This note follows a social media discussion revolving around the implications of postmodern analyses and stances on the matter of knowledge, its social construction and its objectivity conditions. It intends to develop views that I expressed in response to a tweet by Helen Pluckrose, who wrote two articles on the matter, to which this also means to be relevant (although leaving many other interesting aspects aside), as the tweet itself could hardly be seen as anything more than a pretext. Since the aforementioned views have been described by other users as, among other things, poorly formulated, I will now present them in a fashion as orderly, clear and well developed as possible.
The task at hand is the examination of the following claim :
If knowledge is culturally constructed, then it cannot be true.
NB : It appears, in light of the first reactions to this paper, that I should make one or two remarks preventing what seem as tempting misreadings.
I do not regard the claim under examination as a “postmodernist thesis”. Actually, I’ve only seen it explicitly asserted (cf. above links to social media content) by self-designated critics of postmodernist thought, who argued that, since truth was to be allowed to knowledge, then one should deny that it be culturally constructed.
Throughout my close reading of reference texts tagged as “social constructionist” or “postmodern”, I did not find such a metaphysical claim so clearly asserted. I regard as plausible that a version of this claim is operant implicitly, as an element of postmodern folklore, but refuting the claim itself as a “postmodern thesis” would be attacking a strawman.
My only concern is to explore what consequences someone committed to modern epistemic norms should draw from the description of the influence of social determinants on the development of science and said norms.
Namely, I show that in general, an account of the social construction of an epistemic norm, or of the acceptance of a theory, is not sufficient to justify skepticism towards said norm, or theory, on the part of one previously committed to them.
I. Definitions and assumptions.
I will first proceed with defining the terms involved.
A piece of knowledge is a belief, the truth of which the holder of the belief is committed to. One may wish to add that this belief also must be true, but this will not be needed to refute the claim under examination, so we will rely on these necessary conditions only.
Traditional and modern western philosophies, in which the notion of truth associated with the relevant notion of knowledge have been developed and discussed, have known several theories of truth, among which the theory of correspondence, the reliance on the compelling power of “natural light”, the criterion of robustness to falsification attempts.
Any notion of truth relevant to scientific knowledge can be defined by a set of epistemic norms. These are norms which prescribe a manner of inquiry, a method. They include logical norms, norms regarding how to acquire reasoning premises from experience, norms regarding how to test and revise conclusions, and maybe even norms regarding overall features of theories such as generality or conciseness. To believe that a notion of truth is the right one can therefore be understood as no more, nor less than committing to a set of epistemic norms.
Finally, one will be said to be committed to an epistemic norm or method, if, when faced with the alternative of believing or not believing the statement of a given proposition, one is concerned only about whether the epistemic norms that one commits to, i.e. that constitute the right notion of truth, have been enforced in the acquisition of this proposition. The choice of not believing a statement or not committing to an epistemic norm will be called skepticism.
I will understand culturally constructed and socially constructed as equivalent, since the meaning of the latter is more documented, and I will try to keep the analysis at a level of generality at which they can be used interchangeably.
Judging from the relevant literature, one can reasonably consider that there are two main interpretations of social constructionism, one as an area of study, the other as a metaphysical stance.
According to the first one, social constructionism denotes the methodological choice of studying the commitment to norms or theories as social or cultural phenomena, i.e. as caused, determined or influenced by specifically social or cultural factors. These cultural factors include power relations between individuals, cognitive biases, enforcement of social orders or hierarchies, or the influence of social norms seen as distinct entities. In other words, it purports to study, not the logical or metaphysical justifications of knowledge, rather the mechanisms of its development in society.
Social constructionism as a metaphysical stance can be understood as stating that a phenomenon which can be shown to have been socially constructed could have manifested otherwise ; that the way it actually manifests is contingent rather than necessary. Secondly, it approaches social phenomena as the result of the social factors, studied in the manner corresponding to the former definition of social constructionism. Finally, it understands these social factors as the oppression of a social group over another, the intrinsic contingency of which justifies attempts to realize the oppressed manifestations of the phenomenon.
The second notion of social constructionism seems to mobilize the sort of inquiry corresponding to the first notion, prior to applying a metaphysical categorization : that which can be shown to be socially constructed can be said to be contingent and open to change. Going into more details, establishing whether this second take on social constructionism supposes that all social determinants are “contingent” in themselves, or whether only some are of the contingent kind, (for instance those pertaining to power relations), and other are of a necessary kind, is a complicated matter, that seem to admit nuanced interpretations.
However such nuances would be of limited interest to us, as assertions constituting an analysis of the social construction of a norm are themselves sentences belonging to the social sciences, and can be subjected to the same epistemic norms. Yet notions of necessity and contingency are remote from the modern scientific mindset, which deals in relative degrees of influence – if not mere dependence. This is the case whether the phenomenon under scrutiny is the commitment to a social norm, or a material phenomenon. I will thus consider that the only additional question raised by the second notion of social constructionism is that of there being one “necessary” norm of knowledge, and potentially only one “adequate” theory of any phenomenon. Necessity will here be understood as the eventual convergence of any theoretical effort to understand a given phenomenon, assuming that a phenomenon can be well defined independently of its theoretical formalization. If one were to demonstrate this convergence to a sole theory and a sole epistemic norm, one would show that it is not a social phenomenon of a contingent kind, but of a necessary kind.
II. Interpretations of the claims according to the definitions.
Given that one is committed to seek knowledge, when should one recognize that one’s belief is false, and thus fails to be knowledge ?
It may be the case that a justification of a proposition is not held, because the methods were not applied, which would provide justifications according to the norm of truth-seeking one is committed to. It may also be that one did follow a method one is committed to (the use the right measuring device, the correct deduction of the consequences of observations given the accepted theory), but that one’s conclusion contradicts either another accepted theory, or another epistemic norm that one means to commit to. In this case there’s an inconsistency in one’s theory or set of norms, that is discovered thanks to another norm, here logic. In the case of a contradiction between a belief and other beliefs, which one should be preferred? The methodology one is committed to should also provide norms for this choice. For example, if assumed to include a consensual version of logical empiricism, it recommends pondering the ability of the empirical premises of the belief to resist refutation, and choose the one which has proven most robust to the most severe testing.
If one’s belief can be shown not be supported by the observance of the epistemic norms one is committed to, a mild sort of skepticism can be allowed, because the burden of the proof should lie on the believer’s ability to justify her claims. But this skepticism is mild in that it does not exclude believing in the claim later, should a legitimate proof be provided. If it can be shown that one’s belief has been acquired in ways that violate the epistemic norms one is committed to, although it is again difficult to exclude a subsequent legitimate proof, a more serious form of skepticism seems to be allowed. Finally, if this belief can be shown to be inconsistent with beliefs which have received more credit from their confrontation to the epistemic norms one is committed to, for example by resisting falsification, an even more severe sort of skepticism demands that we abandon the weakest belief.
Now, what are we saying when we say that knowledge is socially constructed? We say that we consider the utterance of a certain theoretical sentence (for instance a scientific publication) as a social phenomenon, or that we witness the social manifestation of the commitment to a certain epistemic norm (for instance by witnessing a laboratory experiment), and that we can explain this utterance or this commitment by the action of some kind of social determinant, or social factor. Of course, since the practice of science happens in a social context, it is not surprising that we find such explanations : students and researchers committing to particular epistemic norms in order for their work to be well regarded an deemed more trustworthy, and for their career to advance, research managers selecting research projects that have better funding prospects over projects that lack these prospects, researcher developing concurring theories for the purpose of academic rivalry and competitive advancement, etc. These factors may differ in epistemological flavor, but they are all, on a certain level, social phenomena.
According to our definitions, the claim under examination can thus be reinterpreted either as :
If the commitment to a theory (resp., to an epistemic norm) can be showed to be influenced or determined by social or cultural factors, then skepticism towards the theory or the norm is justified.
Or, accounting for the few additional questions raised by the second interpretation of social constructionism :
If the commitment to a theory (resp., to an epistemic norm) can be showed not to be the one to which all others necessarily converge, then skepticism towards the theory or the norm is justified.
III. Challenging the claim
Remind that one who commits to a given notion of truth, i.e. to its constitutive epistemic norms, will only resort to these norms when evaluating the truth or falsity of a sentence, or the righteousness of another epistemic norm. Then I must ask, should the fact that one’s commitment to a theory can be shown to be influenced or determined by social factors condemn one to skepticism ?
Well, it depends on what exactly the examination of social factors reveals.
Say that it shows that our pursuing this theory has been motivated by academic competition, or the theory’s appeal to funders, or our determination to prove someone else’s theory wrong, or even our belief that this theory justifies a social order we want to preserve, or power relations we want to foster. Say furthermore, that the same analysis does not show that the influence of these factors lead us to neglect the observance of any of the methodological norms one committed to. Say that further analysis of the social context shows that the norms were indeed observed. Then there is no reason to be skeptical, although our belief can be said to be socially constructed.
One may even argue that as long as the scientific institution ensures the enforcement of the right epistemic norms, all social factors that motivate students and researchers to compete and seek social recognition through scientific work and publication, all those social factors actually contribute to establish and demonstrate the virtues of these right epistemic norms.
It may be that the analysis of social factors suggests that an epistemic norm or theory was first adopted mainly for “political or social reasons”, while the scientific community neglected the observance of epistemic norms, and still, in the long run, the reliability of the theory be ascertained. This, according to Bruno Latour was the case with the acceptance of Pasteur’s microbiology. Then mild skepticism may be have been advised at first, but provided that in the long run, the theory is appropriately tested, this skepticism should be relaxed. Moreover, the historical analysis, made when the theory has proved reliable, of the influence of social factors at the time of the theory’s elaboration should not justify any form of retroactive skepticism.
Even clearer is the case of the demonstration of genetic inheritance of traits by Gregor Mendel, that was plagued by forgery of results which would have justified skepticism at the time of the discovery. The forgery may have been necessary to comply to the cultural norms of publication at the time, statistical knowledge supporting the result’s significance being unavailable. Still, the theory proved to be reliable in the long run, and learning today that compliance to a cultural norm caused a forgery at the time does not justify retroactive skepticism about the theory.
What I meant to show in the last paragraph is that the only cases in which the analysis of social factors influencing the course of scientific research should yield serious skepticism is the following. Skepticism is justified when social factors influenced the course of research in a way that prevented the observance of epistemological norm (mild skepticism), or lead to violation of the norms that allowed the possibility of errors (serious skepticism), or lead to conceal errors that were not subsequently addressed (even more serious skepticism).
In these cases, it appears that the analysis of social factors influencing the conduct of scientific research is actually a tool of enforcement of the epistemic norm regarded as the correct one, a tool that should be acknowledged by anyone committed to the norm.
At this point, I think it is clear that the first version of the claim under examination is refuted. In other words, it should be clear that it’s not the case that if the commitment to a theory (resp., to an epistemological norm) can be showed to be determined or influenced by social or cultural factors, then skepticism towards the theory or the norm is justified.
Finally, let us address the additional questions raised by the second interpretation of the notion of social constructionism.
Does the fact that multiple theories of a given phenomenon may have existed, and that one was chosen out of the possible theories because of social factors, justify skepticism towards the chosen theory? Since we have already established that the influence of social factors in the choice of theories or the commitment to norms is not a sufficient ground for skepticism, the question can be refocused. It is that of whether, to be worthy of commitment, a theory or norm should be able to manifest itself as the only possible one.
We need to consider what meaning we may usefully give to “possible” in this context. The mere ability to explain why one theory took over a concurrent theory does not inform us on their necessity and contingency in any interesting way, if understanding the mechanism by which anything is produced amounts to claiming its contingency. Then, because every material phenomenon, including the historical, psychological and social development of knowledge, is governed by mechanisms which, in right, can be understood and acted upon, all commitment would be contingent. Then the developments of true and false theories alike would be trivially contingent.
A less trivial understanding of necessity and contingency in the development of knowledge would be that commitment to an epistemological norm or theory is necessary, if the pursuit of knowledge leads necessarily to the commitment to this norm or this theory. Here, what seems to go unspoken is that, if knowledge is to claim any compelling power, it should be able to develop itself in only one manner, by virtue of some intrinsic norm or constraint. This is not to say that social constructionism itself works under the assumption of an ultimate norm of knowledge. Rather, it is to say that it makes a point of exhibiting the contingencies, i.e. the logical alternatives presented in the historical development of sciences, and the particular mechanisms by which these alternatives were resolved, as proofs that whatever norm is being followed cannot pretend to be the ultimate right one.
The reader can probably feel the intricacy of this position. To ease this feeling, we summarize the position as follows: if knowledge is not socially constructed, then in the long run there is only “one right model of anything”.
While it seems difficult to formally reject such a possibility, I would claim that demonstrating that a theory or norm is “the only possible right one” should not be a condition for commitment. Therefore, I will amend my initial stance of neutrality regarding the notions of truth under consideration in the examination of the claim. I will propose that an epistemological norm that commends skepticism whenever such a hard condition cannot be shown to be met, is too hard. I will now try to justify this proposal.
The pursuit of a form of knowledge that can only be acquired through prolonged effort and the construction of testable hypothesis, through long chains of conditional reasoning, seems to require commitment to a theory or to an epistemological norm in the absence of evidence that it is the absolute best. In this regard, what is commendable in terms of commitment to theories and commitment to norms both differs, and should differ greatly. One often abandons a theory in favor of another, while preserving, or even owing to, one’s commitment to overarching epistemological norms.
Instances of revisions of a commitment to a logical norm justified by commitments to a particular theories seem rare (confined to theories about the subatomic domain, which question the logical norm to be adopted), and I don’t feel entitled in discussing them. This case excepted, the scientific tradition shows many changes in theoretical commitments, but a remarkable consistency in the development of epistemological norms, and even more stability regarding logical norms and other mathematical habits. Regarding the necessity of the mathematical habits themselves, I will not speculate, as it is not required for the mere acknowledgement of their stability within the frame of the logical norm since the works of Euclid (and outside of it before him).
In any case, rather than suffering from commitment to norms in the absence of evidence that they are the only possible ones, the pursuit of knowledge seems to greatly benefit from tentative but prolonged commitments. Such commitments, lacking a proof that they are ultimately necessary ones which is probably impossible, seem to participate into the very normative mechanisms that leads to the accumulation and increased trustworthiness of knowledge. How this happens to be the case is a matter which I encourage my fellow philosophers to pursue, and which I will certainly pursue myself in future work.
In these few pages, I hope to have convincingly rebutted the claim first formulated as : If knowledge is culturally constructed, then it cannot be true.
I did not touch, however, on the reasons why a certain epistemological norm, even culturally constructed, could be preferred over another, and could be considered “the right one”, rather I supposed that one was already committed to a norm believed to be “the right one”. How such a belief is to be acquired is a matter for another time.