Reliabilism Justification I believe that there is a

Reliabilism

Justification I believe that there is a dog tied out in front of the UCen. – I didn’t see the dog on my way here – No one told me about it – I have no past experience with dogs being tied up there at this time of morning. – I am just guessing. But as a matter of fact, there is a dog there!

Justification Do I know that there is a dog there? No!

Justification Knowledge requires more than just a true belief. You need: – Reasons – Evidence – Rational support – Etc.

Justification Call a belief justified if it is believed on a good basis (good reasons, good evidence, rational grounds, etc. ).

Skepticism about Justification Our skeptical arguments so far have concerned knowledge, but they can be easily reformulated to show that we cannot have justification for our beliefs about the external world.

Skepticism about Justification Skepticism about justification is arguably closer to the real issue than skepticism about knowledge.

Skepticism about Justification The skeptic does not try to show that our beliefs are false or that we fail to believe what we think we do. Rather, he tries to show that our reasons for believing what we do are not strong enough to amount to knowledge.

Skepticism about Justification If this is right, then skepticism is even worse than one might think. Not only do you not know anything about the external world, you don’t even have good reasons to believe it!

Skepticism about Justification The Transmissibility Argument (Justification) 1. I cannot be justified in believing that I am not a BIV. 2. My having hands entails that I am not a BIV. 3. If I justifiably believe p and that p entails q, then I can be justified in believing q. 4. So, if I justifiably believe that I have hands, then I can be justified in believing that I am not a BIV. 5. I cannot be justified in believing that I have hands.

Introduction Goldman wants to provide a fully general account of justified belief. His main claim is that what makes a belief justified is that it is formed by a reliable belief forming process.

Introduction If he is right, then we will have a good response to the skeptical argument about justification.

Desiderata Goldman explains what he wants from a philosophical theory of justification: 1. A theory that explains and makes sense of our ordinary attributions of knowledge and justification.

Desiderata 2. Non-circular – Don’t provide an account of justification by employing other epistemic terms (warrant, reasons, knowledge, good grounds, etc. )

Desiderata 3. Not merely extensionally correct. – He wants an account that agrees with all (or most) of our judgments about individual cases. – But he also wants his account to be explanatory. – He wants to be able to say why a particular belief is justified when others are not.

Belief-Forming Processes Goldman thinks that whether or not a belief is justified depends on whether or not the belief was formed by a good or bad process.

Belief-Forming Processes Belief-Forming Process: An operation or procedure that takes an input (e. g. perception, memorial experience, testimony, other beliefs) and outputs a belief.

Belief-Forming Processes Bad Belief-Forming Processes: 1. 2. 3. 4. 5. Confused reasoning Wishful thinking Reliance on emotional reactions Mere hunches or guesswork Hasty generalization What do these have in common? – They are unreliable.

Belief-Forming Processes Good Belief-Forming Processes : 1. Good reasoning 2. Perception 3. Remembering 4. Introspection 5. Properly carried out induction or abduction What do these have in common? • They are reliable.

Belief-Forming Processes like perception, good reasoning, and memory typically furnish us with justified beliefs. Processes like fallacious reasoning, wishful thinking, and hasty generalization typically furnish us with unjustified beliefs.

Belief-Forming Processes Nice Features: 1. Picks out the intuitively good processes from the bad ones. 2. Makes sense of the connection between justification and truth. 3. Allows us to make sense of the fact that justification comes in degrees.

Announcements We will (finally) start on Thomson's "Defense of Abortion" on Monday. Dillon will hold office hours today from 10 AM-1 PM My makeup office hours are Today from 11 AM-1 PM

Belief-Forming Processes Goldman’s Basic Idea: If a belief is formed by reliable process it is justified, and if it is formed by unreliable process, it is unjustified.

Reliability What is reliability? • Reliability has to do with the ratio of true beliefs to false beliefs formed by using a particular process.

Reliability There are problems if “reliable” just means: yielded more true beliefs than false ones. A long series of lucky guesses is reliable according to this definition, but such beliefs are not justified.

Reliability We also need to add in how the process would perform if things had been different. Such conditions are called counterfactual conditions.

Reliability How reliable must a belief-forming process be in order to confer justification? – Perfect reliability does not seem to be a plausible condition. Justification requires a high degree of reliability, but not perfect reliability.

Reliabilism (First Formulation) If S’s believing p results from a reliable cognitive belief-forming process, then S’s belief in p is justified.

Reliabilism Meets Goldman’s desiderata: – Reliability was defined in terms of a ratio between true and false beliefs, hence it is reductive in the way that Goldman wants. (no epistemic terms) – It is also explanatory. It tells us why beliefs are justified: if the that produce them reliably produce true beliefs this is a good explanation for why we should believe thing.

A Problem • Any reasoning process relies for its reliability on factual or likely input beliefs. – If you have a logically valid argument, it seems in one sense to be perfectly reliable: given true beliefs you get only true beliefs out. – However, it becomes immediately unreliable if you start with false beliefs. – Ditto memory. – Ditto testimony?

Two Distinctions 1. Belief Dependent vs. Belief Independent • Some belief forming processes take other beliefs as inputs. – E. g. deductive reasoning, abductive reasoning, inductive reasoning, • Call such belief-dependent belief-forming processes.

Two Distinctions 1. Belief Dependent vs. Belief Independent • Some belief forming processes are that include no beliefs as inputs: – E. g. introspection, perception, • Call such belief-independent belief-forming processes.

Two Distinctions 2. Conditional vs. Unconditional Reliability Conditional reliability: A belief-forming process is conditionally reliable if it yields a high true/false belief ratio given true input beliefs. Unconditional reliability: A belief-forming process is unconditionally reliable if it yields a high true/false belief ratio.

Two Distinctions • What matters for belief-dependent beliefforming is conditional reliability. • What matters for belief-independent s is unconditional reliability.

Reliabilism (Official formulation) 1. If S’s belief in p at t results from a beliefindependent process and that process is reliable, then S’s belief in p at t is justified. 2. If S’s belief in p at t results from a beliefdependent process that is conditionally reliable, and if the input beliefs on which this process operates are themselves justified, then S’s belief in p at t is justified. 3. These are the only ways for a belief to be justified.

Reliabilism (1) If S’s belief in p at t results from a beliefindependent process and that process is reliable, then S’s belief in p at t is justified. This is just Goldman’s first attempt, applied only to belief-independent processes.

Reliabilism (2) If S’s belief in p at t results from a belief-dependent process that is conditionally reliable, and if the input beliefs on which this process operates are themselves justified, then S’s belief in p at t is justified. The added condition deals with the problem presented by belief-dependent processes

Reliabilism Good features of reliabilism 1. Accounts for the justified beliefs of children and smart animals. 2. Accounts for our justified beliefs based on perception, memory, and so on. 3. Does a good job of discriminating between good belief-forming methods and bad ones. 4. Gives us an explanation as to why certain of our beliefs are justified and others are not.

Against Skepticism The anti-skeptical strategy is very similar Moore’s. One could think of it as filling in the details behind his argument.

Against Skepticism Remember Moore’s anti-skeptical argument: 1. There is a table here. 2. If there is a table here, then I am not in the Matrix. 3. Therefore, I am not in the Matrix.

Against Skepticism Moore just claimed to know (1) without any argument. But with Goldman, we have an actual theory that supports the premise.

Against Skepticism 1. There is a table here. Perception is a reliable belief-forming process, and I form this belief on the basis of perception. So this belief is justified.

Against Skepticism 2. If there is a table here, then I am not in the Matrix. Stipulated by the features of the case.

Against Skepticism 3. Therefore, I am not in the Matrix. Inferred from (1) and (2) by valid deductive inference. (1) and (2) are both justified beliefs, and deductive inference is a conditionally reliable process. By the second clause of Goldman’s analysis 3 is also justified.

But Wait… Couldn’t the skeptic respond as follows: – I don’t grant that you know your perception is reliable. – In the Matrix your perceptions are radically unreliable. – If you can’t tell the difference how can you claim to be justified?

Goldman’s Response Goldman’s theory does not require that you must know that your belief-forming method is reliable. It is enough that it is.

Goldman’s Response Of course you can come to know your perception is reliable, but that is not required to justify belief that there is a table here.

Turn your papers in! Please turn your papers in to your TA when they arrive. If you have Jenna, put your paper on the table at the front of class.

Objections to Reliabilism The New Matrix Problem • I am not in the Matrix just as I believe. • But imagine Jim, who is in the Matrix, and has been fed experiences just like mine his whole life I am reliable and he is not.

Objections to Reliabilism The New Matrix Problem • The reliabilist has to say that I am justified in my beliefs, but Jim is not. • But how could this be? Jim’s evidence is exactly identical to mine!

Objections to Reliabilism Bad (Reliable) Belief-Forming Processes We can imagine cases in which intuitively bad processes are employed generating reliably true beliefs – Norman the clairvoyant. (Bon. Jour) – Lucy and the benevolent demon. (Goldman)

Other Responses to Skepticism 1. 2. 3. 4. 5. 6. Transcendental Arguments (Kant) Idealism (Berkeley) Coherentism (Bon. Jour) Contextualism (Cohen, De. Rose, Lewis) Truth-tracking views (Nozick) Hinge propositions (Wittgenstein, Wright)