would pertain not only to propositional (know that) but also practical (know how) knowledge. No one is fully aware of either their propositional or their practical knowledge. You can ask someone how to do something, and you can sensibly ask how they know how to do it, but you would probably earn a funny look if you asked how they knew that they knew how to do it. Practical and propositional knowledge are often intertwined because knowing how necessitates knowing that. Thus, in order to know how to ride a bicycle, one needs to know that bicycles exist in the first place. One also needs to know that in order to move forwards, one needs to put one’s feet to the pedals and exert physical power. Additionally, this practical knowledge even demands further practical knowledge: how to use one’s legs, for instance, which, in turn, presupposes further propositional knowledge, et cetera. Again, knowing how to use one’s legs, or extremeties in general, does not depend on one’s knowing that one know.
3. Endless Justification of Justification?
A problem for justification as an applicable criterion for the analysis of knowledge is pointed out in Agrippa’s so-called Five Modes4, which I mentioned in my ‘Ancient and Cartesian Scepticism’5. The first of the three so-called formal modes states that any justification offered stand itself in need of further justification, so that we cannot help but start an infinite regress. Not only am I not willing to accept this, but also shall I argue that this purportion by the sceptic is false. An intrinsic problem of the sceptic’s statement is that it itself stands unjustified, and if it is the case that each claim stands in need of justification, there is no reason for us to make an exception for the sceptic’s contention. If, on the other hand, the sceptic insists on allowing one purportion to stand unjustifiedly as knowledge, he lacks reasons to forbid the introduction of further unjustified statements as knowledge.
This usually motivates a foundationalist stance, and at this point I prefer such a view over a coherentist one. That is to say, I hold that there be some foundational beliefs which are basic, and therefore not only need no further justification, but cannot be justified by even more basic beliefs. Notwithstanding, I hold that there be several levels or layers of knowledge with respect to one and the same entity (including processes, activities, and the like). Consider again the example I gave above: In order to know that activity A with macrophysical object Omaph1 can cause macrophyiscal object Omaph2’s macrophysical state Stmaphn, I need not know one or both objects’ underlying (constituting) microphysical structures Strmiphn. I can know every single fact there is to know about some macrophysical object Omaphn without knowing anything about the underlying microphysical Structure Strmiphn. In principle, then, if only theoretically, one can know every marcrophysical fact there is without knowing any microphysical fact whatsoever.
In analogy to macro- and microphysics, I propose, we may interpret other branches of knowledge as layers lying above one another. Notice that there may even be causal connections from bottom to top as in the case of microphysical and macrophysical facts, without it being necessary that a Subject Sn needs to know all underlying layers of knowledge in order that they know the n-th layer. If this is correct – and I am convinced it is –, one can be justified in holding the belief that pn without being justified in the sense that one know that one know. This is, by the way, another argument in favour of an externalist view of justification (and probably knowledge, too).
A problem with the analogy of micro- and macrophysics I acknowledge is, of course, that it is difficult to draw a definite line between both. It may thus become a problem for us where to draw the epistemological line between two layers of knowledge in order to help us to find out where to search for the respective layers’ basic beliefs. I shall not try to solve this problem at present, above all because I currently have no solution to offer. I shall resume this issue when an idea comes to my mind.6
1. Confer: ‘Edmund L. Gettier’s “Is Justified True Belief Knowledge?”’: http://wp.me/p1QVNW-o8, 1. The Analyses of Knowledge Gettier Considers, and 2. Gettier’s Two Cases
2. Confer: Platon: Theätet. Griechisch-deutsch. Kommentar von Alexander Becker; Frankfurt am Main: Suhrkamp Verlag; Erste Auflage 2007, 201d [The pagination follows the Greek edition by Henricus Stephanus from 1578.] and my ‘Defining Knowledge’: http://wp.me/p1QVNW-oD, 1. Plato’s Theatetos, fourth section.
3. Confer: Gettier, Edmund L.: ‘Is Justified True Belief Knowledge?’ in Analysis 23, pp. 121-123; Oxford: Blackwell Publishers, 1963, reprinted in Bernecker, Sven and Dretske, Fred (eds.): Knowledge. Readings in Contemporary Epistemology, p. 13, footnote 1; Oxford: Oxford University Press, 2000, reprinted 2005.
4. Confer: http://plato.stanford.edu/entries/skepticism-ancient/#AgrFivMod
5. Confer: ‘Ancient and Cartesian Scepticism’: http://wp.me/p1QVNW-os, 1. Ancient Scepticism and Agrippa’s Five Modes.
6. In the meantime, I should be glad about any suggestion anyone may have to offer.