Weber and Nagel on Values in the Social and Natural Sciences

Due to their different subject matter, the way in which social and natural science inquiries are conducted differs. For some, this difference is constituted in a greater reliance upon values in the social sciences than in the natural. This essay, on the other hand, aims to demonstrate that both schools of science are equally value-laden. This is undertaken via an exposition of Weber’s arguments for social science as more value-laden than natural science, followed by a critique of these arguments through the lens of Nagel’s discussion of social scientific inquiry and accusations of its value-ladenness.

Weber states that the social sciences are those that deal with the “social-economic” elements of life, broadly construed, and that the property of being “social-economic” is not something that an event possesses objectively, but only by virtue of its “condition[ing] our cognitive interest” through being of some cultural significance.1 These “social-economic events” are effectively infinitely multiple because of the nature of social scientific inquiry, such that “we wish to understand on the one hand the relationships and the cultural significance of individual events in their contemporary manifestations and on the other the causes of their being historically so and not otherwise” which leads to, upon investigation, the presentation of “coexistingly emerging and disappearing events, both ‘within’ and ‘outside’ our selves.”2 This multiplicity remains, for Weber, undiminished when a particular object is identified, such as a single act of exchange in economics.

Due to this, all of the analysis that the human mind is capable of consists in an assumption that only a finite portion of reality is worthy of being known. In natural science, this criterion of worthiness is given by the ability of the scientist to produce laws pertaining to an event: as an example, the fact that pure water boils when molecules within the liquid reach a certain kinetic energy and break intermolecular bonds is a relevant scientific fact, the nature of the container of the fluid is not. Such an approach to social science can only have heuristic use, as the matters at hand are not agency-free molecules, but rather things of cultural significance. The significance of an event cannot be given by a series of analytical laws, as the term ‘significance’ implies a certain value-orientation towards an event.3 This is not to say, however, that nomological knowledge is not possible in the social sciences: just that laws cannot explain that significance attached to cultural events and thus are insufficient for explaining the causes of events that only happen due to their cultural significance.

Weber goes on to argue that value considerations permeate the social sciences through the selection of problems and the determination of conclusions, two of the four avenues through which science can be affected by value judgments laid about by Nagel.4 By virtue of investigating something which she attaches significance to, the social scientist allows value-orientation to dictate to the selection of problems: there is no way that a selection from the Weberian infinity of social events could be made without value judgment given that the defining feature of social phenomena is, for Weber, its cultural significance.5

For Weber, this is no[t so much of an issue for the natural sciences as he holds that the defining virtue of ‘relevance’ in the natural sciences is that of the ability to provide laws for events. So far as determination of conclusions is concerned, Weber states that in the method of social science investigation, the guiding “point of view” is of great importance for the construction of the conceptual scheme to be used in the investigation6: even though he states that the social scientist would be bound by norms of scientific thought in the use of their methods, the ability to construct a conceptual scheme to their liking would enable value-biases to enter the way in which groups being investigated were being classified.

For example, if I were investigating unemployment and I was of the bias that the unemployed are such due to their laziness, I would be able to construct a method of investigation that had a criterion of unemployment as being extremely long term, a measure perhaps more likely to provide a group of ‘lazy unemployed’ for investigation. Within the natural sciences, however, this is less of a problem due to the existence of the system of SI units and standardized instrumentation in the practice of science.

It may be, however, that the attention paid by Weber to the social sciences as a discipline affected by these value biases exclusively is misleading. The view of natural science problem selection as being based on what one can determine laws for seems problematic: the only available basis for choosing to perform a study to find a law would be observation of regularity in occurrence. If one could base a natural scientific study on such a premise of observed regularity, it seems somewhat arbitrary for Weber to state that social scientific studies cannot be prompted by the same sort of observation.

It could be accepted that a law cannot be found that explains the causal mechanisms behind a social event, due to the lack of susceptibility of cultural significance to analytic law-making, but it does not follow from this that the interest in the event in order for it to be investigated comes from the cultural significance of the event. Watching people walk into a Job Centre, for example, may hold interest for me as a social scientist even abstracted from the social phenomenon of unemployment: the regularity of appointment lengths may be what interests me, and it is unlikely that tight scheduling has a great cultural significance beyond the practicalities of dealing with people and then dismissing them.

This is still indicative of value choice in problem selection: the social scientist is still investigating what interests her. The importance of this is easily brought into doubt however: one could say the same of a natural scientist who chooses the field of biology over chemistry: as Nagel points out, the way in which the problem for investigation is decided has no bearing on the internal logic of the investigation.7

Weber’s claims of social science necessarily coloring its conclusions due to the social scientist having full control over a conceptual scheme for analysis are also fairly easily subverted: whilst it may be that social scientists have a more granular control over their methodology, the way in which experiments and data analysis are performed (even in line with the ‘norms of thought’ in the field insisted upon by Weber) in natural science may also be colored by personal hopes, aspirations or even the ‘reforming zeal’ mentioned by Nagel of social science.

As an example, natural scientists may be able to bend the norms of thought slightly in order to not classify a result as anomalous, where this result would undermine their conclusions. Even if, however, we were to allow that social sciences were more prone to such manipulation, it is debatable as to whether this additional level of value-ladenness would actually make a difference: if one takes science as a social enterprise, in line with Nagel8, then over time, value biases would be leveled out by competing claims and the ‘truth’ be arrived at by a deductive method comparing results of studies.

In terms of the identification of fact, it has been argued that sociologists or psychologists use value-appraising terms where they claim to be using descriptive, characterizing value judgements.9 As an example, for a psychologist to say that someone is deceitful is to appraise her character, rather than merely describe an attribute: her being deceitful necessarily has negative connotations and the psychologist is imposing them upon the subject here. To put this in Weber’s terms, it may be that the investigation of something ‘culturally significant’ is necessarily going to result in the standpoint of the investigator coming to bear on what should be descriptive, characterizing value judgements: the social scientist, as a concretely socially situated individual, cannot abstract himself from her position when investigating something of importance to her. It is the case, however, that the same can be said of a physicist who claims a pump to be ‘inefficient’: it is not just a claim about the character of the pump, it is an appraisal of its utility to the scientist. In such a light, the value-ladenness of statements in the natural sciences seems to match up with those of the social.

Through the lens of Weber’s account of the value-ladenness of social scientific inquiry, it has been shown that the natural sciences are generally just as describable as value-laden as the social sciences. This is the case as both disciplines necessarily involve a selection of a problem for investigation, analysis of findings and the drawing of a distinction between fact and value when presenting results: this necessitates value-ladenness as people will always require some basis to choose subject matter, people will be incredibly likely to become emotionally invested in their research projects and the fact-value distinction is an untenable one due to the linguistically synonymous nature of characterizing value judgements and appraising value judgements.

---

Endnotes

1. Weber, M. (1949). On Methodology of the Social Sciences, p. 64.
2. Ibid, p. 72.
3. Ibid, p. 76.
4. Nagel, E. (1961). ‘The Value-oriented Bias of Social Enquiry’ from The Structure of Science: Problems in the Logic of Scientific Explanation, Hackett Publishing Company.
5. Weber, op cit, p. 82.
6. Ibid, p. 84.
7. Nagel, op cit, p. 486.
8. Ibid, p. 489.
9. Ibid, p. 494.