Chapter 10 The Analysis of Knowledge: Justification, Certainty, and Reliability
This page is part of my self study project into philosophy. What is written are my own personal notes taken as a means of formalizing what I've read and/or learned. The information may not be accurate, as I may have took away the wrong points. Also the information could be basic or partial - as learning requires taking on small parts before getting deeper. The content may be a tad scatterbrain as well. Either way, it should be an interesting read and maybe you'll learn something with me.
What exactly is knowledge? A false belief is not knowledge. A belief based on a lucky guess is not knowledge, even if it is true.
- 1 Knowledge & Justified True Belief
- 2 Knowledge Conceived as the Right Kind of Justified True Belief
- 3 Naturalistic Accounts of the Concept of Knowledge
- 4 Problems for Reliability Theories
Knowledge & Justified True Belief
What is not true is not known. Where you claim to know something that turns out to be false, you might say you believed it, but would never say that you knew it. This point suggests knowledge is at least true belief. It is also understood that as one acquires more evidence for a belief, one is brought closer to justification for holding it. So justified belief is an element of knowledge.
A necessary condition for propositional knowledge seems that knowledge is at least justified true belief - that we know something only if we believe it, it is true, and our belief of it is justified. A problem here is that a justified belief might turn out to be false in the case of seeing a picture of a person, but presented in such a way that it looks like a person before you.
Knowledge Conceived as the Right Kind of Justified True Belief
The photographic example above is said to be defeated. Undermined or overridden is different such as a testimony and later discovering the person is not credible. the photographic example is epistemic defeat: it eliminates the power of the justification to turn a true belief that acquires justification to turn a true belief that acquires justification into knowledge. The witness example is justification defeat. We might consider knowledge as undefeatedly justified true belief.
Dependence on Falsehood as an Epistemic Defeater of Justification
How is epistemic defeat to be characterized? One natural view is that the justification of a belief is defeated provided the belief depends on a falsehood:
- (1) Depends on a falsehood that would not be justified except on the basis of one's being situationally justified in believing a falsehood about the subject in question.
- (2) Psychological dependence - depend on a falsehood in the causal sense that one has the belief by virtue of holding it on the basis of believing a falsehood.
Knowledge and Certainty
A reason to think knowledge requires conclusive justification is that knowing is often closely associated with certainty. Given the connection between knowledge and certainty, one might hold that knowledge is constituted by conclusively justified true belief, meaning that (1) the believer may justifiably be psychologically certain of the true proposition in question and (2) this proposition is so well grounded as to be itself propositionally certain. Knowledge in this case may be considered epistemic certainty.
Knowing and Knowing for Certain
Sometimes we speak of knowing for certain, which contrasts with simply knowing. It is doubtful that everything known can be known for certain.
Knowing and Making Certain
If we already know something, we can make certain it is so, then there is a reason to think that conclusive justification is not required for knowledge. Getting conclusive justification seems to be the main point of making certain. It might be to think of 'make certain' as 'make it certain' or 'make sure it is certain'.
Naturalistic Accounts of the Concept of Knowledge
A different approach asks whether one needs to appeal to the notion of justification to understand knowledge. Knowledge results from the successful functions of our epistemic equipment - which consists above all of finely tuned perceptual, memorial, introspective, and rational instruments.
The view that knowledge consists in suitably registering true goes well with the idea that we are biological creatures with sense receptors that gather information and with mental capacities that integrate it.
Knowledge as Appropriately Caused True belief
Causal theory - knowledge is true belief caused by something connected with its true in a way that makes it plausible to call the belief knowledge. So believing a green field before me is caused by my perception of seeing the field before me.
A problem with this idea: how to apply the basic idea to a priori. As it's not perceptual, nothing is really causing it.
Knowledge as Reliability Grounded True beliefs
There is another problem with the causal account, in relation to empirical beliefs. The trouble arises when has a justified true belief that is not knowledge. Suppose testimony that isn't credible. Hence not reliable. This may suggest it is plausible to analyze knowledge as reliably grounded true belief.
Reliable Grounding and A Priori Knowledge
In the case of a priori, our understanding of entities entails a direct contact with these entities, a kind such that the entities form an essential part of the very content of that understanding. Then they play the indirect role in sustaining beliefs which justified a priori knowledge. ie: a causal sustaining ground.
Problems for Reliability Theories
the reliability theory holds that the more reliable the testimony, the more justification. But it might be argued that my knowledge has a presuppositional dependence on the proposition that someone is reliable to justify accepting testimony.
The Specification Problem
A different problem faced by the reliability theory is specifying what is reliable in the first place. We could always say in good viewing conditions our perception can be reliable, but this doesn't deal with life size pictures or wax models that look real. Another issue is the quantity of factors that affect reliability. It would be tough to list them all and often terms would be like 'to far away'/
Reliability and Defeat
Let's say we have a reliability process where something seems reliable. What about the case of chance, is that knowledge? I will lose the lottery is a fair assumption as you have a one in a million chance, but there's also a chance that you could win. Your perception might see something, but there's a change it could be off. You see Jan, but there could be a chance you see a life size picture of Jane.
We wouldn't call the lottery example knowledge as the resulting outcome is chance. So how does one describe the right kind of dependence, called a functional dependence (or discriminative dependence). A useful metaphor is tracking. As we track a person in the snow, causally guided by the path, our belief system can be sensitive to the changing evidences that indicate the truth.
Reliability, Relevant Alternatives, and Luck
Suppose I see Jane standing 12 feet away from me. But in this case it is her identical twin, of who I don't know exists. If I had not learned to tell them apart, I would have taken Jane's sister for her. Suppose Jane was walking toward the same destination (and the twin went somewhere else out of view) I would have believed I saw Jan this whole time. Knowledge cannot be arrived at by luck.
Relevant Alternatives and Epistemological Contextualism
In the example above, context seems crucial. If I know Jane's sister is halfway around the world, I know Jane is before me. The idea here is that when there is a 'relevant' alternative to what we believe to be, the so is actually present in the context and we cannot discriminate it from the case we believe to obtain, our belief does not constitute knowledge when if it is both true and justified.
Judged by different standards in different contexts is often called contextualism.
Contrastivism - ordinary unqualified knowledge ascriptions are implicitly contrastive and the relevant set of contrasting cases is determined by contextual features.
For a contrzstivist, standards also vary with differences in meaning; for contextualists, who do no posit different meanings in those attributions, it is just the standards for true attribution that change.