In a 1957 BBC program, Elizabeth Anscombe was famously asked “Does Oxford moral philosophy corrupt the youth,” to which she ironically answered “No.” Instead, Oxford moral philosophy was already “perfectly in tune with the highest and best ideals of society.” Given that these ideals were already perverse and corrupting, there was no more damage Oxford moral philosophy could do. In “Modern Moral Philosophy,” Anscombe challenges not only her fellow academics, but the values of society at large. She does so with the following three theses.
- It is not worth doing moral philosophy until we have an adequate philosophy of psychology (1).
- There is no specifically moral sense of “ought” – concepts of moral duty and moral obligation are remnants from a law conception ethics that no one seriously considers (1, 6).
- The differences between Sidgwick and modern Oxford moral philosophy are so minute, and their similarities so strong, they might as well all be grouped under one term: “consequentialist” (1,12).
In addressing these theses, I find it best to work in reverse order. I want to focus on the following train of argumentation. (3) The prevailing moral theory of our time takes right action to be that which produces the most favorable state of affairs. Call these moral theories “consequentialist.” A consequentialist ethics fails to produce moral absolutes. Insofar as we care about these absolutes, we should abandon consequentialist moral theories. Deontology is no better a moral theory to subscribe to. For though it can generate absolutes, it does so on the basis that right action is that which one has a moral obligation to do or not do such-and-such an act. But this specifically moral obligation is a survival from divine law ethics. (2) Deontology needs a God figure to ground the morally obligatory sense of the word ought. If we want to talk about moral absolutes without a God figure, then one should look to Aristotle. In Aristotle’s virtue ethics, illicit action is that which contradicts the set of virtues, where Aristotle saw no need to talk about a God. (1) But, before we can do this, one needs a requisite philosophy of psychology to describe human nature, action, what a virtue is, and human flourishing. Once these are in place, one could formulate a virtue ethics around the ordinary sense of ought, generate absolutes, and not have to deal with a God figure.
Third Thesis – The Attack on Consequentialism
- What is consequentialism? Page 9
Consequentialism is that set of ethical theories which takes “right action” to be that which produces the best possible consequences. That is, regarding whether I should do A or B, I should not look to A or B in themselves, but to what A or B will cause. Anscombe does not attribute this exact formulation to any philosopher. Rather, this is the first step of any consequentialist theory. To undermine this first step is to undermine the entire consequentialist project. In a footnote, Anscombe addresses Oxford Objectivists, who do acknowledge the intrinsic value of an act. But, this difference is only superficial. Objectivist theories collapse into standard consequentialism and so fall prey to the attack Anscombe begins in this paper. I would extend this to the divisions between act and rule utilitarianism as well. These theories are more similar than dissimilar.
- Can consequentialists reconcile absolute prohibition? Page 10-11
If the criteria for what one should do in some situation is the state of affairs they will bring by acting in such a way, then one is always left with a robust set of “live” options. Hypothetical scenarios can always be constructed for this person such that some especially distasteful act might be the “right” thing to do. That is, under consequentialist logic, there might be some scenario where murder, torture, willful conviction of the innocent, etc. is the right thing to do. We might consider someone for whom these are even an option a rather vicious person. These consequentialist theories are in direct contradiction to the Christian ethic. When the commandments say “Thou shalt not murder,” it’s meant to be understood that there is no conceivable scenario in which murder (being a distinct concept from “killing”) is the right thing to do, regardless of the consequences. The consequentialist and Christian ethic are irreconcilable on this front. (Candace Vogler has a great paper on this topic titled “In Defense of Moral Absolutes.”)
- Millian utilitarianism – Page 3, 9-10
Anscombe denotes a distinction between consequentialism and the utilitarianism she attributes to Mill. Mill is more sophisticated in his regarding of an act in itself. Mill conceives of acts as falling under a principle of utility quite separate form their actual outcomes. He conceives of an analytical principle of utility such that acts of certain kinds might ultimately be prohibited. If a particular action falls under this principle, then it should not be pursued even if it produces good consequences. (Cora Diamond has a paper discussing this titled “Consequentialism in Modern Moral Philosophy.”) Though, for Anscombe, it’s not clear how one might conceive of some act falling under any one principle of utility, given that one needs some singular description of the action. Any act may be described in ways such that it falls under multiple principles of utility. If this is the case, then one could always morally cheat their way out of a vicious action description and into an acceptable one. Though you may be able to describe my act as coming under the principle of X, I can also describe it as Y, and so justifies my action. It can come under some aspect which justifies my action. (Bernard Williams, in Morality, makes the further point that these principles of utility are subsumed under the utility principle, and so if cheating under this species of action were to bring more utility, and the agent knew this, why would they follow the convictions under the genus of action?)
- Intended vs foreseen consequences – Page 11-13
Anscombe’s secular concern with consequentialism begins with Sidgwick’s treatment of the doctrine of double effect, or, consequences I foresee but do not intend. When right action concerns consequences, so too does praise and blame. I.e. bad consequences are worthy of blame, good ones are worthy of blame. So, am I worthy of blame when my action produces ill consequences? It seems that consequentialists are willing to admit one isn’t worthy of blame if one did not foresee these consequences. (It’s worth acknowledging Anscombe’s point that in many cases, if foresight is all that’s necessary for blame, then all one needs to do to avoid reproach is to show how they could not have foreseen such a consequence. And if foresight is the requirement, then how is one expected to know what will happen in unfamiliar circumstances or those with complex causal relations? An agent doesn’t see with perfect accuracy the consequence of their action but should know perfectly well their intentions. The consequentialist can’t say in any particular practical case “Don’t do X,” only “Don’t bring about consequences xyz” as morality is not located in any doing, but what your doing brings about, and so in difficult cases, consequentialism remains silent.) But what about if I foresaw these consequences and did not intend them? In another paper, Anscombe uses the example of two bomber pilots to elucidate this distinction. Both are tasked with bombing the same location. In the first case, the bomber is told to bomb this area in order to destroy a weapon cache. Destroying this cache means killing the innocent people living in the surrounding area. In the second case, the bomber is told to bomb the cache in order to kill innocents and stir up fear. In both cases, killing innocents is foreseen, but only intended in the second case. When one collapses this distinction as Sidgwick does, defining intentional as that which one can foresee, then one has to hold the second case as equal to the first.
Second Thesis – The Moral Sense of ‘Ought’
- Moral sense of the word “ought” is a holdover from divine law? Page 2, 5-7
There is a popular notion in deontological moral philosophy of the obligatory ought. This concept holds that there is some objective moral imperative for which one is required to oblige should you want to be moral. That is, there is some set of imperatives with objective grounding which have the weight of moral goodness and badness. Adherence to these imperatives is to be morally good and diversion morally bad. Deontological theories have a linguistic similarity to divine law conceptions of ethics, the difference being the objective grounding in divine law is a God figure who promulgates the law. (It’s not clear in Anscombe’s account if divine law theory is a suitable substitute to consequentialism or deontology. Anscombe was a catholic, often times writing to catholic audiences, and merely states such a view would not be generally held.)
- What’s at stake for deontology without a God? Page 2-3, 13-14
Deontological ethics without a God figure has no leg to stand on. There are two strands of deontology without a God which Anscombe attacks: Kantianism and contractualism. Both of these ethical theories need to find a way to replace the objective grounding that God provided. She makes short shrift of both. Kant tries to utilize the idea of self-legislation. For two explicit reasons, Anscombe denies this solution. Legislation assume a legislator, and in order for one to independently respond to legislation, it must be issued with authority. This means there must be a power of legislator not available to legislatee. Second, one cannot legislate to themselves, because there is no possibility of disobeying that legislation. Related to the issue of action description relevant in Mill, if one thought one was acting under a certain maxim, then he just would be acting under that maxim since there is no independent criterion for determining whether or not one was following such legislation. Then, there is contractualism, the aim to replace the objective grounds of God with a contract. In divine law theory, this law, this contract, is promulgated to man by God through man’s knowledge of good and bad. In social contract theory, there is no such promulgation of the laws. Further, one must know one is contracting in order for a contract to be binding. There is no explicit knowledge that one is contracting at all. Anscombe speculates one might be able to work out a philosophy of implicit contract through the use of shared language, institutions, and signs, i.e. that participation in various shared forms of organization are ways of contracting without the need of an explicit signature. However, she is skeptical this speculative account of implicit contract will have anything substantive to say about particular practical circumstances.
- The return to Aristotelian ethics – Page 14-16
There is still a line for us to take in the search for a secular, prohibition-based system of ethics. There is hope in replacing the objective grounding of God with some form of existing norms. The first attempt Anscombe explores is the norms of society. These fail in light of the fact that they are not suitably “objective” to be the objective grounding for a system of ethics. We could imagine the norms of society to be anything, even things we find distasteful. There is still another set of norms, the norms of man. Just as man has a certain number of teeth, he has a certain set of virtues. So, while this man might not have that many teeth, man considered as a logical object, not a biological one, sets a standard of normativity. By having knowledge of man in this way, we have the requisite grounding for a new system of ethics. Yet, when we make this transition in the search for divine law without a God, we’ve lost our original sense of “law.” So, it is in this sense that they aren’t “moral” at all, or at least not moral in the way that deontologists use the word. For the way that it’s used in “one ought to be virtuous” is the same ought as “he ought to have this many teeth.” The first is no more “morally” ought than the second.
First Thesis – The Need for Moral Psychology – Page 1-5, 14-16
- The naturalist fallacy and brute facts
In proposing a replacement of God with norms about man, Anscombe now turns to addressing Humean concerns, sometimes called the is-ought problem, as well as refuting G.E. Moore’s naturalistic fallacy. Anscombe anticipates uneasiness towards her proposal to look to man’s norms for a grounding in ethics as violating one of these two canonical principles. My formulation of the problem she faces is the following. There are statements about what is and there are statements about what ought to be. Statements about what ought to be cannot be derived from the way things are. Things are the way they are, and we can make true judgments about those things. But no combination of true judgements about the way things are will add up to a true judgement about the way things ought to be. Anscombe introduces brute facts and institutional facts in response to the is-ought problem. She wants to first show how we can make the transition from “is” to “owes,” or “is” to “needs,” and these will show us why we can make the transition from “is” to “ought.” If we look to certain “is” statements, we will notice a relation of bruteness where one is more primitive to the other. For example, a sack of potatoes arriving at my door when I have requested them is brute relative to having bought a sack of potatoes. That is, the given circumstances (1) I requested potatoes, (2) they arrive at my door, and (3) the background institutions which constitute exchange, all add up to constitute the fact that (4) I bought potatoes. Now, the fact that I bought potatoes combined with the background institutions of business and payment constitutes the fact that I owe the grocer money. In this way, we started with the factual statement “there is a sack of potatoes at my door” and arrived at the statement “I owe my grocer money.” The first is brute relative to the second, but we were able show the truth of the second by means of the first. Similarly, the fact that a plant is arranged in such a way is brute relative to it needing sunlight and water, but the truth of the first is part of the relation of bruteness to which we conclude the second. If it were not the case that is is such a way, it might not need such a thing. In light of our discarding the moral sense of the word ought, it has the same sense as a plant needing water, according to Anscombe. Through this relation of bruteness, we should be able to arrive at a similar process for transitioning from is to ought. I.e. man is a certain way and so needs the virtues and so ought to be virtuous the same way a plant is a certain way and so needs water. (Anscombe addresses brute facts explicitly in her paper “On Brute Facts.”)
- The gap in ethics and the place of philosophy of psychology
In order to get to the place where we can say a man ought to be virtuous, or, an unjust man is a bad man, we need to take a step back from ethics and look to the philosophy of psychology. Until we have some requisite concepts in place, our virtue ethics can’t get off the ground. Among these concepts are a description of human nature, human action, what a virtue is, and human flourishing. That is, we need to substantively nail down substantive claims about human nature in order to get practical advice, we need to know what human action and the virtues are such that we know why certain actions are prohibited and what it means to intend them, especially since the list of illicit actions are those which fall contrary to the virtues. Anscombe begins this project in Intention, what she considers a book containing no ethics, and you can see similar attempts at filling this philosophical gap in Philippa Foot’s Natural Goodness or Michael Thompson’s Life and Action.