Subject: Re: JH: The Design Argument Date: Mon, 20 Aug 2001 14:29:39 -0700 From: "Brian Holtz" To: "Paul Filseth" wrote: > My criteria aren't "understand everything" and > "predict everything". They're "understand something" and "predict > something". The miraculous priest theory satisfies those criteria. > > So what exactly did you mean by "unifies the > > phenomenon with others"? > > It means a hypothesis for the cause of some mystery phenomenon > also implies some other unexplained phenomenon will occur. Are you saying that if a theory doesn't predict and explain some other phenomenon previously thought to be unrelated, then the theory cannot be considered a successful explanation of the original phenomenon? > > > > while no natural theory can predict them at all. > > > > > > That's not a prediction. > > > > Father Jones announces [..] > > The next day, miracle X happens. How is that not a prediction? > > I meant the _class_ of events isn't predicted That depends on who's defining the class. What principled method of defining event classes makes this not a prediction? > Why do you say the naturalist theory is silent? Because no theory exists that can predict the phenomena without making reference to the will of the priest. > Anybody in your scenario who insists after 10,000 years that > there must be an unknown but natural reason why all those crucifixes > appear will predict the next one just as confidently as a theist Only if he includes the will of the priest as part of his unknown but "natural" reason. Of course, such a "reason" is by my definition supernatural, but you dodge this conclusion by claiming that my definition (and all others you've seen) is unclear. Are you also claiming positively that a clear definition of supernatural is impossible (or, equivalently for our purposes, that supernaturality is by proper definition impossible or oxymoronic)? > > > According to your scenario, _first_ we observed that priests could > > > make this happen, and _then_ people said it was because goddidit. > > > > Newton first saw an apple fall, and then said gravity did it. Does > > that mean his theory of gravity makes no predictions? > > His theory predicts that if you plug [..] the height of an > apple tree branch into a certain formula, you'll get the time > it takes for an apple to fall. That wasn't known before he > discovered it. The priest theory predicts that if you plug in the announced intention of the priest, you'll get the amount by which conservation laws are about to be violated. Before this theory, it wasn't known whether or by how much conservation would be violated. > > So, to clarify, you agree that there is a conceivable case in which > > a deity exists and has observable consequences and is the best > > explanation for those consequences? > > Certainly not. [..] I'm taking no position on whether > such a case could occur, because I don't have a definition of "deity" > to work with that's capable of partitioning the set of possible beings > into deities and nondeities. Are you claiming that no such definition is possible? And do you worry at all that you'll seem to be disingenuously ducking the issue by stubbornly claiming that millenia of philosophical debate has been about a concept that you now claim is too ill-defined (and undefinable?) to justify taking a position? Let me operationalize this somewhat. Do you agree that there is a conceivable case in which something exists and has observable consequences and is the best explanation for those consequences and would be considered by most people to be a "deity"? And before you cite cases of advanced UFO super-science or whatever, can you cut to the chase and tell us what is the closest case you could imagine to the traditional human monotheistic notion of a deity? > > such a notion of "being" makes it impossible to rebut the > > proposition that an arbitrary thing (e.g. Santa Claus) exists. > > one can easily rebut > the proposition for many such [things]. One simply checks to see if > [its alleged] properties are mutually incompatible. Are you saying Santa Claus has mutually incompatible properties? If so, then substitute for him any non-self-contradictory but non-existent thing, and address my point. Viz., such a notion of "being" makes it impossible to rebut the proposition that an arbitrary non-self-contradictory thing (e.g. Santa Claus or whatever) exists. > > > > Can you tell me what *is* standard usage [for "exists"]? > > > > > > Sure. I'll point out the concept it stands for when somebody > > > uses it. > > > > I'm not asking for instances, > > Instances are what determines meaning But instances are not all there is to meaning. In my book I define meaning as "the context-sensitive connotation ultimately established by the relevant denotation and use". Connotation (i.e. associated properties and concepts) is ultimately more important (for philosophy, not lexicography) than denotation. > they're the usage definitions try to model. Can definitions never have any prescriptive utility? Can the exercise of precisely defining something never lead to new insight? I think not. > > since they are everywhere (like "*is*" above or "are" in this > > sentence). > > So you were using them in the conventional sense, and not in the > sense of your causality-definition? Those senses overlap enough that either works for the present purposes. The non-overlap only matters in corner cases such as parallel or fictional universes, and not at all when I straightforwardly ask you "what is standard usage?". > Why then are you trying to stop > everybody else from using them that way? They can use "is" and "exist" all they want in the conventional sense for conventional cases. But when they talk about unconventional cases like gods and say (as Rick did) god-did-it is never a better explanation than "I don't know" I will point out the implications of their statements. Which are (again): Either Rick is claiming gods are impossible, or he has no defense against people who say (unparsimoniously) that Santa Claus exists and causes Christmas presents and also arranges it so that his existence is never the best explanation for Christmas presents. So my conclusion that Rick is claiming gods are impossible is based not on my dialect, but on my presumption that he wouldn't want to embrace a silly ontology. > You seem to be under the impression that usage should follow > definitions rather than vice versa. Philosophical usage should follow precise definitions, and precise definitions should follow from and clarify common usage. > It's okay to use concepts we haven't figured out how to define. Why is it, then, that when I use the concepts of "supernatural" and "deity", you (when it suits you :-) act as though you don't know what I mean? > People make mental models of stuff, but the stuff doesn't have > to match any particular model. The stuff can do anything it wants, > so to speak. Since it can do anything it wants, as long as a model > doesn't make contradictory assertions, the stuff can match it. > ("Anything it wants" may match either of two contradictory assertions, > but any relation the stuff may have to _both_ won't satisfy what > people mean by "match".) > > Some models take the form of graphs with vertices mapping to > stuff and edges mapping to cause-effect relations among stuff. In > some such graphs the edges collectively form paths connecting every > vertex to every other; in other graphs they don't. The stuff might > match one (or more) of the former or the latter. (Or it might match > no graph model at all.) > > "There exists an X with property P" means you have a graph model > in mind, and you have an algorithm in mind that partitions vertices > (or sets of vertices) into "P" and "not-P" classes, and the algorithm > outputs "P" when given at least one input, So far this sounds like logical existence, not ontological existence. > and the stuff happens to match the model. It appears you have just defined existence as stuffness, which hardly solves the problem. :-) > > > > The only elements of non-circularity we find above are: verifiability, > > > > knowability, recognizability, understandability, presence in the > > > > universe, and concreteness/physicality. [..] > > > > > > If so, the lexicographers botched their job. > > > > So you disagree not only with my definition but also with those > > of all the lexicographers I could find? :-) > > For an atheist, you appeal to authority quite a lot. :-) To determine "standard usage", what else can one do but appeal to the habits of the linguistic community, especially as recorded by lexicographers? > This isn't about me and my "disagreement" with them. I _exhibited_ defects > in the definitions. It's not me you have to deal with, it's the _defects_. You are the one who called my definition "non-standard". When I show that the non-circular elements of all the published definitions I can find resonate quite well with my definition, you call those definitions "defective". I agree they are defective in a perscriptive sense (in ways that my definition corrects), but this has nothing to do with their correctness in a descriptive sense. What lexicographic or other evidence do you have to show that the "standard usage" of 'existent' (and related words be/real/actual) bottoms out as anything else than what I say it does (causal relatedness)? Note that this doesn't require that English speakers recognize that 'existence' bottoms out as 'causal relatedness'. It merely requires that every accurate description of what speakers mean by 'existence' bottoms out (perhaps indirectly) as 'causal relatedness'. > I observed that standard usage doesn't match their definition You did? You wrote: > the lexicographers botched their job. Most of the above > elements belong to epistemology, while "exist" and "being" belong to > ontology. You can't settle the immaterial-soul question by word games, > which defining "exist" as requiring concreteness/physicality tries to > do. "Presence in the universe" merely shoves the problem off onto the > definition of "universe". Where is there an observation here about "standard usage"? > > > You can't settle the immaterial-soul question by word games, > > > [such as] defining "exist" as requiring concreteness/physicality > > > > Right, which is why I propose to settle it by requiring causal > > relatedness. > > That hardly settles it. The immaterial-soul theory is quite > specific that soul-events cause physical events and physical events > cause soul-events. I'm not sure what you mean by "the immaterial-soul question" here. The issue I see is how to decide whether souls or other purported immaterial things "exist". Dictionary definitions of "exist" often cash out as concreteness/physicality/materiality. I agree that such a criterion is a worthless way to decide whether souls etc. "exist". As an alternative, I propose the criterion of whether they are causally related to our universe. I think my criterion *can* help settle the issue. Now, if anybody adamantly insists souls etc. "exist" even though they are causally unrelated to our universe, I'll make no sudden movements, back slowly away and and say in reassuring tones that we'll have to agree to disagree. ;-) > > I'm not saying they don't use it, I'm just saying (as you say about > > "supernatural") that they are confused when they use it. > > That they (or maybe we) are confused and not clear on the concept > doesn't mean everyone should pretend we have a completely > different concept in mind. I make no such pretense, especially since the lexicographic data I cite suggest that in fact standard usage can be summed up and clarified as I do. So I am just pointing out that people must either clarify their concept along the lines identified, or embrace what appears to be an untenable implication of not doing so. > > > or (b) jumping to a correct conclusion not supported by evidence. > > > > Then for any thing (e.g. Santa Claus) that in standard usage is > > positively considered not to exist, this option would make that > > thing's existence irrefutable. > > If by "Santa Claus" you mean "a" Santa-like being in a detached > part of the multiverse, it's existence _is_ irrefutable. If by "detached" you mean will never have had any possible causal relationship with us, then its "existence" is indeed in principle without any possible observable consequences. But if by "detached" you mean something weaker, then its existence would in fact have possible observable consequences and would in principle be refutable. > If you mean "the" actual Santa Claus of Earth-legend, [..] > causal interaction is part of what's understood by "the" Santa Claus. Not for my hypothetical Santa Claus who arranges that his existence is never the best explanation for Christmas presents (or any other phenomena). That's the very Santa whose "existence" I say Rick has no argument against. > > Only because you stipulate that those inhabitants "exist" > > Only for the purpose of exploring that scenario. The scenario > exists. The stuff might match it. If you reject it a priori you're > doing armchair physics. I'm not sure what your point is here. You originally wrote: > if reality contains two causally disconnected segments [..] > then the inhabitants of each segment would have to call the > other "non-existent", even though they're both part of the whole. My point is that it wouldn't matter if "reality" (whatever that might mean) contained things that we would mistakenly say it doesn't contain if those things can never have any possible causal relationship with us. So if you're looking for an uncomfortable implication of my definition, keep looking. :-) > > > > > "No non-interacting parallel universes exist" is an unfalsifiable > > > > > synthetic statement, and making it analytic by redefining "exist" > > > > > doesn't settle it > > > > > > > > A statement like that whose truth or falsity seemingly has no possible > > > > consequences is hard to consider as having any content whatsoever. > > Positivism is what you're doing if you choose not to > consider certain statements. No, it's stronger: positivism asserts that certain statements are propositionally meaningless. > It's Newspeak when you try to _push_ > that choice on others by _taking their words away from them_. If I knew how to "take words from" people, then I would have already taken enough from _you_ to wrap this up. :-) I hardly think I'm "taking words away from" anyone by pointing out that certain of their statements have untenable implications or are propositionally meaningless. > Applying your own definition of a word to what others say even though that's > not what they mean by it, on the grounds that you think what they _do_ > mean is confused or philosophically unsound, is putting words in their > mouths. If pointing out the implications of what others say is "putting words in their mouths", then I plead guilty. I'd rather spend time debating the substance of my allegations than arguing over whether making those allegations is rhetorically polite. Debates about the recent course of the debate is what makes most Usenet discussions so lame... > > > "No non-interacting parallel universes exist" is synthetic. > > > > I doubt there's a way to rephrase it that doesn't > > give parallel universes the same status as fictional universes. [..] > > Sherlock Holmes has the same status as a parallel universe. I should have written: "the universe of Sherlock Holmes has the same status as a parallel universe". > If Sherlock Holmes were real he'd have a[n] effect [..] > If a parallel universe is real it doesn't have any effect [..] > That's not the same status. If the fictional universe of Sherlock Holmes were by some amazing coincidence a "real" non-interacting "parallel universe", it would not have any effects. A fictional universe would have precisely the same effect-causing status as a non-interacting parallel universe, and should therefor be considered to have the same ontological status. > > Then by what criteria could Rick ever assign anything to the > > category of (not existing) instead of to the category of > > (existing but can never be the best explanation for the evidence)? > > a low Bayesian prior probability. I.e., implausibility. How does Rick determine a priori that it's implausible that a deity could exist and therefor that god-did-it is never a better explanation than "I don't know". > If a hypothesis > explains, predicts and unifies enough, it seems improbable that the > universe would coincidentally make something that doesn't exist seem > to do such a good job. How do you determine the probability or improbability of such a coincidence? You have to pick your ontology (i.e. decide how to decide what counts as real) before you can start drawing conclusions from the patterns and frequencies you notice amongst those things you do count as real. > > Operationally, he would have to use the paired categories > > inseparably. > > Hardly. There are all sorts of unfalsifiable hypotheses that > flunk the laugh test. Only if you smuggle in the hidden assumption of a parsimonious notion of existence, and thus beg the question. > But there are some that pass it -- parallel > universes being the prime example. How could one conclude that a particular parallel universe belongs to the category of (existing) instead of to the category (not existing but is the best explanation for the evidence)? -- brian@holtz.org http://humanknowledge.net