H2 Knowledge and Inquiry — Knowledge as JB (25/30)

Critically assess the view that only two conditions are required for knowledge: justification and belief. [RI Prelim 2023] Epistemologists have long sought to construct a definition of knowledge in the form of individually necessary and jointly sufficient conditions. While some propose the possibility that knowledge requires only justification and belief, a reasonable view in some fields, this requires a largely discredited and unipolar view about all knowledge in general, because truth is widely regarded as another necessary condition for justified belief to become knowledge. Ultimately, our need for belief and quest for justification in knowledge — a pursuit undertaken to mitigate the risk of epistemic error — implicitly reflects our need for knowledge to be true, making justification and belief jointly insufficient overall to constitute knowledge in the vast majority of our fields of inquiry. Before we tackle the necessity of the truth criterion, it is necessary to first consider whether belief and justification respectively are necessary for knowledge. Ostensibly, belief seems to be a distinct concept for knowledge in everyday parlance — often, we might hear a confident athlete declare before a game that he does not “believe” he will win, but he “knows” he will win. In this example, it appears that belief is not necessary for knowledge: we can know something without believing it to be the case. However, epistemologists have generally managed to dispel the intuitive, commonsense appeal of this illustration — what the athlete means is not that he does not “believe” he will win at all, but that he does not just “believe” he will win. This linguistic expression of confidence and certitude therefore should not render belief separate from our conception of knowledge in epistemology. In fact, Moore has observed that it would be contradictory and bizarre to claim one does not believe something that one knows — for instance, we would find it strange for someone to say that “It is raining, but I do not believe it is raining.” The absurdity of claims of the form “P, but I do not believe P” reflects that knowledge implies belief: when we make the knowledge claim “P”, it implies strongly that I indeed believe “P” to be the case. Hence, it is clear that belief must be a condition for knowledge — we encounter Moore’s paradox otherwise.

Similarly, justification is an important necessary condition for knowledge, even when it seemingly does not add to the utility of a belief. Detractors of the justification condition often claim that a belief without justification can be just as useful as a belief with justification — for instance, even though the Egyptians and the Mesopotamians were unable to offer a proof for the Pythagorean Theorem like the ancient Greeks did, they were equally able to use the theorem to construct right-angled triangles and build magnificent architectural feats. Hence, if the reason we value knowledge is that it is applicable in our lives, it seems that we would say that the Egyptians and the Mesopotamian peoples equally ‘knew’ the Pythagorean Theorem — in every meaningful sense, their belief (though unjustified), was just as useful as the ancient Greeks’ justified belief. This argument, however, remains unpersuasive for most philosophers, because the presence of epistemic luck does not detract from our capacity to use unjustified beliefs in our lives. For instance, a gambler would not be able to use the unjustified belief that the next lottery number will be 1234, a belief he obtained from guessing alone — even if the lottery number is 1234 this time, such belief obtained from guesswork will one day fail to win him a prize. In this matter, the utility of most beliefs in our lives are intricately connected to the strength of our epistemic justification for believing it — a rational person would hesitate to act on his unjustified belief about the lottery number, but he would be far more willing to act on his justified belief that he is likely to lose money from gambling (obtained from statistical analysis) and thus abstain from placing a bet. Thus, we cannot say that justification is not necessary for knowledge — the gambler does not know that “1234” will be the number that appears, because the belief’s poor justification makes it prone to error and by extension less useful in his life. At this stage, it is clear that justification and belief are both conditions for knowledge — an unjustified belief or a justified ‘non-belief’ cannot be knowledge. Some epistemologists go further to propound that these are the only conditions for knowledge — it need not be the case that knowledge is true. There is some limited merit to this view, particularly in some fields where the concept of truth seems to be elusive and inapplicable. For example, in aesthetics, it is unclear how we would judge the ‘truth’ of a belief that the Mona Lisa is beautiful — it would be strange to evaluate whether the physical artwork of the Mona Lisa corresponds to the abstract, intangible ideal of ‘beauty’, since we would not have

any epistemic access to the abstract realm of these ideals as entities living in space-time. Additionally, we seem to hold the intuitive conviction that beliefs about beauty are subjective and do not require correspondence to the ideal of ‘beauty’ as coherence with others’ aesthetic judgements — beauty is in the eye of the beholder, after all. As such, it is indeed the case that aesthetic judgements only need to constitute justified beliefs to become knowledge — even without correspondence to reality or coherence with other judgements, we seem to be able to know objects to be beautiful. Justified belief seems to be jointly sufficient to constitute knowledge in this case. However, aesthetics seems to be the exception to the rule rather than the rule itself — the unique nature of aesthetic knowledge that precludes truth is not found in most other kinds of knowledge we seek. It is easy to verify whether our beliefs about the natural world correspond to physical reality — for example, we can check whether the earth is a globe or a flat plane by looking at satellite images or performing calculations based on the earth’s curvature. In fact, we would consider it intuitively necessary for our beliefs to be true before we consider them to be knowledge. Although humans in the 13th century believed the earth to be flat, a belief justifiably obtained through the usually reliable senses that can typically identify the shapes of objects, we would hesitate to say that they ‘know’ the earth to be flat — they merely believed it to be so. A similar requirement of truth seems to exist in other fields too — in history, we would be uncomfortable with the statement that “Holocaust deniers know the Holocaust did not happen”, because the Holocaust did in fact happen — it corresponds to the experience of Jews in the past and coheres with our records from the 1940s. Hence, the fact that we reject justified but false beliefs as knowledge in a vast range of fields suggests that truth is an unimportant condition for knowledge as well. In fact, our need for justification and belief seems to imply our requirement that knowledge must be true. The reason the aforementioned Moore Paradox arises in the first place is because our beliefs pertain to truths in the world — when we believe “P” we also believe that “P is true, making it illogical to not believe what one claims to be true.” If our beliefs are inherently connected to truth, it stands to reason that our conception of knowledge should account for this condition of truth. Similarly, our need for justification is also tied to

our quest for truth in knowledge — we want to arrive at our beliefs in the right kind of way so that we minimise the possibility of epistemic error, i.e. the possibility that our beliefs are false. Hence, justification is an attempt to secure the truth of our beliefs, making it natural for truth to be a c