A Summary of Anscombe’s Criticisms
G.E.M. Anscombe, a Christian philosopher and student of Ludwig Wittgenstein, presented a paper critical of Lewis’s argument from reason, as presented in his book Miracles, in February 1948 to the Socratic Club in Oxford, and which was published later that year as the premier essay in The Socratic Digest. She was not the first to criticize Miracles,{1} but the objections she raised were by far the most significant. Her essay primarily argues that Lewis employs terms with dubious definitions, and when they are corrected, the argument no longer holds.
Irrational and nonrational
Anscombe quotes Lewis’s dictum that “no thought is valid if it can be fully explained as the result of irrational causes.”{2} However Lewis chose examples where the natural processes which influence someone to reach particular beliefs are known to mislead him into false beliefs (such as delirium tremens). It simply does not follow from this that all natural processes do so. “A causal explanation of a man’s thought only reflects on its validity as an indication, if we know that opinions caused in that way are always or usually unreasonable.”{3} “… it is only because we already know that men with delirium tremens see things that are not there … that we dismiss a man’s belief by ascribing it to delirium tremens.”{4}
Part of the problem here is that Lewis uses the term “irrational” in too broad of a sense. An irrational cause for a belief would be an invalid argument that lets one believe what one wants to believe. On the other hand, when we say that a belief is caused by something like delirium tremens, these are not irrational in the same sense as the former causes are. Rather, “they are conditions which we know to go with irrational beliefs or attitudes with sufficient regularity for us to call them their causes.”{5} Such natural causes may be “non-rational” in the sense that they are just bare facts rather than propositions, and as such, have no truth-value. This is not the same thing as being irrational, however. By conflating irrational causes and nonrational causes, Lewis has committed a category mistake, and this calls his argument into question.
Anscombe illustrates this with Lewis’s example of a man afraid of a dog. If a man says a dog is dangerous and, when asked how he knows, gives insufficient grounds for this belief, it is irrational. However, if a man says a dog is dangerous and, when asked how he knows, begins to tremble and shake his head, his belief is not based on insufficient grounds: it is completely groundless. It is caused by some event in his past or some dysfunction of his psyche, and so is nonrational rather than irrational.
Paradigm case
Anscombe goes on to challenge Lewis’s claim that we have to believe in the validity of reason. “You can talk about the validity of a piece of reasoning, and sometimes about the validity of a kind of reasoning; but if you say you believe in the validity of reasoning itself, what do you mean?”{6} Here, Anscombe is challenging Lewis’s claim that if we can call an isolated belief irrational if it springs from irrational causes, we can equally call all of our beliefs irrational if they are all the result of irrational causes.
Her point is that in order to understand valid reasoning, we would have to have an example of it.{7} As such, to question the validity of all reasoning appears nonsensical, since it would imply the possibility of there being no valid example of reasoning, and so no concept of valid reasoning could ever be formed. Moreover, part of our understanding of valid reasoning comes from contrasting it with invalid reasoning. Yet we would need at least one example of each in order for such a contrast to take place. “Anscombe here is employing the Paradigm Case argument, an argument against the possibility of meaningfully raising certain skeptical questions.”{8}
Antony Flew -- an atheist philosopher who participated in the Socratic Club, and even took part in a debate on Christianity and Plato later that month -- illustrates this by comparing it to hallucination. There is nothing problematic about asking whether a particular perception is hallucinatory or real. “But it is preposterous to ask whether all perceptions taken together are hallucinatory. The term ‘real perception’ and the term ‘hallucinatory perception’ derive their usual significance from their mutual contrast, and from the tests used to decide which is applicable.”{9} Suggesting that all perceptions might be hallucinations is not just a ridiculous claim: it is incoherent. “Hallucination” does not mean anything without the contrast of real perception. Similarly, an irrational belief does not mean anything without the contrast of a rational one.
Naturalistic explanations
Anscombe then asks what exactly in the naturalistic worldview would prevent any of the reasons for a belief from applying. If we are asked for an explanation of a particular belief, “what in the naturalistic hypothesis prevents that explanation from being given and from meaning what it does?”{10} The naturalist scenario, at least as naturalists understand it, does not preclude someone believing something and giving an explanation for this belief when challenged.
In fact, this leads to a great irony in Lewis’s argument. He condemns attempts to refute beliefs based on their allegedly irrational credentials, such as Freudians claiming that traditional beliefs are the result of psychological processes in the subconscious. This, however, is presumptuous and inappropriate. We should judge a belief on whether or not it is true; any other factors are simply irrelevant.{11}
Yet after condemning this fallacy, he turns around and commits it himself. Any belief with irrational causes is thereby invalidated; so if all beliefs have irrational causes, all beliefs are invalid. Recall his claim that, if naturalism were true, “The finest piece of scientific reasoning is caused in just the same irrational way as the thoughts a man has because a bit of bone is pressing on his brain.”{12} But to dismiss a belief because it has irrational causes is precisely the Bulverism fallacy. To argue that a belief is made invalid by such irrational causes “does not follow at all. Whether his conclusions are rational or irrational is settled by considering the chain of reasoning that he gives and whether his conclusions follow from it.”{13} Regardless of whether someone reached a belief for irrational or nonrational reasons, we cannot, on those grounds, reject it. The validity of a particular belief is not determined by whether the person who drew it did so in accord with the correct logical procedures, but whether we can do so upon further investigation.
Reasons and causes
This leads to the heart of Anscombe’s critique, that one type of explanation does not rule out another type. Lewis had assumed that giving a naturalistic explanation of a particular belief was incompatible with giving a rational one, in fact that any given phenomenon has only one complete explanation. Anscombe argues to the contrary that the rational and naturalistic explanations are just two different ways of describing the same phenomenon; both can be correct simultaneously.
As a Wittgensteinian, Anscombe was content to view different types of explanations as different “language games.” A “full” explanation would be one which fully answers the questions of an inquirer. As Victor Reppert writes,
… there can be many explanations for the same event. For example, if we ask, “Why is the soda can sitting on the bookshelf?” I can answer correctly, “Because I put it there yesterday,” or “because I wanted it to be recycled,” or “because no one has knocked it over,” or “because the shelf holds it up,” or “because of the law of gravity,” or even “because it is cylindrical,” which explains why it stays put on the bookshelf and doesn’t roll around. We must admit, with Anscombe, the question-relativity of explanations and also that different explanations can be given for the same event.{14}
Such explanations “are not mutually exclusive. They are not even in competition.”{15}
Lewis, essentially, has confused the causes of a belief with its grounds. His failure to distinguish between irrational and nonrational causes leads him to use other terms in an ambiguous manner as well, specifically the terms “reason,” “why,” “cause,” “because,” and “explanation.” When we ask why someone believes something, we can answer in terms of what caused the belief, or we can answer in terms of what grounds the belief. The former would yield a nonrational answer, while the latter would yield either a rational or an irrational one. Both answers would begin with “because,” but would be a different type of explanation.
Anscombe’s point is that these two types of explanation do not contradict each other. They are merely describing the same thing from different perspectives. Anscombe even adds on to the naturalistic and rational explanations of a belief two more: one can give a psychological explanation why one has a belief, and one can give a personal history explanation of why one has a belief.{16} None of these explanations are in competition with each other, and all can be true of the same belief simultaneously.
Anscombe goes into some detail about the difference between a causal-type of explanation and a grounds-type of explanation. Lewis seems to be arguing that if a belief has an irrational or nonrational cause, the person believing it did not reach that belief by reasoning. If the belief turns out to be true “we regard it as accidental.”{17}
But this is not the role that reasons play in our beliefs. Causes are mechanical, physical regularities based on observation. Reasons “are what is elicited from someone whom we ask to explain himself.”{18} So, in contrast to Lewis’s claim that reasons are “a special kind of cause.”{19} Anscombe seems to think that reasons are not causes at all. Thus, reasons and causes are completely different spheres, different language games, that have nothing to do with each other. “It appears to me that if a man has reasons, and they are good reasons, and they are genuinely his reasons, for thinking something -- then his thought is rational, whatever causal statements we make about him.”{20}
In the notes of the discussion following Anscombe’s presentation, and a supplemental comment by Lewis,{21} he acknowledges the difference between causes and grounds, but argues that a belief could only be considered rational when its cause is the recognition of its grounds. If one only arrives at a belief because of causes that have nothing to do with the grounds, then it seems that the grounds for that belief play no role in one’s holding of it. The final assessment of those in attendance was that Lewis “would have to turn his argument into a rigorous analytic one.”
Summary
Anscombe’s critique can be reduced to the following points, not necessarily of equal importance, and some of which bleed into each other:
1. Lewis conflates different types of nonrational processes: just because some nonrational processes lead to false beliefs, it does not follow that all do so.
2. Lewis conflates nonrational processes with irrational processes. If his argument is that having irrational causes for a belief invalidate it, it does not follow that nonrational causes do likewise.
3. The paradigm case argument. Suggesting that a position would invalidate all reasoning is nonsensical, since we would not be able to understand what these terms mean without examples of both. It would erase the distinction between valid and invalid reasoning. However, this distinction may be the only way we can understand what valid or invalid reasoning is.
4. There is nothing in the naturalistic worldview which would prevent a person from giving a rational explanation for a belief and meaning it.
5. Bulverism. Lewis, ironically, commits his own fallacy. A belief is not justified by how it was formed, but by whether or not it is true.
6. Lewis fails to distinguish between a belief’s grounds and its causes. Both may use the same terms (“reason,” “why,” “cause,” “because,” and “explanation”), but they are two distinct types of explanation that do not preclude each other.
In the next post in this series I will analyze these points in more detail.
Notes
{1} See, for example, Robert Eisler, Review of Miracles: A Preliminary Study, by C.S. Lewis, Hibbert Journal 45 (1946-47): 373-77.
{2} C.S. Lewis, Miracles: A Preliminary Study, 1st ed. (London: Bles, 1947), 27.
{3} G.E.M. Anscombe, “A Reply to Mr C.S. Lewis’s Argument that ‘Naturalism’ is Self-Refuting,” in The Collected Philosophical Papers of G.E.M. Anscombe, vol. 2: Metaphysics and the Philosophy of Mind (Oxford: Basil Blackwell, 1981), 231.
{4} Ibid., 224.
{5} Ibid., 225.
{6} Ibid., 226.
{7} It should be pointed out that Anscombe leaves this point open: she writes, “Whether you would adopt this method or some other (though I do not know of any other) …” (ibid., 226)
{8} Victor Reppert, “The Lewis-Anscombe Controversy: A Discussion of the Issues,” Christian Scholar’s Review 19 (1989): 37.
{9} Antony Flew, “The Third Maxim,” The Rationalist Annual 72 (1955): 64-65.
{10} Anscombe, “Reply to Mr C.S. Lewis,” 226.
{11} C.S. Lewis, “‘Bulverism’: or, The Foundation of 20th Century Thought,” in God in the Dock: Essays on Theology and Ethics, ed. Walter Hooper (1970; Grand Rapids, MI: Eerdmans, 1996), 271-77; idem, “Meditation in a Toolshed,” in God in the Dock, 215.
{12} Lewis, Miracles, 1st ed., 28.
{13} Anscombe, “Reply to Mr C.S. Lewis,” 227.
{14} Victor Reppert, C.S. Lewis’s Dangerous Idea: In Defense of the Argument from Reason (Downers Grove, IL: InterVarsity, 2003), 66.
{15} John Beversluis, C.S. Lewis and the Search for Rational Religion (Grand Rapids, MI: Eerdmans, 1985), 74.
{16} Anscombe, “Reply to Mr C.S. Lewis,” 230-31.
{17} Ibid., 228.
{18} Ibid., 229.
{19} Lewis, “Bulverism,” 275.
{20} Anscombe, “Reply to Mr C.S. Lewis,” 229.
{21} Ibid., 231-32; C.S. Lewis, “Religion Without Dogma?” in God in the Dock, 144-46.
(cross-posted at Agent Intellect)
Discuss this post at the Quodlibeta Forum