From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Philosophy (Rated Start-class)
WikiProject icon This article is within the scope of WikiProject Philosophy, a collaborative effort to improve the coverage of content related to philosophy on Wikipedia. If you would like to support the project, please visit the project page, where you can get more details on how you can help, and where you can join the general discussion about philosophy content on Wikipedia.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.
 ???  This article has not yet received a rating on the project's importance scale.

Underdetermination and indeterminacy[edit]

Do people usually learn in university that underdetermination and indeterminacy are the same thing? To me they are patently not (as I have tried to explain below). I think to equate them is to drain underdetermination of its important, specific meaning. It's a regrettable simplification. Is my education completely different to everyone elses? In that case I won't push the issue further here. Kronocide 23:34, 28 January 2007 (UTC)

I added this as an alternate name because confirmation holism linked here, presenting a concept called "indeterminacy of data to theory", which was completely identical to the underdetermination issue (modification of theories by ad-hoc hypotheses so that they may never be falsified). Plus I recalled coming across the two used interchangably numerous times. Seeing the wiki entry for "indeterminacy (philosophy)", though, it's clear that apart from indeterminacy of data to theory this concept has a whole other meaning, so I edited accordingly. --AceMyth 07:07, 29 January 2007 (UTC)

Requests for claification requested[edit]

This page was written rather quickly and I (the author) fear that it might rely on too much jargon. I'm really immersed in the philosophical discussion of this problem (in the philosophy of science) and it would help me, if you found places where this article digresses into jargon or just over-technicality, if you would point them out to me here. I'm usually pretty good at making things clear once I know what to clarify, and I'll do my best. philosofool

I don't think the problem is one of jargon, but of a somewhat wide view of the problem itself. There seems to be no difference, in your view, between underdetermination and indetermination. Underdetermination ought to signify those situations were evidence limits the range of acceptable ("true", "good", take your pick) theories, but not to a single theory. If there is no limitation at all (Cartesian dreams/demons, the Matrix) then it's not strictly speaking underdetermination, but indetermination.

Examples of true underdetermination arguments are David Hume's (which only limits theory acceptance with regards to past observations, but not to future predictions, hence the underdetermination) and especially Pierre Duhem's holistic argument against "crucial experiments." (Quine had a similar idea a half-century later.) In fact, pretty much all of the contemporary philosophical debate on underdetermination problems takes its cue from Duhem, so he ought to be mentioned. I might have time to add something later.

Other relevant figures are Nelson Goodman (the Grue argument) and Kripke's Wittgenstein's skeptical argument about rule-following. Even more peripheral is Harry Collin's "Experimenter's Regress." For people who have written extensively about underteremination problems, see for example Larry Laudan, Donald Gillies, and Imre Lakatos. Kronocide 19:35, 27 May 2006 (UTC)

I will second Kronocide in his assesment. Also, it could be considered jargony everywhere, but probably inextricably so. The solution, rather than trying to find the best synonyms in the thesaurus would be to give more examples, wherever possible. AnandaTree 02:44, 11 September 2006 (UTC)

'Indetermination' is not really a word, but would probably be a synonym for underdetermination if it were. Cartesian demon scenarios are really just one form of underdetermination: Two rival theories (real world? or demonic deception?) and evidence (sensory evidence, anyway) can't decide between them. Bing! Underdetermination.


This sentence is nonsensical, from the "then" onwards: "For example, if all that was known was that 10 dollars was spent on apples and oranges, and that apples cost 1 dollar and oranges 2, then we would know that 9 apples were not purchased, but we could not know which combination of apples and oranges were purchased." — Preceding unsigned comment added by (talk) 17:09, 30 June 2011 (UTC)

That a theory is underdetermined is a premise upon which skepticism often rests. Since skepticism is a view that challenges beliefs about knowing and people value knowledge, people often speak of "the underdetermination problem."

This makes sense. Yet, underdetermination is not mentioned at skepticism?

He rejects all such arguments but they have been very influential in epistemology.)

What does this mean?

Skepticism, and consequently underdetermination, has a history at least as old as the ancient Greeks. This issue has continued to be one in philosophy ever since.

Does this add anything? Not sure.

It actually adds an error. Scepticism and underdetermination are not the same thing. There can be scepticism without underdetermination, so it's not a consequence of that there were scepticism in ancient Greece that there were also underdetermination (arguments. Note that if theories are underdetermined by evidence, then this has always been the case, but the sentence above seems to want to be about what types of arguments existed, not if underdetermination existed.) There is, as I've mentioned above, generally a confusion between underdetermination arguments and general scepticism in this article. Descartes's demon, for example, has nothing to do with underdetermination. Kronocide 14:44, 17 November 2006 (UTC)
Quoting you from below, "Something is determined, but not one unique theory." Seeing as that "something" contains both theories where there is no such demon and theories where there is such a demon, and this is after having considered the entirety of the evidence (which is the only thing we can hope to be guided by), it seems to me that the demon hypothesis certainly highlights what kind of a problem underdetrmination implies for scientific enquiry. --AceMyth 07:38, 29 January 2007 (UTC)

[Descartes], who attempted to argue from a skeptical position to a position in which he had significant knowledge.

Does this mean anything? I think it needs to be spelled out to add value.

I've reworked the examples, and trimmed. Possibly too much.

A question I have is, are the example theories provided really inconsistent?

David Hume is another source of an important underdetermination argument, the problem of induction.

Is he 'the' source or 'a' source? I'll punt on 'the'. Please fix if I've guessed wrong.

A set of beliefs that has been shown to be false by new evidence can adapt via ad hoc hypotheses, replacing the hypotheses that have been contradicted with new ones that are consistent with the new evidence, while retaining the core content of the theory. In addition to claiming that for a particular theory there is at least one rival theory consistent with all evidence for the theory, it is often claimed that there are infinitely many such theories.

I love this paragraph, but I don't think it belongs where it was. Any suggestions?

This is the holistic underdetermination argument, the most widely discussed today, so it should be described somewhere. Kronocide 14:44, 17 November 2006 (UTC)

Lastly, am I right that faith could be defined as accepting one of two+ contradictory theories in the presence of underdetermination? And that skepticism is the refusal to accept an accepted theory on the grounds of underdetermination?

I think the important question is if it is defined that way, and it isn't, so that should not go into the article. Scepticism could be described as the refusal to accept a generally accepted theory, for some reason. Not necessarily for reasons of underdetermination. (Descartes's scepticism, for example, was not based on underdetermination.) Kronocide 14:44, 17 November 2006 (UTC)

I think there's still some work needed, but at the least, it's a bit more readable?

Regards, Ben Aveling 06:30, 29 October 2006 (UTC)

In my opinion this page needs more or less a complete rewrite. There is already a page about scepticism, so this page should be about underdetermination, properly so called. Underdetermination means that there is a determination deficit, not that nothing at all is determined by evidence. That would be indetermination (not a standard philosophical term, I should add) and general scepticism. To wit: inductive arguments exclude all theories that disagree with already observed events, but does not choose between the ones that don't, hence there is underdetermination with regard to what a theory says about the future; in holistic underdetermination, the set of beliefs consisting of theory and background assumptions is actually deductively falsified, but different theories compatible with the evidence can be constructed by properly changing one of the beliefs in that set. This is underdetermination. Something is determined, but not one unique theory. Kronocide 14:44, 17 November 2006 (UTC)
That's true, but I don't see how it's so incredibly incompatible with how the article is now to warrant a complete re-write. The only difference between what you wrote here and the point of view the article takes is whether we admit that for most useful purposes, a theory that has been subject to slight ad-hoc modification to move it from the "excluded" camp to the "not excluded" camp is essentially still the same theory. If you want to embetter the article by making a sharp distinction between the concept of underdetermination proper and its implications regarding skepticism etc., go ahead - but I don't think it'd be a good idea to just sweep said implications under the carpet just because technically there's merely underdetermination at work and not indetermination. The problem isn't that the evidence isn't ruling any theories out, it's that no matter how many theories it rules out, there are still plenty more left. --AceMyth 07:24, 29 January 2007 (UTC)
Actually, the article is a lot better now than I remembered it when I wrote that comment. Nevertheless, in the description of underdetermination I still feel the following changes to be important: Descartes should go unless he is actually discussed in university courses where underdetermination is analyzed in some detail. That is not the case where I am, and I think it is a mistake where it happens, but the important thing is that the entry reflects actual practices. Hume's problem of induction really needs to be elaborated in its underdetermination form. It is the first of the two major underdetermination arguments. Duhem and/or Quine really need to be included, since their form underdetermination is the most important one in current philosophy. I might do something about the Hume and Duhem-Quine if I can find a moment. Kronocide 16:51, 29 January 2007 (UTC)


Perhaps the article should say something about Wittgenstein's Rule following argument, which is closely related. Ben Finn 13:13, 1 May 2007 (UTC)

Merge proposal from Epistemological problem of the indeterminacy of data to theory[edit]

The article Epistemological problem of the indeterminacy of data to theory badly needs to be merged somewhere. I am open to other suggestions about where it should be merged though. Anarchia (talk) 21:04, 11 December 2007 (UTC)

Totally agree. Merge it. —Preceding unsigned comment added by (talk) 19:50, 10 June 2009 (UTC)
It's gone! Merged? — John Harvey, Wizened Web Wizard Wannabe, Talk to me! 12:14, 3 March 2012 (UTC)


Hi there, not to annoy anyone, but the lead section is rather weak. First of all, it does not do any justice to the concept of underdetermination, it is merely an account of indeterminacy. That is at best a specific case of underdetermination. I believe that the main idea of underdetermination is that, even if there seems to be a straightforward relation between an effect and its supposed cause, there are always other conceivable causes of that effect. Hopefully, someone can add something along those lines to the lead.

Furthermore, the example given is just false: When you have 10$ and oranges cost 2$, you can perfectly spend 10$ on apples and oranges by buying 5 oranges. Nothing in the previous sentence implies that you actually have to buy both oranges and apples. What you cannot do is buy more than 5 oranges... — Preceding unsigned comment added by (talk) 10:21, 18 April 2014 (UTC)