Wednesday, July 15, 2009
Good Science, Bad Philosophy
Here we go again. Scientists are justly upset with people who have no experience or knowledge in their field nonetheless presuming to lecture them on science. For some reason, scientists who have obviously never put much effort into studying philosophy think themselves competent to babble on about it.
The latest to fall victim, in a rather spectacular way, to this syndrome is the physicist Sean Carroll, who is admittedly uninterested in philosophy, but who does not let that stop him. He gets off on the wrong foot (the one with the bullet in it) by claiming that, on the issue of what sort of questions science is competent to answer:
... one popular but very bad strategy for answering this question [is]: first, attempt to distill the essence of "science" down to some punchy motto, and then ask what questions fall under the purview of that motto. At various points throughout history, popular mottos of choice might have been "the Baconian scientific method" or "logical positivism" or "Popperian falsificationism" or "methodological naturalism." But this tactic always leads to trouble. Science is a messy human endeavor, notoriously hard to boil down to cut-and-dried procedures.I'm quite sure that the generations of philosophers (and scientists interested in philosophy) who have formulated, studied, critiqued and attempted to improve on those systems of thought over hundreds of years, would have been, and are, quite surprised to learn that they've only been working on "punchy mottoes."
I'm afraid that it doesn't take any great skill in prediction to surmise that anyone who thinks those philosophical systems are just popular mottoes, unsuitable to cover the "hard to boil down to cut-and-dried procedures" of science, is getting ready to boil science down to some cut-and-dried procedure embodied in something very like a motto. Carroll does not disappoint the prognosticator, even if he disappoints those who hope that scientists might be a bit more cogent and self-aware thinkers.
Carroll sweeps all those mottoes aside by first presenting an example:
Here is my favorite example question. Alpha Centauri A is a G-type star a little over four light years away. Now pick some very particular moment one billion years ago, and zoom in to the precise center of the star. Protons and electrons are colliding with each other all the time. Consider the collision of two electrons nearest to that exact time and that precise point in space. Now let's ask: was momentum conserved in that collision? ...Let's just stop here and note that Carroll is appealing to induction as a process that delivers "truth." But how does he know that? Over 200 hundred years ago, one of those motto producers, David Hume, pointed out that anyone who looks to empiricism cannot justify the truth-delivering qualities of induction because the only possible empiric evidence on the subject comes from experience -- in other words, from induction. We "know" induction produces truth because, in our experience, induction produces truth. That itself is an induction and trying to justify induction by an induction is circular reasoning, a logical fallacy. No one has solved this conundrum in the time since Hume but that doesn't bother Carroll, probably because he is unaware of it and ignorance is bliss.
... The scientific answer to this question is: of course, the momentum was conserved. Conservation of momentum is a principle of science that has been tested to very high accuracy by all sorts of experiments, we have every reason to believe it held true in that particular collision, and absolutely no reason to doubt it; therefore, it's perfectly reasonable to say that momentum was conserved.
[S]cience does not proceed phenomenon by phenomenon. Science constructs theories, and then compares them to empirically-collected data, and decides which theories provide better fits to the data. The definition of "better" is notoriously slippery in this case, but one thing is clear: if two theories make the same kinds of predictions for observable phenomena, but one is much simpler, we're always going to prefer the simpler one.This is, of course, Occam's Razor, which is far closer to an empty motto than the Baconian method, logical positivism, falsificationism or methodological naturalism. In fact the razor is a rule of thumb for making a first approximation (less formally known as "a guess"). Carroll cites to the replacement of determinism with quantum uncertainty as an example of science appealing to "the inference to the best explanation." I'd dearly like to know by what metric he determined quantum mechanics to be "simpler" than determinism. In any event, what guarantee of truth does the razor deliver even if you can manage to wield it rationally? As Samir Okasha's Philosophy of Science: A Very Short Introduction states:
The idea that simplicity or parsimony is the mark of a good explanation is quite appealing, and certainly helps flesh out the idea of [the inference to the best explanation]. But if scientists use simplicity as a guide to inference, this raises a problem. For how do we know that the universe is simple rather than complex? Preferring a theory that explains the data in terms of the fewest number of causes does seem sensible. But is there any objective reason for thinking that such a theory is more likely to be true than a less simple theory? Philosophers of science do not agree on the answer to this difficult question.Simplicity is surely attractive to those who want to think simplistically but does it have any truth-delivering capability? After all, it is the inference to best explanation and Occam's razor that the Intelligent Design Creationists appeal to. As Carroll notes, the IDers, like Carroll, don't see methodological naturalism as standing in their way:
There's no obstacle in principle to imagining that the normal progress of science could one day conclude that the invocation of a supernatural component was the best way of understanding the universe. Indeed, this scenario is basically the hope of most proponents of Intelligent Design.Carroll may not understand, or may not care, but, if his account of science is correct, then ID cannot be barred from American public classrooms. It is, under his version of science, a valid attempt at science and, even if he thinks it is wrong or unsupported, it cannot be barred from public classrooms just because it has religious implications. Nor is there any basis under our law to bar it just because it is "bad" science. Of course, simply because it will have bad consequences doesn't mean that Carroll's definition of science is wrong but, if your version of science includes something so clearly not science, it may be time to reexamine your definition.
But Carroll isn't done with the razor. Based on it he declares:
In the real world, by far the most compelling theoretical framework consistent with the data is one in which everything that happens is perfectly accounted for by natural phenomena. No virgin human births, no coming back after being dead for three days, no afterlife in Heaven, no supernatural tinkering with the course of evolution. You can define "religion" however you like, but you can't deny the power of science to reach far-reaching conclusions about how reality works.In short, Carroll is maintaining that philosophical naturalism is a scientific result. I wonder when he will be publishing this in the scientific literature? I'd suggest he try publishing in a philosophical journal but I think philosophers would take him even less seriously.
Labels: Accommodationism Incompatiblism
Let's see - a claim which is in principle falsifiable, yet no one has managed to falsify it despite centuries of trying. It is consistent with every shred of empirical evidence we have, makes testable predictions regarding every facet of the physical world, and has unrivalled predictive power. Remind us which other criteria need to be met before it can qualify as a scientific result? (Not a publishable one, of course - originality is a requirement for that.)
Do tell us. What scientific test would demonstrate that an event had a supenatural cause (as opposed to a natural cause of which we are yet unaware)?
The question is whether science has, in fact, explained "every facet" of the physical world (in which case, why are we still doing science?). If you are inducing from the fact that science explains much of the world to the claim that it can explain all of the world, you are arguing circularly.
The criteria of science that philosophical naturalism lacks is empiric testing. No one as far as I know has published such testing or the results thereof in the scientific literature and, therefore, originality is not a problem.
On the other hand there are so many different variants of philosophy that you have to suspect that some of them merit less respect than others...
Plus there is just a smidgeon of suspicion that some experts over-elaborate their expertise to the point of parody. This applies in many walks of life, not just philosophers and scientists. Have you ever listened to a wine buff drone on and on?
Plus I think you go a step too far when you infer that, based on Carroll's article, ID could be taught in US classrooms now. He said that ID could be taught in the future - if ID had gained scientific credibility through scientific methodology.
Carroll is essentially making an ontological claim, via induction, about some proposed entity which cannot be observed. That may be permissible if the proposed entity had a logical relation to some macroscopic property of Alpha Centauri A that is readily observable, but the behaviour of two individual electrons has no such relation. They are not only invisible, but also irrelevant. It's similar to arguing about how many angels can dance on the head of pin.
Carroll also doesn't seem to consider the possibility that directly observing the past (if possible), would actually alter or define the past, an idea explored by John Wheeler that has some mind-blowing implications. And it's actually not impossible when you're dealing with billions of lights years instead of four.
I think you go a step too far when you infer that, based on Carroll's article, ID could be taught in US classrooms now. He said that ID could be taught in the future - if ID had gained scientific credibility through scientific methodology.
Carroll's argument won't do it itself, of course. But if that definition was widely accepted, particularly by a court deciding whether the teaching of ID is constitutional, that would be the outcome. If ID's method is scientific, the mere fact that it is a minority (even a very small minority) view within the "scientific community" is not enough to make it unconstitutional to teach in public school classrooms. There is no requirement that what gets taught has scientific credibility, as long as it is "science." And, of course, the politicians, school boards and teachers who what to teach it as if it is science don't give a damn if it is credible.
'You can't prove a scientific hypothesis but you can disprove it.'
To coin a motto!
Note: searching on the internet is not allowed.
I wanted to leave a comment on this topic (I'm all for science) but my English is not as good as yours...
But, as you know, the scientific method is a set of processes including repeatable tests of falsifiability.
If the ID crowd could put forward a well formed hypothesis that could be tested, then yes they would be doing science. They have not managed that yet.
Not according to Carroll, who thinks that it can make an inference to best explanation based on nothing more than Occam's Razor. After all, the DI is correct that the things that they appeal to are empiric facts revealed by testing: that some structures/functions in biology are "irreducibly complex," that cellular processes are highly complex, etc. It is the inferences that they draw from those empiric facts which we dispute. But in neither Carroll's examples nor the DI's are the inferences being put to empiric testing. Carroll simply appeals to Occam's razor to say one is better than the other. The DI can and does the same thing. Without more, Carroll's account of science cannot say the DI's inferences are unscientific.
Mauritius. Which I knew before I looked it up to make sure I was right and to get the correct spelling. I am a font of useless and near useless information.
If you haven't been there, I suggest you plan a vacation. It is an amazing island (and, having traveled quite a bit, I am quite difficult to please).
I am flying out to Cuba tomorrow for two weeks.
Ockam's Razor is an indispensable explanatory tool. Consider the situation with ID again.
IDers might claim that ID is simpler than natural evolution, that Ockam's Razor weighs in their favor. The question is, are they right?
The answer is: of course not. Natural evolution does not postulate any entities beyond our explanatory framework, and it does not postulate anything superfluous. It does not postulate entities beyond necessity.
ID, on the other hand, postulates an "intelligent designer" which is outside of our explanatory framework, which is not (and, some IDers would say, cannot be) explained. ID does not explain how the "designer" has done anything. It does not explain anything.
Ockam's Razor is not the principle of least effort. If it were, then any predictive theory would lose against hand-waving. No, the razor does not favor the argument which requires the least amount of work. Rather, it says, the best explanation is the one that does not postulate unnecessary entities. (Necessity is recognizable by comparing two competing theories.) Clearly the razor favors natural evolution, and nothing "supernatural."
Okasha's argument is not convincing, because the razor does not require us to claim that nature is "not complex." Rather, it requires us to recognize the uselessness of postulating unnecessary entities. Our theories should be as complex as necessary to make good predictions, and no more. This does not require making any assumptions about how simple or complex nature really is.
As far as the problem of induction and the Alpha Centauri exmaple: I think Carroll did misrepresent the scientific view there. It is not that scientists claim outright, "of course momentum was conserved!" Rather, they predict that momentum is conserved, and maintain that attitude unless given strong enough reason to change it.
There is no problem of induction in science, however, because predictions are not induced by finite sets of examples. Rather, they are deduced from theoretical frameworks which define examples as such.
It may be "useless" for us to postulate unnecessary entities, but why must it be necessarily true? People make systems with unnecessary steps requiring involvement of "unnecessary entities" all the time. I'm typing on a computer run by an operating system notorious for that. Looking at it from the outside based only on what it does, I could surmise that, in any one function, it might not "need" to utilize, say, a memory cache (i.e. in theory, the function could be done better without involvement of that "entity"), but my computer might still do so.
It is still just a rule of thumb (perhaps a very good one) and not a rule of logic or some empirically demonstrated fact of the world.
It is not that scientists claim outright, "of course momentum was conserved!" Rather, they predict that momentum is conserved, and maintain that attitude unless given strong enough reason to change it.
Of course, the only source of any "reason to change" the theory or prediction is from induction (i.e. from observations that are not of, in Carroll's example, events a million years ago in Alpha Centauri, but from observations of a very tiny number of events, out of the total of such events in the universe, in our local area within the very recent past, out of the total age of the universe). Since all our theories are built on induction, any deductions from those theories and any tests of them, still rely on induction delivering "truth."
2. Why would you claim that "all of our theories are built on" it? I don't think any scientific theories are built upon it.
Theories are supposed to relate to the empiric facts of the world and tested against them, right? We don't just make shit up and call it a "theory" and then declare it right or wrong based on a whim. Furthermore, those theories are not restricted to the events that occur in one experiment or one observation.
The only empiric facts that count toward constructing theories or their testing are delivered by induction. Without induction, no theories -- theories are built on induction.
Which was my original point. What "truth" science delivers (and it is considerable) is not "Metaphysical Truth," the way Carroll wants to make out, with or without the aid of Occam's Razor.
There is no controversy between accommodationists and incompatibilists that "scientific truth" does not support religious beliefs -- at most, accommodationists maintain (or do not dispute) that, for certain sorts of religious beliefs, science does not rule out those beliefs. The controvery is whether it supports Carroll's metaphysical beliefs or, instead, merely does not rule them out.
Obviously theories are tested against observation. Predictions are deduced from theory, and thus theories are falsifiable. Where does induction factor in?
As for Ockam's Razor, imagine science without it. Imagine how much sense it would make if we allowed for the multiplication of entities beyond necessity. I just don't see the sense in thinking of the razor as optional, and I wouldn't underestimate its power or relevance to topics such as Intelligent Design.
As for your claims re Carroll's alleged "metaphysical truth," I am sure I don't know what you are talking about.
Well, I can't make you open your eyes. Why is any experimental or observational evidence relevant to anything other than the particular time, place and circumstances, if not for induction? If it isn't relevant to anything other than particular times, places and circumstances, what good is it to theories about the workings of the universe as a whole?
As for Ockam's Razor, imagine science without it. Imagine how much sense it would make if we allowed for the multiplication of entities beyond necessity.
It might not, but what guarantee do we have that the universe will, in whole or in any part, make sense to us? That the universe must make sense to us is, itself, a metaphysical claim not demonstrable by science. It is the difference between doing the best we can and having access to absolute truth. It is not unlike the problem of using scientific models which simplify complex forces and events to something we can understand ... without guaranteeing that the models are doing anything more than delivering approximations of greater or lesser accuracy.
re Carroll's alleged "metaphysical truth," I am sure I don't know what you are talking about.
First of all, tinkering in evolution, virgin human births and resurections are postulated by their proponents on the premise that philosophical naturalism is not true, not on the premise that our theories of how life works can be better explained if they have exceptions. Science, as Carroll conceives it, is testing the wrong hypothesis. When he comes up with a test of philosophical naturalism, then he may have a start.
There is no theory at all in science as to whether there is or is not any afterlife or what its manifestation in the physical world would look like if there was. There is no alternative scientific hypothesis to test (and if there was, it would still have to test philosophical naturalism to really test the claim).
The things he is claiming science can answer are, in fact, only subject to metaphysical assertions ... which is what Carroll is, therefore, making.
About Ockam's Razor, I said "what sense would it make" and you interpreted me as saying "what sense would the universe make." That was not what I meant. I meant, what sense would science make if we allowed for the multiplication of entities beyond necessity?
As for induction . . . You still have not supported your assertions with argument. And your response to a request for argumentative support has been to accuse me of not having my eyes open. I would suggest showing a little more respect to people who are trying to understand your point of view, and a little more openness to disagreement.
I've recently posted a bit on my blog about the philosophy of science, and I address the problem of induction. You might want to take a look, to see why I question the role of induction in scientific practice:
A Brief History of the Philosophy of Science.
To answer your question: "Why is any experimental or observational evidence relevant to anything other than the particular time, place and circumstances, if not for induction?"
I could equally ask, "how could induction demonstrate the relevance of any experimental or observational evidence to anything beyond that particular time, place, or circumstance?"
What is your answer?
The point, which I have already stated, is that meaning is not induced from observation to theory. Rather, it is deduced from theory to observation. Again, see my recent blog post for a broader discussion of this issue.
Pointing out how ubiquitous induction is within science is not an argument but simply denying that science uses induction (or something) is? The third time you do that I tend to lose patience. Your article says little about induction except noting Popper's failed attempt to recast science as deductive.
Carroll's example of Alpha Centuri was clearly an induction in classic form but beyond that, induction includes other forms of reasoning in science. As the Stanford Encyclopedia of Philosophy says:
Until about the middle of the previous century induction was treated as a quite specific method of inference: inference of a universal affirmative proposition (All swans are white) from its instances (a is a white swan, b is a white swan, etc.) The method had also a probabilistic form, in which the conclusion stated a probabilistic connection between the properties in question. It is no longer possible to think of induction in such a restricted way; much synthetic or contingent inference is now taken to be inductive; some authorities go so far as to count all contingent inference as inductive.
I could equally ask, "how could induction demonstrate the relevance of any experimental or observational evidence to anything beyond that particular time, place, or circumstance?"
Which was Hume's point exactly. You can't demonstrate it, i.e. demonstrate that induction delivers "truth." Which is not to say it is useless. Again as the SEP says:
The great advantage of induction is not that it can be justified or validated, as can deduction, but that it can, with care and some luck, correct itself, as other methods do not.
But like a sharp knife that cuts both ways, it needs to be handled with care, not in the careless fashion Carroll waves it about.
... meaning is not induced from observation to theory. Rather, it is deduced from theory to observation.
I observe the words but can deduce no meaning from them. Are you saying that science makes up any old shit and then goes out and finds observations that support them? Are you claiming Popper's falsification schema? What is that supposed to mean?
That was not what I meant. I meant, what sense would science make if we allowed for the multiplication of entities beyond necessity?
I have said repeatedly that Occam's Razor may be useful or necessary to do science. The question here is the ability of science to deliver "truth" as Carroll clearly claims it does. Saying that science's data:
... is one in which everything that happens is perfectly accounted for by natural phenomena. No virgin human births, no coming back after being dead for three days, no afterlife in Heaven, no supernatural tinkering with the course of evolution.
... is clearly a metaphysical claim that science can and does pass judgment on those issues. If we agree that he is wrong on that point, then we agree on my point.
"... predictions are not induced by finite sets of examples. Rather, they are deduced from theoretical frameworks which define examples as such."
The question I have is this: if a theory is rooted in observation at all (which I think most scientists would claim they are) and those observations are confined to limited space and time (which they have to be, since we can only observe a limited number of "events" out of the total that have and are occuring in the universe) and our theories are supposed to to describe the general workings of the universe, including making predictions of future events (which, again, I think scientists would agree), how can our theories not be inferences to universal affirmative propositions from the specific instances of observations ... or, in other words, inductions?
The blog post will of course look quite familiar, since it is a copy of my original response to your blog. The comment should look quite different.
As for your questions . . .
1. You say theories are "rooted in observation." That is somewhat misleading. Theories are tested by and evaluated according to observation. But the theory does not come from observation. Again, see the comment I linked to in my last post for more on this.
2. There are two points to make here.
First, scientific theories do not rest upon any inherently mysterious or inexplicable notions. Mann'sWord does not seem to distinguish between pseudoproblems (like freewill and absolute morality) with legitimate scientific issues, like cognition and DNA. Such things as time, space, morality, cognition, and DNA are all scientifically explicable. In fact, scientists have produced good ways of approaching them already, as you probably know. Scientists do not "scramble" with any of those things, nor is there any evidence that scientists are postulating more entities than are necessary.
Second, the "one theistic paradigm" does not explain anything at all. It dismisses all requests for an explanation by pretending that the word "God" explains everything. Theism is only an alibi for an explanation. It has no explanatory value.
Yes, I did. But it seemed to confuse methodological naturalism with philosophical naturalism (or else was declaring philosophical naturalism to be true by definition; by fiat) and I could not make out how it bears on these questions.
Theories are tested by and evaluated according to observation. But the theory does not come from observation.
Really? So no observations of the conservation of momentum were ever made before the theory was proposed? It was just made up of whole cloth? But in any event, if theories are tested by and evaluated according to observation, aren't our evaluations of which theories are true and useful and scientific themselves inductions? The simple fact is, no matter how you twist and turn, our science, indeed, all our "knowledge" is founded on induction.
Such things as time, space, morality, cognition, and DNA are all scientifically explicable. In fact, scientists have produced good ways of approaching them already, as you probably know.
There is a scientific consensus on explanations of those phenomena? No, I didn't know that. Have any references? More importantly, are promises of future answers the same as "scientifically explicable"?
... the "one theistic paradigm" does not explain anything at all. It dismisses all requests for an explanation by pretending that the word "God" explains everything. Theism is only an alibi for an explanation. It has no explanatory value.
[Looking] No, Occam's Razor doesn't say anything about "explanatory value" ... just about the number of entities (in your restricted version). And I asked for an objective standard, not your opinion as to what counts as "explanatory value".
Now, obviously, I'm not disputing that science is our best "knowledge" or that ID fails as science or any general inference to best explanation. But ignoring the fact that science is not the clear, neat and easy process that Carroll (and you, to a lesser, but still significant extent) make out is doing it no favors.
Please tell me where you think I confused matters.
More interestingly, you have ignored a great deal of my response.
I linked to a scientific study which supports the conclusion that Popper's hypothetico-deductive model, and not induction, is an accurate picture of how scientists think.
I cited Albert Einstein's own disavowal of induction as a significant factor in his construction of the theories of Special and General Relativity.
These points are not even deserving of a response, in your view?
You ask: "no observations of the conservation of momentum were ever made before the theory was proposed?"
Right. No observations of the conservation of momentum as such were made before there was some theoretical/experimental framework for recognizing the conservation of momentum as such.
You ask: "if theories are tested by and evaluated according to observation, aren't our evaluations of which theories are true and useful and scientific themselves inductions?
No. Why would you think they were?
The theories are not produced by or constructed from the observations. So why call it induction?
You say: "The simple fact is, no matter how you twist and turn, our science, indeed, all our 'knowledge' is founded on induction."
Does citing scientific research and scientists who have actually constructed dominant theories count as "twisting and turning?"
You say: "There is a scientific consensus on explanations of those phenomena?"
Did I say that?
And: "are promises of future answers the same as "scientifically explicable"?
No. I didn't say that, either.
You say: "No, Occam's Razor doesn't say anything about "explanatory value" ... just about the number of entities (in your restricted version)."
My restricted version? What do you mean?
About explanatory value, the razor is about constructing explanations for phenomena. Not building a ship or buying eggs. So obviously "necessity" is determined with respect to explanatory value.
You say: "science is not the clear, neat and easy process that Carroll (and you, to a lesser, but still significant extent) make out"
Please tell me what I've said that implies science is clear, neat, or easy.
I only have time now to comment on the abstract to the article you cited (you don't think I'm going to buy the article to prove your point, do you?).
Here is a quote from the abstract:
Therefore, support has been obtained for Popper's hypothesis that enumerative induction does not exist as a psychological process. Instead, people appear to process information in terms of increasingly abstract cycles of hypothetico-deductive reasoning. Consequently, science instruction should provide students with opportunities to generate and test increasingly complex and abstract hypotheses and theories in a hypothetico-deductive manner.
Needless to say, one article in the Journal of Research in Science Teaching can hardly be said to definitively answer this long-standing issue in the philosophy of science and, more importantly, it does not appear to address the question I have been posing. The issue here isn't how people psychologically process information, the issue is the nature of the information that are processing and the nature of "knowledge" that comes out of the process. The fact that people do not consciously utilize "enumerative induction" to reach a result does not mean that the inferences they ultimately reach are not, in fact, inductions, anymore than the fact that people do not mentally "process" what they see as individual photons causing chemical/electrical changes in their retinas means that that isn't exactly what is happening. Remenber the SEP article I cited to on the ubiquity of induction. As I said, if you rely on observations, either in the formulation of theories (and frankly, I think it is ludicrous to propose that we don't, no matter how "theory-laden" you may think our observations are) or in the evaluations of them, what you wind up having is an induction to a general proposition from limited observations. As to why our evaluations of theories are inductions, I explained it before, which you did not respond to:
... if a theory is rooted in observation at all [i.e. either in formulation or evaluation] (which I think most scientists would claim they are) and those observations are confined to limited space and time (which they have to be, since we can only observe a limited number of "events" out of the total that have and are occuring in the universe) and our theories are supposed to to describe the general workings of the universe, including making predictions of future events (which, again, I think scientists would agree), how can our theories not be inferences to universal affirmative propositions from the specific instances of observations ... or, in other words, inductions?
Hume's Problem of Induction does not address what the pysychological benefits or disadvantages of induction are, it addresses the "truth-value" of inductions, whatever they may be called.
I did respond to what you are referring to.
I'll respond to your other points after you've responded to mine.
one article in the Journal of Research in Science Teaching can hardly be said to definitively answer this long-standing issue in the philosophy of science
You're right, of course. But, then, I suppose you will also agree that your insistence that science relies on induction will not, in itself, answer the long-standing issue, either.
more importantly, it does not appear to address the question I have been posing.
I disagree about the relevance of that study. In any case, your comment about the meaning of information is interesting. I was wondering how much of our disagreement had to do with the way we understand knowledge in general.
I think we might need to get into a broader discussion of epistemology if we want to make any progress towards agreement. With that in mind, I've just posted something you might want to read on my blog:
Induction and Scientific Reasoning
Oh, and about this: "I was supposed to read YOUR comment to YOUR blog post at the link you gave."
I linked directly to the relevant comment. Sorry that wasn't sufficient to avoid confusion.
This is taking the discussion in a whole new direction and, frankly, I don't know when I can get to it because of personal circumstances. I have no intention of spending what free time I have all on a private discussion.
Already I see one question: if science is all about prediction -- good theories make good predictions -- are the theories judged by each (and only each) successful prediction or do we, in fact, at some point decide that they are so successful that we then demand more than one anomalous result to bring the theory into question?
Nor, I should say, am I going to accept your restriction on the meaning of induction by fiat. Hume may only have framed his problem in terms of the "classic" formulation of induction but that does not mean the problem is restricted only to it. The reason the term has been extended is not out of some whim but because of the similarity of the underlying inference.
About not having the time to respond in full, no worries. You certainly don't owe me any of your time.
But, to answer your question:
"if science is all about prediction -- good theories make good predictions -- are the theories judged by each (and only each) successful prediction or do we, in fact, at some point decide that they are so successful that we then demand more than one anomalous result to bring the theory into question?"
You are supposing that theories gain some special status upon observational corroberation; that they become "true." That is the point Popper contested, and which I also reject. Theories are always in question. Scientific advancement does not occur when a theory stops being questioned. Rather, discovery occurs when a new predictive framework is adopted which organizes our behavior in new ways.
"Nor, I should say, am I going to accept your restriction on the meaning of induction by fiat."
I explained why I am using the term as I do in my last blog post, which I linked to in my last post here. My usage is consistent with the traditional debate over the problem of induction. I am not adamantly opposed to adopting a different usage; however, as I noted, I am waiting for sufficient reason.
But that's exactly what Carroll did with conservation of momentum and why I said it was an induction. If you are claiming epistemological behaviorism, do you get to reject how scientists behave?
I am not adamantly opposed to adopting a different usage; however, as I noted, I am waiting for sufficient reason.
If it walks like a duck ...
I think the most we can say is, perhaps. After all, he did explicitly perform an inductive inference. I wouldn't assume he was thinking inductively. Yet, by failing to express the principle of conservation in purely predictive terms, he did leave room for confusion.
You say: If you are claiming epistemological behaviorism, do you get to reject how scientists behave?
I don't understand that question.
As for the "if it walks like a duck" comment, the problem is, I don't see a clear alternative definition of "induction" on the table.
If you do not take "induction" to mean "enumerative induction," then what do you take it to mean?
Should I take your comments about induction to include or exclude enumerative induction?
You have, as far as I can tell, a very minority view of the philosophy of science (which doesn't mean it isn't interesting). If you have any references to books or papers expounding on it, I'd be interested.
Thanks for an interesting discussion.
My point about Carroll was that he might simply be guilty of sloppy thinking (or writing), and not induction. But to understand why I would say that, you probably have to understand why I would think that comments like "of course momentum was conserved" could be read a number of ways, not all of which are open to Humean criticism.
In my view, the "problem of induction" is more of a confusion than a legitimate cause for concern, as it is based on an under-developed notion of knowledge and language. I think anyone of a pragmatic or Wittgensteinian bent would find more to agree with than criticize in my perspective.
This leads to the one small point of disagreement: I'm not so sure my view is "very minority." It may be somewhat minority, but then, I am not opposing any view which might not also be considered somewhat minority. In any case, I am not so concerned about remaining consistent with dominant trends in the philosophy of science as I am with making sense of science itself.
As for references which might elucidate where I'm coming from . . .
I took the term "epistemological behaviorism" from Rorty's Philosophy and the Mirror of Nature, which I highly recommend as an historical overview and critique of modern epistemology. Rorty attributes epistemological behaviorism to both Dewey and Wittgenstein.
My views are more or less consistent with the later Wittgenstein's approach to language as well as Dewey's conception of inquiry.
I cannot recommend Wittgenstein's Philosophical Investigations enough, if you haven't approached it yet.
Also, here is a relevant and interesting reflection on some similarities and differences between Wittgenstein and Hume: Mannison, "Hume and Wittgenstein: Criteria vs. Skepticism"