Talk:Problem of future contingents
|WikiProject Philosophy||(Rated Start-class, Mid-importance)|
What a load of ...
"Something will not happen tomorow. Therefore this (first) statement was true yesterday."
Really? It does not implies it. The battle could have been today and first statement still stays true. BUT! First statement says nothing about past. This whole implication is wrong.
- You'll find a lot of odd reasoning in the ancient philosophers; probably because they learned philosophy from a few masters only, while we have access to the thoughts of thousands upon thousands (thanks, Gutenberg!)
- However, what he's actually talking about is that truth statements in classical logic are invariant over time. If the proposition that the Earth is round is true, then it is also true tomorrow and it was also true yesterday; or stating it more accurately, classical logic does not concern itself with the concept of time. Also, it helps that the ancient Greeks had a tense that they could use specifically for timeless truths, whereas we have to do with the present, which implies a boundary in time.
- There are some logical calculi which can be used to model propositions that vary over time, should one need them. They are all twentieth-Century inventions, though.
- That being said, the problem as stated has a fault even in classical logic, which lies in the use of tomorrow. Even though the proposition is invariantly true, every day one enunciates the proposition, it refers to a slightly different proposition compared to the one enunciated the day before. Specifically, if I say there won't be a sea battle tomorrow, I am saying the same thing as there won't be a sea battle in the twenty-fourth; Tomorrow, when I say the same words, I'm actually saying the same thing as there won't be a sea battle in the twenty-fifth, and these two prepositions, one referring to the twenty-fourth and one to the twenty-fifth, may have differing truth-values. Wtrmute (talk) 21:50, 23 June 2009 (UTC)
Imagine the following universe:
- "Moment X"=TRUE; "There's light"=FALSE; "The fact is empirically true"=TRUE
- "Moment X"=FALSE; "There's light"=TRUE; "The fact is empirically true"=TRUE
- Any other combination between "Moment X" and "There's light"...; "The fact is empirically true"=FALSE
Considering this universe, what sense makes the expression ("There's light" AND NOT("There's light"))?
(Also posted at http://en.wikipedia.org/wiki/Talk:Contradiction#A_Question) --Faustnh (talk) 22:13, 20 March 2009 (UTC)
I propose moving the current article Problem of future contingents to Future contingent, since future contingent is widely used in the literature in such contexts as Medieval Theories of Future Contingents or talk of future contingent sentences/propositions/statements (e.g., in Putnam, 1967, 'Time and physical geometry'; or Mignucci, 1996, 'Ammonius on Future Contingent Propositions'). It also makes sense to me to introduce what a future contingent is before saying what the problem is. Objections? — Charles Stewart (talk) 08:27, 10 July 2009 (UTC)
Careful. 'Future contingents' is a direct importation into English of the Latin plural 'futura contingentia'. The Latin plural form nearly always refers to future contingent events or propositions, and is found in discussions of God's foreknowledge in tracts like Ockhams 'De futuris contingentibus'. The Latin singular form nearly always refers to a proposition, and is usually in the ablative form, as in 'de contingenti', meaning a contingent proposition. The English tends to reflect this - Googling the plural 'future contingents' returns the stuff about God's foreknowledge, Googling the singular 'future contingent' gives stuff about propositions. Note also that the English plural usage is normally a noun, whereas the plural usage is normally an adjective. (The SEP article throughout conforms to this convention).
Thus it would be odd to have the article titled 'Future contingent' as the English singular naturally suggests the adjectival use. 'Future contingents' would be OK. As you say, the article should introduce what future contingents are before saying what the problem is. Peter Damian (talk) 16:45, 13 July 2009 (UTC)
- The following two sentences seem natural to me:
- Note that the last proposition is a future contingent.
- We are interested principally in the domain of the future contingent.
- So I think the singular form can be used as a noun. There's a WP article naming policy that says to prefer the singular, because you can usually add the -s to get the plural form without an additional redirect. I lean to the singular, even if it might sound slightly artificial, since we can phrase the lede how we like. — Charles Stewart (talk) 07:07, 14 July 2009 (UTC)
Sorry, don't agree. My golden and never to be broken rule is that what is Wikipedia should never be far from anything in a standard reference work. Wikipedia should not be an outlier. Thus, if we look at the SEP, there are 28 occurences of 'future contingents' in the plural form, i.e. as a noun, including some important articles from the literature. There are 20 occurrences in the singular form, but always as an adjective qualifying a singular or plural noun. Specifically 9 occurrences of 'future contingent proposition', 8 occurrences of 'future contingent statement', 2 of 'future contingent event' and 1 of 'future contingent prediction'. If you can find important and significant uses of the singular noun in this context, let me know. Otherwise the article could be called 'future contingent proposition', but then you lose the ambiguity of the plural form, which sometimes means 'event', other times 'proposition' or 'statement'. I appreciate that Wikipedia has a policy on use of the singular, but that does not require us to have articles on 'scissor', 'trouser' or 'pant'.
- I'll look a bit more for natural examples. I agree that my first example is an ellipsis, though I managed to convince myself that it wasn't before posting.
- Peter Øhrstrøm, 2009, In Defence of the Thin Red Line: A Case for Ockhamism contains "There are future contingents. But no future contingent is true." (p20). — Charles Stewart (talk) 13:57, 14 July 2009 (UTC)
Aristotle vs Diodorus
|“||Sorabji has suggested that it is, for reasons of intellectual economy, attractive to rule out the possibility that the neither Aristotle's argument against the fatalist nor the Master Argument was a response to the other. [...] Now it has been made very likely by Sedley, on the basis of historical considerations, that Diodorus was a younger contemporary of Aristotle's than had previously been supposed, and hence if we respect Sorabji's constraint, we should conclude that the Master Argument was probably a response to Aristotle.
Purely logical considerations certainly make it unlikely that DI 9 was intended as a response to Diodorus.
[next page] But I have argued that it is unlikely that Diodorus himself intended his Argument to run on temporally definite expressions, or that that he noticed its implicit commitment to necessitarianism. So it is most likely that neither the Master Argument nor DI 9 was directly conceived in opposition to the the other, and hence that Sorabji's proposed constraint should be rejected.
[conclusion] Instead of identifying the possible with what is and will be true, Diodorus now merely identifies it with what is or will be true. However this may be, Diodorus' attempt to prove that all possibilities will in time be realised still presumes that no one will question the principle of bivalence. Aristotle meanwhile has dared to do just that.
More than one problem
Branching time models
Also, the following sentence seems ambiguous: “Suppose that a sea-battle will not be fought tomorrow. Then it was also true yesterday (and the week before, and last year) that it will not be fought, since any true statement about what will be the case was also true in the past.”
By "not fought tomorrow" does this mean on a specific day X? If so there is no paradox, as this merely relates to that specific day. I presume "not fought tomorrow" is instead taken to be a true statement on all days, past and future. If it is intended in this latter sense, then mathematical induction would indeed seem to imply that the battle is never fought, if the original assertion is true, classical bivalent logic is used (thus validating the law of excluded middle and the law of non-contradiction), and a classical linear (i.e. determinate and non-branching) timeline is assumed.
Looking at the article as a whole, if a branching model of time is assumed, then there seems (to me) to be no paradox. Along some branching timelines the battle might be fought, and along others it might not be fought. The proposition as to whether the battle is fought or not in the future can then be considered to be true and false depending on the chosen timeline. When looking at the overall time tree, as long as the battle is fought at some point, along some timeline branch section, the statement "the battle was fought" is then true.
Even then though, there is the question of how you define "the battle" (a given sequence or state space of quantum states?) opposed to a different version of the battle or a different battle altogether. The term "happened" also has issues regarding definition. Definitions are a problem even if a linear (i.e. non-branching) model of time is assumed.
- It's meant to refer to an unequivocal fact that obtains at some particular time. See philosophy of time, but I'd imagine there are bigger fish to fry than future contingents in regard to branching time models; an implicit presentism for one (er… many).—Machine Elf 1735 19:55, 9 November 2011 (UTC)