Axiological Neutrality Research Paper

1. The Logical Foundation Of Axiological Neutrality

The source of axiological neutrality is to be found in a double logical impossibility:

(a) the impossibility to deduce a value judgment from factual propositions;

(b) the impossibility to deduce a factual proposition from value judgments.

A factual proposition is true if it corresponds—and false if it does not correspond—to the empirical phenomenon that it is supposed to describe. A value judgment (‘social state A is better than social state B’), on the other hand, is a preference between social facts, and it does not ‘correspond’ to any empirical phenomenon: it cannot be said ‘true’ or ‘false.’ Since a proposition logically derived from factual propositions is also a factual proposition (true or false), it cannot be a value judgment. Conversely, a proposition logically derived from a value judgment is also a value judgment (neither true nor false), and thus cannot be a factual proposition (Ayer 1936).

The consequence of this dichotomy between facts and values is that social science theories alone are unable to express value judgments. These theories are supposed to provide causal (deductive) explanations of social facts. They exclusively consist of factual and logical propositions: their premises cannot contain any value judgment, because it would prevent them from explaining empirical phenomena (impossibility (b) above). And it is therefore impossible to deduce from these theories any value judgment (impossibility (a)). From this point of view, social science is no different from any other empirical science. In and of itself, a theory seeking to explain the origin of an inequality cannot yield a value judgment approving of or condemning morally this inequality. For the same reason, a physical theory explaining the solidity of a building cannot yield a value judgment in favor of or against this solidity.

In order to prevent a possible misunderstanding, it must be observed that the factual propositions making up the theories of the social sciences may be propositions describing the moral preferences, the esthetic judgments, or the tastes of the agents of the system under study. But when the social scientist describes or explains these preferences, they are considered as mere facts (the fact that an agent has a given preference). It may of course happen that he or she approves of or condemns these preferences, but this value judgment plays no role whatsoever in the explanatory (deductive) theories themselves.

As a consequence of the double impossibility above, social scientists cannot prescribe or recommend a policy—unless they combine the scientific facts and theories with an ethical value judgment specifying the end(s) to be achieved. Given an end E, scientific theories can in principle determine the means M that can be used to reach it (or they can tell us whether this end is impossible to reach). Actually, the proposition ‘If M, then E’—‘If the means M are used, then the end E will be reached’—is a factual proposition: it can in principle be said to be ‘true’ or ‘false.’ On the other hand, propositions such as ‘It is morally desirable to attain end E’ and ‘It is morally desirable to use means M’ are value judgments: they are not factual propositions, and are neither true nor false. In other words, a social scientist can use scientific knowledge of cause and effect in order to determine the means to an end, but not in order to advocate this end or these means from a moral point of view (Simon 1976). It goes without saying that if the theories are wrong or the relevant facts not fully known, the means in question will (generally) not actually permit one to reach the end.

In the 1940s, the founders of the ‘New Welfare Economics’ (economists such as John Hicks and Paul Samuelson) believed that they could overcome the dichotomy between facts and values by resorting to the unanimity principle; that is, by using only value judgments accepted by all the members of the society (‘Pareto-optimality’ seemed to be such a unanimous value judgment). Against this view, a number of objections have been raised. First, even if a value judgment is unanimous, it remains a value judgment and can never be ethically neutral (Sen 1970). Second, really unanimous (and thus nonconflicting) value judgments are probably less numerous than economists would like to believe. Last, such value judgments are morally questionable insofar as they tend to promote the status quo (Rothbard 1997).

2. Axiological Neutrality As A Moral Imperative

This dichotomy between facts and values implies that social scientists have a responsibility towards their audiences, be it their scientific colleagues, the general public (which they may reach through the media), or more especially the ‘captive’ audience of their students. The imperative of axiological neutrality commands social scientists not to make their audience believe that science alone—as a system of factual and logical propositions—can demonstrate the value of their own moral or political convictions. The social scientist, in other words, must not indulge in moral teaching while pretending to be engaged in pure empirical science. In his classic text about axiological neutrality, Weber (1917) emphasized that the professors of social sciences had to respect this imperative, first by being aware of it themselves, and second by making their audience and especially their students aware of the ethical value judgments founding their prescriptions. When social science theories are put into practice, value judgments are necessarily required. Such judgments are not scientific (empirical) propositions, and their existence and content should be explicitly stated—especially when these propositions may elicit interpersonal disagreement.

Weber went very far in the defense of neutrality since he recommended that professors confine themselves to purely scientific presentations, and if possible totally refrain from expressing their ethical preferences in front of their students. But on this point, his conception of axiological neutrality was too demanding. For if empirical theories are devoid of value judgments, the latter cannot be excluded from scientific Activity. A social scientist cannot do science—try and discover the true explanations of phenomena considered as important—without first laying down value judgments. To decide to do science is to express a preference and thus to follow a value judgment. The selection of phenomena considered as important and deserving of a scientific explanation also rests on a value judgment. Social scientists may thus be guided by their ethical or political convictions in the choice of their subject matter: poverty, unemployment, violence, ethnic relations, etc. This is in part what makes the imperative of axiological neutrality necessary, since a clear distinction must therefore be made between value judgments and scientific, factual propositions. Axiological neutrality does not mean that scientific activity can get rid of all value judgments. It is obvious that if scientists had to remain perfectly neutral towards value judgments, they would not be able to choose an interesting subject matter, or to prefer truth to error. Science would simply become impossible (Gouldner 1963, Homans 1978).

Axiological neutrality is a moral duty inherent in intellectual honesty. Transgressions of this imperative may of course originate in dishonesty, but also in honest error. Here is a famous example of an involuntary confusion between facts and values chosen from the field of economics. After the law of marginal utility was discovered in the 1870s, some of the greatest economists at the time (such as Edgeworth and von Wieser) thought they could scientifically justify an egalitarian policy. The law of marginal utility, from which contemporary price theory originates, shows that when the stock of a good possessed by an agent grows, the subjective utility (satisfaction) for this agent of each additional unit of the good diminishes. The first unit is used to satisfy the most urgent need, the second one a less urgent need, the third one a still less urgent need, and so on: each additional unit is allocated to the satisfaction of less and less urgent needs. From this law is deduced the fact that, all other things being equal, the greater the quantity of money (monetary income) possessed by an agent, the smaller the subjective utility for this agent of an additional unit of money. Conversely, the smaller the monetary income of an agent, the greater the subjective utility of an additional unit of money. It would then seem as if the following reasoning could be made on purely empirical grounds: if the State decides to reduce the inequalities of income by a monetary transfer from the rich to the poor, and if this transfer does not reduce production (i.e., does not have any adverse effect on work and investment), then the total utility of society will grow, since the utility lost by the rich will be less than that gained by the poor. Robbins (1935) has shown that this deduction is not scientific: the law of marginal utility can only be applied to one agent, and does not enable us to make comparisons of utility between persons. It is thus impossible to deduce from this law the proposition that the utility of the income taken from the rich is smaller than the utility of the income given to the poor. This proposition is an ethical value judgment—a preference given to what the poor gain over what the rich lose—that has been confused with a factual proposition. In order to deduce the egalitarian policy, this ethical value judgment is indispensable.

3. An Objection Against Axiological Neutrality

Strauss (1953) has raised an important objection against the weberian imperative of axiological neutrality: if value judgments are not scientific (see impossibility (a) in Sect. 1), then they are not based on reason and are thus arbitrary; but if value judgments are arbitrary, why should they be obeyed? Why should social scientists respect the moral imperative of axiological neutrality? Furthermore, if values are arbitrary then scientific activity is also arbitrary since it requires a series of value judgments (see Sect. 2). In other words, if the logical foundation of axiological neutrality is accepted, then the rational basis for the formulation of value judgments is undermined, and scientific activity becomes utterly arbitrary. The ideal of neutrality is supposed to protect science from the unwarranted interference of value judgments, but it results in the destruction of science since it implies that the selection of scientific problems and the practical utilization of science are purely arbitrary. According to Strauss, in order to avoid these nihilistic consequences the foundation of axiological neutrality must be questioned, and it must be admitted that a science of morals, a science of value judgments is possible.

Very briefly, a reply to this argument of Strauss is that the impossibility scientifically (empirically) to derive any value judgment does not imply that these judgments are necessarily irrational. The notion of rationality is not limited to the empirical domain. There also exists a rationality of norms, of values, an axiological rationality in the sense of Boudon (1995): a value can be justified by reasons, by arguments whose soundness and validity can be rationally discussed. Just because value judgments cannot be established on a purely empirical basis does not mean that they are arbitrary. It is true that among the advocates of axiological neutrality are authors believing in the pure irrationality and subjectivity of values (like Weber himself), but there are also among them authors defending the rationality of values (Boudon 1995, Rothbard 1997).

Bibliography:

1. Ayer A J 1936 Language, Truth, and Logic. V. Gollancz, London
2. Boudon R 1995 Le Juste et le Vrai: Etudes sur l’Objectivite des Valeurs et de la Connaissance. Fayard, Paris
3. Gouldner A W 1963 Anti-Minotaur: The myth of a value-free sociology. Social Problems 9: 199–231
4. Homans G C 1978 What kind of a myth is the myth of a value-free social science?. In: Homans G C (ed.) Certainties and Doubts: Collected Papers, 1962–1985. Transaction Books, New Brunswick, NJ
5. Robbins L 1935 An Essay on the Nature and Significance of Economic Science, 2nd edn. Macmillan, London
6. Rothbard M N 1997 Praxeology, value judgments, and public policy. In: Rothbard M N (ed.) The Logic of Action, I: Method, Money, and the Austrian School. Elgar, Cheltenham, UK
7. Sen A K 1970 Collective Choice and Social Welfare. North-Holland, Amsterdam
8. Simon H A 1976 Administrative Behavior: A Study of Decision-Making Processes in Administrative Organizations, 3rd edn. The Free Press, New York
9. Strauss L 1953 Natural Right and History. University of Chicago Press, Chicago
10. Weber M 1917 The meaning of ‘ethical neutrality’ in sociology and economics. In: Schils E A, Finch H A (eds.) Max Weber on the Methodology of the Social Sciences. The Free Press, Glencoe, IL