Download Agrippa's Trilemma and Scepticism about Knowledge of the External World and more Study notes Philosophy in PDF only on Docsity!
Chapter 8 Agrippa’s Trilemma
The thesis we are considering is this: No one knows anything about the world external to his or her own mind (not even that there is such a world). We can call this thesis the thesis of scepticism about knowledge of the external world. Before attempting to determine whether the thesis of scepticism about knowledge of the external world is true or false, let us try to understand better both it and the argument we have given for it.
1. Epistemic justification If you know that presently you are reading a page in a philosophy text, then you can know that you are not being deceived by an evil scientist into falsely believing that presently you are reading a page in a philosophy text. That is because knowing that presently you are reading a page in a philosophy text implies that presently you are reading a page in a philosophy text, and that , in turn, implies that you are not being deceived into falsely believing that presently you are reading a page in a philosophy text. But you cannot know that you are not being deceived by an evil scientist into falsely believing that presently you are reading a page in a philosophy text. Therefore, you do not know that presently you are reading a page in a philosophy text. For all you know, you are reading your economics text, having mistaken it for your philosophy text. Or, for all you know, you are merely imagining or dreaming you are reading. The argument just given does not rest on the contention that you are, in fact, reading your economics text under the misapprehension that it’s your philosophy text, or that you are merely imagining or dreaming that you are reading a text. It does not rest on the contention that it is in any way likely that you are wrong about the text in front of you or merely imagining or dreaming that you are reading. It might not even require that it be possible that you are wrong about what you are reading or that you are merely imagining or dreaming something or that there could be an evil scientist able to deceive you about anything he wants to deceive you about. Our sceptic, who thinks no one knows anything about the world external to his or her own mind because he endorses as sound the argument above, is not trying to convince you that you are mistaken about anything (except, of course, about whether you or other people can know things). Nor is she trying to convince you that you are dreaming. That is not how the argument for scepticism we are considering works. The argument for scepticism we are considering does not work by proposing that you have false beliefs about what is in front of you or even that it is likely or possible that you have false beliefs about what is in front of you. It is not an argument that challenges the truth of what you believe. Maybe what you believe is true. Maybe presently you are indeed, as you believe, reading a page in a philosophy text. What the argument for scepticism we are considering challenges is the strength or quality of your justification for believing what you believe. According to that argument, you are not well enough justified in believing that presently you are reading a page in a philosophy text for your belief to count as a piece of
knowledge. Because you lack adequate justification for knowledge, you do not know. Because you lack adequate justification for knowledge, you do not know, even if what you believe is true. Recall from Part I of this text the three conditions necessary for knowledge. If you know that presently you are reading a page of a philosophy text, then 1) presently you are reading a page of a philosophy text, 2) you believe that presently you are reading a page of a philosophy text, and 3) you are very well justified epistemically in believing that presently you are reading a page of a philosophy text. To say that each condition is a necessary condition for knowledge is to say that if any one condition is not fulfilled, then you do not know that presently you are reading a page of a philosophy text. The sceptic who endorses as sound the argument given in the first paragraph of this section holds that you are not well enough justified epistemically for your belief that presently you are reading a page of a philosophy text to be a piece of knowledge. You don’t know, even though you believe with confidence and even if your belief is true, because the epistemic grounds on which you believe are not strong enough to make your belief a piece of knowledge. You believe that presently you are reading a page in a philosophy text. Why are you not adequately justified epistemically for that belief to count as knowledge? It would seem that you have plenty of good evidence in favour of your belief. Why is all the good evidence you have not evidence enough for knowledge? Even more, why must you always remain inadequately justified for your belief to count as knowledge? Why will your collecting more and more evidence in favour of your belief never turn your belief into a piece of knowledge? Recall again from Part I the standard account of the nature of epistemic justification. If a person is well justified epistemically in believing that penguins eat fish, then that person is aware of a good argument to the conclusion that penguins eat fish. Each of the premises of that good argument must express something that that person believes. A good argument is an argument that moves from acceptable premises to its conclusion through strong or valid inferences. Keeping the standard account of epistemic justification in mind, suppose that Sally believes that penguins eat fish and that she is aware of an argument to the conclusion that penguins eat fish, an argument each of whose premises she believes true. Suppose that each inference in that argument is strong or valid (and that Sally is aware of this). Now suppose that for one of the premises in her argument, a premise, we said, that Sally believes to be true, Sally is unaware of a good argument that has that premise as its conclusion. That means, on the standard account of epistemic justification, that Sally is not justified epistemically in believing that premise. Sally, then, is aware of an argument that proceeds from premises she accepts to the conclusion that penguins eat fish through strong or valid inferences—an argument, though, one of whose premises she is not epistemically justified in believing. Is Sally epistemically justified in believing that penguins eat fish? Sally, we can admit, is fairly well justified epistemically in believing that penguins eat fish. She has no evidence against her view and some evidence in favour of it. But she is not entirely well justified epistemically in believing that penguins eat fish. After all, Sally has no reason for thinking that one of the pieces of what she takes to be evidence that penguins eat fish really is evidence that penguins eat fish. Remember that Sally has no argument in favour of one of her beliefs about what she takes to be evidence.
justified in believing that presently you are reading a page in a philosophy text will do one of three (and only one of three) things, each of which prevents that line from conferring justification sufficient for your belief to count as knowledge. Any line of justificatory argument will regress from belief to belief endlessly or terminate in an undefended and, thereby, arbitrary assumption or circle back such that some belief in it supports that very same belief. This contention, that a justificatory argument will regress to infinity or halt arbitrarily or contain circular reasoning, is called Agrippa’s trilemma. (Nothing is known today about the Hellenistic philosopher Agrippa except that he proposed five modes of casting doubt on a person’s claim to know something, three of which form the trilemma we have just described.) Here, again, is Agrippa’s trilemma:
For any person S and any proposition “p” that S believes, any line of argument purporting to show that S is justified in believing that p will:
a) regress from belief to belief to belief without ever stopping. The argument that p will contain the premise that q; the argument that q will contain the premise that r; the argument that r will contain the premise that s; the argument that s will contain the premise that t..., and so on to infinity. or b) terminate in an undefended assumption. The argument that p will contain the premise that q and no line of argument will support the proposition “q.” S’s belief that q will not itself be supported by argument. or c) form a circle in which a conclusion of an argument close to p in the chain appears as a premise further along in the chain. The argument that p will contain the premise that q; the argument that q will contain the premise that r; the argument that r will contain the premise that s; the argument that s will contain the premise that q—but the statement that q has already appeared in the chain of argument; and so the chain will loop around in a circle.
We can make the trilemma clearer by returning to your belief that presently you are reading a page in a philosophy text. A premise in your argument that presently you are reading a page in a philosophy text is that your eyes are fine. You have an argument that your eyes are fine. A premise in that argument is that your doctor said they were fine. You have an argument that indeed the doctor did say they were fine. A premise in that argument is that you heard your doctor say they are fine. You have an argument that you heard your doctor say your eyes are fine. A premise in your argument is that your ears are fine. You have an argument that your ears are fine. A premise in that argument is that.... And, so, a line of argument that ends with the proposition that you are reading a page in a philosophy text has no beginning. That is the first of the three possibilities in Agrippa’s trilemma. Or, alternatively: A premise in your argument that presently you are reading a page in a philosophy text is that it looks to you as though presently you are reading a page in a philosophy text. You have no argument that it looks to you as though presently you are reading a page in a philosophy text. It just looks to you as though presently you are reading a
page in a philosophy text. End of story. There’s nothing in your mind that supports this belief. That is the second of the three possibilities in Agrippa’s trilemma. Or, alternatively: A premise in your argument that presently you are reading a page in a philosophy text is that how things look to fine eyes in fine conditions is a good guide to how those things are. You have an argument that how things look to fine eyes in fine conditions is a good guide to how those things are. A premise in this argument is that often in the past how things looked to your fine eyes in good conditions turned out to be how those things were. A premise in your argument that often in the past how things looked to fine eyes in fine conditions is that the dish looked white to your fine eyes when you viewed it in fine conditions and it was white. A premise in your argument that the dish was white is that how things look to fine eyes in fine conditions is a good guide to how those things are. A premise in your argument that how things look to fine eyes in fine conditions is a good guide to how those things are is that often in the past how things looked to your fine eyes in fine conditions... and so on. Your line of argument has formed a circle, with propositions functioning to lend support to themselves. Agrippa’s trilemma states that any line of justificatory argument will follow one (and only one) of the three patterns described above. There is no fourth pattern a line of justificatory argument can take. A line of argument can do no other than regress to infinity, end at an undefended belief, or circle around on itself.
3. The implications of Agrippa’s trilemma The sceptic who says that no one is ever well enough justified in believing something for her belief to be a piece of knowledge makes these two claims. First, that if a person is justified in believing something, then that person is aware of an argument that has that something as its conclusion. Second, that any line of justificatory argument will follow one of the three patterns described in Agrippa’s trilemma. Your task, now, is to explain how these two claims support the contention that no one is ever well enough justified in believing something for their belief to count as knowledge. The question for you to answer is: How does one get from the standard account of epistemic justification plus Agrippa’s trilemma to the conclusion that no one knows anything about anything external to her or his own mind? Take ten minutes or so to compose in writing your answer to the question in the previous paragraph. (Hint: Take the patterns one at a time, and explain how a chain of argument in each pattern will fail to confer strong justification on its target belief.) Do you think that the sceptic is right? Do you think the argument from Agrippa’s trilemma that you have constructed succeeds in showing that no one knows anything about the external world? *** Now that you have completed your task, read on and compare your answer to the answer given below. The question is how Agrippa’s trilemma supports the sceptical contention that no one knows anything about objects or events beyond their own minds. We will take the three possibilities in the trilemma one at a time. We want to show that no matter which of the three patterns attends a target belief, that target belief will fail to count as knowledge because it is
reasoning. In circular reasoning, some line of justificatory argument simply presupposes as true a statement that it must show to be true. For instance, suppose I believe that men have been to the moon, and that I believe this on the grounds that rocks I saw in the space museum are from the moon. I believe that the rocks are from the moon on the grounds that people who have been to the moon brought them back. But that belief implies that men have been to the moon. On what grounds do I believe that men have been to the moon? On the grounds that the rocks I saw in the space museum are from the moon. Now my line of argument is going in a circle. I’m actually presupposing that men have gone to the moon; I have no argument independent of this claim to support that claim. An argument that goes around in a circle is not a justificatory argument for that argument simply presupposes that the claims within the circle are true. Thus, if a line of justificatory argument for a target belief turns back on itself to form a circle, that target belief cannot stand as a piece of knowledge. A line of argument in favour of a belief can only either regress to infinity, end in an undefended belief, or form a circle. But a line of argument in favour of a belief that takes any one of any of these three patterns will not confer strong justification on that belief. Thus, since one must be strongly justified in believing something for that belief to count as knowledge, no one knows anything.
4. Agrippa’s trilemma and the evil scientist argument The evil scientist argument for scepticism about knowledge of the external world is this: 1) If you know that presently you are reading a philosophy text, then you can know that you are not being deceived by an evil scientist into falsely believing that presently you are reading a philosophy text. 2) You cannot know that you are not being deceived by an evil scientist into falsely believing that presently you are reading a philosophy text. Therefore, 3) you do not know that presently you are reading a philosophy text. Since 4) both you and the proposition that presently you are reading a philosophy text were picked at random for this argument (any other person and any other proposition supposedly about the external world would have worked just as well), we must conclude that 5) no one knows anything about the world external to her or his mind. Strictly speaking, the evil scientist argument isn’t really needed to establish scepticism about knowledge of the external world. Agrippa’s trilemma seems all by itself to imply that no one knows anything about the external world. Yet the evil scientist argument is rhetorically very powerful; it brings the sceptical thesis home to a person as no other argument can. Agrippa’s trilemma, though, can play an important role in the evil scientist argument. One can appeal to Agrippa’s trilemma in defending the second premise of the evil scientist argument. Putting the two arguments together can make a person feel vividly that he knows nothing at all about the world around him. The implications for the possibility of knowledge of Agrippa’s trilemma are what make the second premise of the evil scientist argument so difficult to deny. According to that premise, you cannot know that you are not being deceived by an evil scientist into falsely believing that presently you are reading a philosophy text. “Yes I can know that I am not being deceived by an evil scientist!” one wants to say. But each line of justificatory argument in favour of what one wants to say will regress to infinity, halt arbitrarily, or turn in a circle. Any line of justificatory argument that regresses to infinity, halts arbitrarily, or turns in a circle
fails to confer on the target belief justification sufficient for that belief to count as knowledge. So, you cannot know that you are not being deceived by an evil scientist. For all you know and all that you can know, nothing is really the way you believe it is.
