|WikiProject Philosophy||(Rated Start-class, Mid-importance)|
Problem of Counterfactual Conditionals is missing
Goodman's problem of counterfactual conditionals is not mentioned in the article. I think it's worth a blurb if not a second page. Jordan 23:33, 15 March 2012 (UTC)
- I'd like to expand on this problem, despite my not being well read on the subject.
- The "problem of counterfactuals" is that it is not necessarily clear what hypothetical assumptions must or should accompany a stated counterfactual condition, in order to make sense of a question about that condition.
- Consider "If you had been a plumber instead of a mathematician, would you still read The New York Times?" Here the question can't really be answered without understanding what psychological and environmental differences could have caused the mathematician to have instead become a plumber. And quite possibly some of those sets of conditions might result in a plumber who did read the NYT, and other conditions to one who didn't. In this case the counterfactual question would have no one single answer.
- But some counterfactual questions might at least seem to hypothesize a clear set of conditions. "If the roulette ball had landed on red, would you have won that casino bet?" would seem to have a clear-cut answer for someone who had just bet on black.
- A deeper analysis could show that things aren't necessarily that simple. What would it have taken for the ball to land on red instead of black? The croupier might have spun it differently. Or the wind currents may have been different. One or both of these conditions would require a whole backwards cascade of previous causes going back in time. At some point one of these causes could entail that the person betting would not exist -- say, if the different air current led backwards to a major gale 1000 years earlier that kept the bettor's direct ancestor from meeting their spouse.Daqu (talk) 05:10, 24 July 2012 (UTC)
The definition of counterfactual conditions is that it uses the subjunctive mood, asserting what would be the case if such and such were also the case. Yet, the final section gives an example that does NOT use the subjunctive mood: "The counterfactual condition is the basis of the comparison or control group in medicine, natural and social sciences. The experimental or treatment group demonstrates 'if X is present, then Y is present' (i.e. if a patient takes antibiotics, then a bacterial infection will be cured)."
Someone needs to clarify this -- "If x, then y" is an indicative conditional, not a counterfactual conditional. 184.108.40.206 20:47, 17 July 2007 (UTC)
Why is there a Nazi reference here (In "Possible World Semantics")? I think it would make more sense to say something like: "so, for example, it is not a world where hat-eating is common"... this ties in more with what the discussion is about. Of course, it's usually good to assume that the world is not run by Nazis when considering a counterfactual, but in the hat-eating example, I think that there's a better alternative. If anyone disagrees, please voice your opinions; otherwise, I'll make the change in a few days (or someone else can handle it). --Nichenbach 06:58, 17 May 2006 (UTC)
Perhaps the introduction could be less technical. It should be possible to understand what "Counterfactual" means without requiring formal logic. More examples would help also. Some of the links at the bottom of the page have some good ones. —The preceding unsigned comment was added by 220.127.116.11 (talk • contribs) 00:00, 23 November 2005 (UTC)
I'll be back to edit this more later. Needs plenty of work. KSchutte 18:21, 19 December 2005 (UTC)
While it may well be experts in rhetoric and philosophy define conterfactual conditionals along the lines the article currently makes I think the presentation is extremely misleading. Counterfactual conditionals do not require, as all the examples suggest, the past tense, negated clauses, or complex pre and post conditions. A statement like "If you make your bed you will sleep better." is a perfectly good example. Labeling a statement as counterfactual, or running against the facts, is a critique. The speaker, by assigning this label to a statement, is raising the question of how strong the connection from the precondition to the postcondition really is. He is not denying the connection but he is suggesting that the connection is weak, how firmly he assigns the label allows him to adjust how weak he thinks the connection is. The label is typically used to provocatively. I think it's more often used to label entire arguements, rather than individual statements. The redirect here from counterfactual isn't helping the situation. Bhyde 13:22, 2 February 2007 (UTC)
Does anyone think it would be relevent to talk about Lewis' counterfactual account of causation. I could add this but I'm not sure if it would be verging off topic. Ralphmcd 19:04, 17 February 2007 (UTC)
Confusing Section on Pearl
This comment refers to this revision of the article.
The section on Judea Pearl is confusing because it first asserts that Pearl rejected the possible world semantics, but then notes that Pearl's interpretation is in fact compatible with possible world semantics. (I know Pearl often casts his work more revolutionary than it actually is, so the first might well be a paraphrase of a claim of Pearl's, while the second part is a statement about research since then.) Be that as it may, either the rejection or the claim of compatibility need to be clarified, restricted, or removed. --18.104.22.168 (talk) 16:33, 28 December 2008 (UTC)
The (sub)section "English grammar" appears to be original research. Please provide a citation to where this is published. Tizio 13:07, 23 March 2007 (UTC)
This page needs work
This page needs a lot of work. At the very least, a section on Goodman needs to be added, as he is the father of the rival school of counterfactual analysis, that which rejects possible worlds semantics. Secondly, general clean up would be good. Anyone oppose if I try out some sweeping (but not too sweeping) changes? Pjwerner (talk) 02:28, 24 October 2008 (UTC)pwerner
I see this as a bit muddled. How can any conditional - indicative or otherwise - ever be true, given that truth relies on facts (events that have materially taken place)? Doesn't a conditional remain a hypothesis until then?
"If Oswald did not shoot Kennedy," and "If Oswald had not shot Kennedy," differ only in their temporal contexts - tenses; so up to this point, neither is actually counterfactual, where counterfact would surely imply that Oswald (had) actually saved Kennedy's life. Nor would I class them as 'afactual', since both imply (if the truth were told), Oswald's involvement, (but ask us to leave him out of the equation while we investigate what that might entail). At trial, in a court of law, this would be hypothesis. As it is, knowing after the fact that Oswald did it, perhaps it can be best termed as 'non-factual'? This would imply you could replace it with any other non-fact, (such as 'the Moon is made of cheese' ), and claim that someone else killed Kennedy.
In the first example, the subordinate clause (which we are supposed to take as being true), relies for its truth on a semantic falsehood, thus involving it in the proof of a negative. Try this out with: "If red is not a colour, then another colour must be red".
In the second example, "then someone else would have" is given as not necessarily true. That implies that it can be true or false, but always on condition of the antecedent. That can give us the case of : 'Someone else would not have [shot Kennedy], when Oswald hadn't shot him.' Examinator (talk) 14:48, 18 November 2009 (UTC)
- For me too, the JFK-Oswald example is a bit misleading or not totally explaining. In my course on Belief revision, the example (which I think is much more clear) states: "If I were the president of the European Union, I would become rich quite quickly" versus "If I were the president of the European Union, Michael Jackson would still be alive." Both are logically spoken true, because their antecedent is false, but the first seems intuitive, the other sounds counterfactual. It would definitely clear out the difficulties with times and tenses, since both implications have the same antecedent. Is it a good idea to change this example? BertSeghers (talk) 15:15, 10 January 2010 (UTC)
- Quite muddled, indeed, I'm afraid. But it's not the example that's bad, it's your understanding of it! The Oswald/Kennedy example is the classical example in the literature, so I suggest you stick with it. Granted, it's not the absolute best example one could come up with, but it's clear enough to most people. I'll try to clarify your confusions one by one. (1) "How can a counterfactual ever be true? Aren't they all at best just hypotheses?" Consider the following counterfactual: If I had let go of my wine glass 2 seconds ago, it would have fell to the ground. Almost everyone is going to grant that that counterfactual is true. Since our talk of counterfactuals pervades most of human inquiry (including science, law, history, and so on), rejecting the truth of all counterfactuals (or the possibility of our knowledge thereof) amounts to a very radical skepticism about most knowledge. For example, we wouldn't be able to ever truly say 'If the dinosaurs hadn't gone extinct, mammals wouldn't have been able to flourish in the post-Jurassic period' or 'If I hadn't defended myself, he would have shot me!' It seems reasonable to wonder how we can know about the truth of counterfactuals, but radical skepticism about all knowledge of counterfactuals seems to many too radical to be tenable. (2) Perhaps it would help if we thought for a moment of conditionals in general as expressing relations between propositions. Consider the pair of propositions (assume there can be negative propositions for the moment): Oswald doesn't shoot Kennedy (call it A) and Someone other than Oswald shoots Kennedy (call it C). Specify a time frame for the propositions, if you like. It doesn't matter. We can think of the two conditionals 'If Oswald hadn't Kennedy, then someone else would have' and 'If Oswald didn't shoot Kennedy, then someone else did' as both involving a relation between these same two propositions R(A,C). Don't let English grammar confuse you; the conditionals are both talking about the same states of affairs. So if the truth values of the two conditionals differ, then the relations expressed by the two conditionals differ. And the intuitive truth values of the two conditionals do differ, assuming that our world is a world where someone or other shot Kennedy and that the Warren Commission was true about Oswald acting alone. In logic-speak, we say that the conditionals involve different logical operators. In the indicative case, we write →(A,C) or A→C, whereas in the counterfactual case we write >(A,C) or A>C. Different operators! Two side notes: (a) Don't confuse either of these operators with the material 'horseshoe' operator, which is equivalent to ~A or C. All philosophers agree that the counterfactual operator is not the material horseshoe; it's controversial whether the operator at work in regular indicatives is. Never just assume that if the antecedent of a conditional is false, it's automatically true! (b) The proposition expressed by 'If Oswald hadn't shot Kennedy...' and 'If Oswald didn't shoot Kennedy...' is counterfactual because it is counter-to-fact. I'm not sure how you argue for the conclusion (reductio?) that we can replace 'If Oswald hadn't shot Kennedy...' with 'If the moon were made of cheese...' and still get a truth... (3) Subordinate clauses in conditionals are called antecedents. I don't understand what you mean by 'semantically false' or what your red example is supposed to show. (4) It's true that the counterfactual is contingent; for example, consider a possible world where Oswald didn't act alone. I don't know what your last example is supposed to show. (5) "If I were the president of the European Union, I would become rich quite quickly." Intuitively true, as you say. "If I were the president of the European Union, Michael Jackson would still be alive." False! You claim that it's true because it's antecedent is false. That's confused. Counterfactual conditionals are not the same as the material conditionals (see above)! (Many philosophers also think that regular indicative conditionals are not truth-functional either!) You can't just assume that A→C is true if A is false. Never can you assume that A>C is true if A is false. Also, Bert Segher's proposed example fails for another reason: it doesn't involve the same two propositions, as do the Oswald/Kennedy examples.--Jkhall (talk) 05:55, 29 November 2010 (UTC)
Possible world semantics
I undid the changes made to the possible world semantics section by Mike Rosoft (I was also the original author of the changes he undid). I incorporated what stylistic changes Mike included to my edits that I thought were helpful, and removed the ones that were false or that suggested that a controversy had been settled where an open debate remains. For example, in one of Mike's edits of my changes, he suggested that David Lewis's metric of closeness was supposed to be something like run-of-the-mill overall intuitive similarity. Anyone who has studied the literature on counterfactuals knows this is not true (see Kit Fine's 1975 paper and Lewis's 1979 response). In another place Mike's edit suggest that the Limit Assumption is false. This is presented as an established conclusion, when in fact many philosophers continue to believe it is true.
I apologize for making the changes I did without having logged in, but this page is terrible. It's clear from the content of the article and the comments on the discussion board (especially the terrible confused comments in the section entitled 'misgivings') that many of you know nothing about the logic of or literature on counterfactuals. —Preceding unsigned comment added by Jkhall (talk • contribs) 04:55, 29 November 2010 (UTC)
Confused about the protasis
I am confused about two points.
Firstly, the article starts by talking about an antecedent, then talks about the protasis, then goes back to talking about antecedents again. I have no idea what a protasis is, is it the same thing as an antecedent or not? Shifting from the terminology of philosophy to the terminology of linguistics then back to philosophy is not, I think, a good idea. If it needs to be done it should be explained why. Can somebody please explain?
Secondly, I do not understand what “The protasis of the first sentence is not necessarily true” means.
Does it mean: 1) It is not necessarily true that Oswald shot Kennedy.
or does it mean: 2) It is not necessarily true that Oswald did not shoot Kennedy.
It would not take too much effort to spell out exactly what the protasis is in this case, and this would be very helpful to people who are unfamiliar with the term “protasis”. Can somebody help with this?--Logicalgregory (talk) 16:01, 11 January 2011 (UTC)
In response to Logicalgregory, protasis is the linguists equivalent to the logician's antecedent. The page should decide whether it's addressing logicians or linguists. Because it speaks more in terms of logic than language, I suggest eliminating reference to "protasis" and instead keep just with antecedent. Objections? Jordan 23:37, 15 March 2012 (UTC) — Preceding unsigned comment added by Jordanotto (talk • contribs)
I added a dated "No footnote tag". The article has a reference section but all are general references. Since 2004 this article has had 149 edits by 81 editors so would someone please look at this? Otr500 (talk) 00:01, 23 December 2011 (UTC)
The Subtle Advancing of a Spurious Agenda
This article frequently uses the Kennedy assassination to demonstrate alternate uses of verbiage/tense for counterfactual examples. I think this is obviously, OBVIOUSLY an attempt to push the ludicrous idea that Oswald shot Kennedy, when he was clearly assassinated by either Fidel Castro or a very convincing body-double.