Anscombe on Belief: “Motives for Beliefs of All Sorts”

Elizabeth Anscombe (G. E. M. Anscombe)
G.E.M.  Anscombe (1919-2001) was a British analytic philosopher often credited with initiating the return to the study of the virtues in her famous essay, “Modern Moral Philosophy” and with revitalizing interest in the philosophy of action with her enormously influential monograph, Intention.  In this series of posts, we will explore some of her later writings about beliefs and how our epistemic agency might be part of our overall character formation. Note: This is part one of a three part series. 

The St Andrews Studies in Philosophy and Public Affairs, a series run under the auspices of the Centre for Ethics, Philosophy, and Public Affairs at the University of St Andrews and overseen by John Haldane, has now published the fourth and final volume of Elizabeth Anscombe’s hitherto uncollected and unpublished papers, Logic, Truth and Meaning: Writings by G.E.M. Anscombe, Edited by Mary Geach and Luke Gormally. For those who have been waiting to see what remains of her Nachlass that has been deemed worthy of publication, this is a significant event. Of special interest in this volume are three essays on beliefs, all of which are undated typescripts. These are “Motives for Beliefs of All Sorts,” “Grounds for Belief,” and “Belief and Thought.”


In “Motives for Beliefs of All Sorts,” the question at issue is whether we can have motivated beliefs (that is, beliefs based not on grounds that show or tend to show the truth of the proposition believed, but motives that explain beliefs by reference to some end or good).


That motives can and do play such a role is evidenced by the fact that we sometimes make remarks such as “he was under a strong temptation to think that p” or “out of loyalty, he remains convinced that such and such.” One is only tempted by something one desires, and one desires what one finds (in some sense) good. Moreover, some beliefs, such as the belief that my spouse loves me and only me, may be too painful to give up in spite of compelling evidence to the contrary; contrariwise, some beliefs are so satisfying or so tied to my self-identity that I don’t even see the things that would count as evidence against them. In all such cases, the explanations of what I believe bring in factors about what I want (or what I am desperate to avoid). But all the same, Anscombe concedes that the idea that motives can and do play a role in our belief formation is “a bit queer” (190).


The queerness of the idea of motives for beliefs comes from certain features of the concept of motive that appear to divorce it from our concept of belief. First, Anscombe remarks that it’s “crazy” to announce that one believes something on purpose. If one does something on purpose, this implies that one might have done the thing accidentally (that is, not on purpose). But no one can believe something accidentally. It is absurd to say “By a slip of the believing mechanism I believed p when I meant to believe not p.” Anscombe thinks this reveals a deep metaphysical truth about belief: it is not the product of a mechanism that might misfire – to believe something is to mean to believe it.


It is possible, by contrast, to think of something on purpose in order to achieve some end: one might think of something boring in order to fall asleep, or think of something sad in order to elicit a certain emotion, as actors do. One can also believe something with a further purpose in mind, such as cases of wishful thinking, in which one believes that one will do something – make this free throw shot – in order to bring it about that one actually does it. But, she writes, “believing something on purpose … in the way in which one can think of something on purpose, though not with a further purpose: this concept has no foothold” (191).


It is also impossible to believe something just for the fun of it, or because one is feeling rambunctious or depressed, or for no reason at all. Nor is it possible to obey a command to believe a random proposition for which one has no reason other than the fact of the command. All this suggests that belief is simply involuntary but Anscombe rejects this. The very idea of motives for belief is the idea that one can give a reason for believing that does not point to the truth of its content, and Anscombe has already allowed that this is possible and even common in human life.


It is possible that one believes something because one sees that “it would be better to do so.” An example:


“It is rather beastly to harbor suspicion against a man if one hasn’t got to. It’s better – pleasanter or nobler or better general policy for the sake of human relations, if in the particular case it is not unwise – to think well of someone than ill, or to think well of him than to remember he may equally well deserve to be thought ill of, if one hasn’t got to, so let’s accept his story”(193).


One might be motivated by one’s conception of the good life to believe something, but only if one is faced with a case in which the evidence is underdetermined for believing one way or another. In such cases, one doesn’t dwell or seek out evidence against the belief, and one might have to remind oneself of other interpretations when such evidence is brought to one’s attention. These motives for belief, even though they do not relate to truth but the good, seem to be reasons we can give in response to a doxastic sense of the question “Why?” that do not immediately call the belief into question; for this reason Anscombe declares they are “announceable.” They are also “reputable” because they do not call into question one’s doxastic credentials.


Some motives for belief are of obvious ill repute. An example is “I believe it because I hate him.” This is announceable because in saying it one does not invalidate the claim to believe (though one does show bad character). But some motives are “unannounceable” because putting them forward shows that one doesn’t really believe; an example of such a motive is, “I believe it because, if it is true, the inheritance is mine.”


How can we come up with a satisfactory explanation of the difference between the unannounceable motives and the announceable ones? Anscombe says that of the former group, the general form of explanation is: “it is better, more pleasing to me, if p is the case” (194). This tends to show that one doesn’t really believe it at all, because beliefs cannot be indifferent to the way things are, what is actually true. Contrast this with “I believe that because I hate him.” This is a possible expression of belief, because the general form seems to be that “it is better to believe that p is the case” rather than “it is better if p is the case.” And while the former is of ill repute it is still possible as an expression of belief, whereas the latter is not.


Anscombe thinks that a distinction between belief and believing will help make the subject matter clearer. If I am under the strong influence of someone, a powerful man in my field who is mentoring me, I may be inclined to believe things just because he says them. But if you ask me why, the answer does not involve any appeal to truth, but just that he said it. His saying it leads me to believe. So I have a motive for believing but no ground for the belief. This is possible, but as soon as I admit that I have no grounds, the belief is called into question. In the end, Anscombe seems to be saying that motivated belief is possible but psychologically tenuous. It seems that she is suggesting that the more reflective we are about our beliefs and our reasons for holding them, the less likely we are to be motivated in our beliefs in a problematic way. This in turn would suggest that reflection is not similarly motivated, but we may ask why we are entitled to this assumption.


Tomorrow, Part II: “Grounds for Belief.”

This series originally appeared as “The Capacious and Consistent Mind of Elizabeth Anscombe” in International Journal of Philosophical Studies, Vol 24, 2016, Issue 2.

Jennifer A. Frey is a principal investigator with Virtue, Happiness, and the Meaning of Life.