Most resources I've found on underdetermination approach the subject within the context of science. That's definitely a fascinating area of study, but I'd like to explore ways of applying underdetermination more broadly. I'm interested in the notion that people's beliefs about themselves and the world they live in may be massively underdetermined by the evidence available to them. If underdeterminism poses a threat to scientific theories, then it must pose an even greater threat to lay beliefs, as they seem to be developed far less rigorously.

In terms of beliefs about the self, I'm interested in the way our narratives of self and our identity are underdetermined and therefore liable to shift as different interpretations of the evidence are considered. I'm interested in how sudden realizations and epiphanies about one's life may be comparable to Kuhnian "paradigm shifts", catalyzing a radical realignment of self-image without a considerable change in the already-known facts of the matter.

In terms of beliefs about the world, I'm interested in the way underdetermination relates to ideology. It seems that people with differing ideologies can interpret the same event in different ways, which suggests that these ideologies are much like underdetermined scientific theories, actively competing to explain the empirical data.

Is this a topic that philosophers have addressed? Or have I accidentally strayed into the realm of psychology? Would appreciate any resources you fine folks could provide. Thanks!


Underdetermination by evidence does not pose a threat to scientific theories, it only poses a threat to realistic interpretation of scientific theories. In other words, empirical adequacy of scientific theories, on which their usefulness is based, is completely unaffected by the underdetermination. What is affected is ontological interpretation of them as representing "true" reality. Since different ontologies (e.g. interpretations of quantum mechanics) are compatible with the same empirical predictions it becomes difficult to argue that even one of them represents "the reality". But this also works the other way, whatever doubts underdetermination raises about correspondence to reality such doubts are empirically irrelevant. They may however be sociologically relevant, e.g. if funding choices favor adherents of a particular interpretation. Ideologies rarely submit themselves to the strictures of empirical adequacy, so applied there "underdetermination" is at best a metaphor.

Quine, one of the most prominent proponents of scientific underdetermination wrote a famous book that extends it to linguistics, Word and Object. His terms of choice are "inscrutability of reference", which means that parts of a sentence can change what they reference in such a way that the meaning of the sentence as a whole remains the same, and "indeterminacy of translation", which means that there exists no fact of the matter as to whether "radical" (from scratch) translation from one language into (never before encountered) another is correct. Indeed, Quine's ideas are the source of Feyerabend's and Kuhn's "incommensurability" of paradigms, see Is Feyerabend confusing discovery and justification when he criticizes the scientific method?

Together with the similarly minded Derrida's "deconstruction" on the continental side the idea was picked up and extended to cultures by post-modernists at large. If you are looking for such an expansive view Rorty's Philosophy and the Mirror of Nature should do nicely:

"...thinking of the entire culture, from physics to poetry, as a single, continuous, seamless activity in which the divisions are merely institutional and pedagogical... we shall say that all inquiry is interpretation, that all thought is recontextualization... To say that something is better 'understood' in one vocabulary than another is always an ellipsis for the claim that a description in the preferred vocabulary is more useful for a certain purpose."

A word of caution. The founding fathers of underdetermination, Quine and Kuhn, walked back their original polemically radical claims. And while social and cultural biases are there, exuberant post-modernistic quest for finding them everywhere ended up producing way too much inane nonsense, partly exposed by the Sokal hoax. Check out Zammito's Nice Derangement of Epistemes, which gives a philosophical history of post-modernism from underdetermination and incommensurability to "sociology of scientific knowledge" and ultra-feminism, and to its discreditation in 1990s. Here is his take on Rorty in particular:

"Setting out from Quine, Kuhn, and Davidson, Rorty has executed several elegant turns through Gadamer and Heidegger to come more and more to partner with Derrida... What is left is language and the arbitrary "poetics" of conversation. Rorty dissolves too many distinctions; his new "pragmatism" entails a cavalier disdain for rational adjudication of dispute. There has been a derangement of epistemes. Philosophy of science pursued "semantic ascent" into a philosophy of language so "holistic" as to deny determinate purchase on the world of which we speak. History and sociology of science has become so "reflexive" that it has plunged "all the way down" into the abime of an almost absolute skepticism".

  • 1
    Interesting point about empirical adequacy being safe... would this imply that it IS a problem for scientific realists but not for constructive empiricists? – Matt Diamond May 16 '16 at 1:00
  • 1
    @Matt There is a nuance, undetermination goes hand in hand with the theory-ladenness of observations thesis, so adequacy is not to be taken as individual matching of theory to "neutral facts", but as overall effectiveness of scientific schemes, methodology included, in their domains of application. This is the gist of Quine's naturalized epistemology, developed around the time of Word and Object, and its current popularity has much to do with it being far more defensible philosophically than traditional scientific realism. Zammito even uses it to criticize Quine's own excesses. – Conifold May 16 '16 at 19:11
  • 1
    Well done Conifold. Do not recall having the pleasure of reading this earlier. Is this pretty much what you believe today? – gonzo Aug 5 '20 at 3:17
  • 1
    @gonzo Thank you. Yes, I do think that after classics overdid it with absolutes the pendulum has swung the other way much too far. – Conifold Aug 5 '20 at 10:10
  • 1
    It seems that what we have ended up with is a radical/global skepticism and anti-realism as to the traditionally conceived “real world” (your “classics”), increasingly coupled with radical realism with respect to systemic abstract objects [with causal powers]: cf. structural realism, systemic realism, moral realism, etc. In this way, the “goal of inquiry” remains, “truth,” but truth of an entirely different sort. One which, because founded upon linguistic idealism, rather than, say, ontological realism, is highly unstable. – gonzo Aug 5 '20 at 17:33

Yes, one example is Noam Chomsky's Poverty of Stimulus thesis in psychological linguistics, which suggests that the data language learners have access to about grammar underdetermined those general rules. Chomsky uses this claim as a premise for defending the idea that we must be born with mental structures for handling language, since otherwise we would not be able to do it as well as we can.

  • The POS "thesis" was always wild speculation (Chomsky himself at one point explicitly rejected the idea that we need empirical evidence!) that has been convincingly refuted. The OP is interested in the relation between underdetermination and ideology. Chomsky is pure ideology so that's a good place to start! – user20153 May 15 '16 at 22:05
  • Yep -- agree with all of that. Perhaps hypothesis is better than thesis, but thesis is what's usually said. – ChristopherE May 15 '16 at 23:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.