Another ID knockdown, by Sarkar 8 Jun 200918 Sep 2017 Intelligent design (ID) is perhaps the most widely-discussed non-idea of all time. There seem to be three reasons why real scholars discuss it: 1. It is historically an idea that had influence on intellectual history, up to, say, 1860 2. It is an idea that needs to be discussed because the religious put it forward in opposition to science and 3. It allows one to discuss interesting ideas by acting as a kind of Gedankensexperiment. It is the latter that Sahotra Sarkar’s forthcoming paper in Synthese addresses. As Sarkar notes, may people have asked whether ID qualifies as science. The standard answer is that the designer is opaque to scientific investigation, in which case the hypothesis is not science (because it is uninvestigable), or it is not, in which case it is failed science. But Sarkar uses it to riff off the issue of the demarcation criterion which I have discussed here before. He takes the lack of precision in ID to not be, ipso facto, reason to deny ID is science, because, as is well known these days, many theories (he uses Elton and Macarthur’s hypothesis of the relationship between diversity and stability of ecosystems as an example) fail to have precise terms. I even think that a certain degree of imprecision is usually necessary for a hypothesis in its early days. The problem is that demarcation criteria assume a context independence which is historically and sociologically false. What counts as a theory in science is relativised to the state of the science at a time and place. The use of demarcation criteria in this context is usually political, says Sarkar. But for a theory to be subjected even to contextual criteria, there needs to be a theory there in the first place, and this requires that there be some terms that have shared meanings, and shared methods, that are employed in the theory under test. Terms like mass need to be relevant, which is to say, they need to be interpretable in empirical ways. ID fails in this sense en masse. He writes, “In the voluminous corpus of its [ID’s] proponents, there seems to be no attempt to de?ne “intelligence” in any way whatsoever. Except in analogy to intelligent human agents, we are not told what it means.” They do not give accounts in tractable cases like animal intelligence. They do not explicate what it means to be “specified” or “complex”. Sarkar makes a number of criticisms of the analogies used by ID: why is the fragility of “irreducible complexity” a sign of competence, rather than incompetence? We use redundancy to guard against failure, why wouldn’t the Designer? [A side issue, Wesley Elsberry and I introduced the notion of a rarefied design in contrast to the ordinary design to which it is analogous, and pointed out that the analogy fails – where design is like human ordinary design, the inferences used by ID fails, and where it is not, it tells you nothing.] He dissects Dembski’s actual precisification of ID: it is impossible to evaluate the probability of a pattern without knowing ahead of time how likely it is, and Dembski introduces a bit of special pleading in his criterion of complex specified information. Sarkar writes: Where are we left? We have no positive speci?cation of “intelligence” whatsoever. We only have, at best, an incoherent attempt at a positive speci?cation of “design.” In other words, we have no theory of ID at all. It follows, that we are in no position to judge whether the theory meets some demarcation criterion should we want to play that game. That is, we can’t even apply a demarcation criterion if we have one. There’s no “there” there. Late note: Brian Switek at Laelaps also has a discussion. Sarkar, S. (2009). The science question in intelligent design Synthese DOI: 10.1007/s11229-009-9540-x Creationism and Intelligent Design Philosophy Science
Creationism and Intelligent Design Tautology 2: The problem arises 23 Aug 2009 After Williams and others had made the comment that fitness is a tautology, it came around that the point needed to be discussed in more detail. One such discussion was by a student of Dobzhansky’s, Richard Lewontin. Read More
Biology Rise of the Planet of the Moralists 2: chains and trees 18 Oct 201122 Jun 2018 Rise of the Planet of the Moralists Series1: Introduction2: Chains and Trees 3: Clades and grades4: Predicting traits5: Social dominance and power Very famously, Darwin came up with the idea of the evolutionary tree. What is not often realized is that it is the tree that is more uniquely Darwin’s than natural selection, for which… Read More
Evolution Evolution quotes 1 May 2010 To understand evolution we must first understand the historical development of ideas on evolution. But to understand its history, we must first understand evolution. – Donald Forsdyke [H/T Piers Hale] Read More
/delurk This reminds me of George Reisch’s critique of pluralism, especially of the John Dupré sort. (This is from Phil Sci, like, 1998. I can track down the exact cite later of anyone wants.) Reisch argues that (a) we need something like a demarcation criterion to distinguish biology from ID, (b) this criterion can’t be a priori and decontextualized, à la falsifiability, (c) pluralism can’t deliver such a criterion, and (d) the unity of science, in Neurath’s sense, can constitute such a criterion. Very roughly, the proposal in (d) is that a proposal is judged scientific if it coheres with established and accepted science; since ID posits a supernatural designer, it fails this test, and hence doesn’t qualify as science.
One criterion which everyone should agree is necessary for science is that its inferences are _rational_. The reason I give for saying ID is not science (or is pseudoscience) is that the inference to ID is not rationally justified, and the arguments made by ID advocates are nonsense. If I thought a purported scientific inference was mistaken, but not by too large a margin, I would probably describe it as mistaken science, rather than pseudoscience. Of course, I haven’t addressed here the question of how we decide whether an inference is rational. That’s a much more difficult question. But those of us who agree that the inference to ID is not rational should have no compunction about saying that ID is not science. And that doesn’t preclude the possibility in principle that there could be a rational, scientific case for ID. Of course there are other criteria too. Not all rational inferences are labelled “science”. Some are labelled “history”, or “geography”, or are just everyday matters such as inferring why my car’s engine won’t start. Such categorisation is necessary for some purposes, like deciding what material to teach in different classes. But it’s not the most important thing. What matters most to me is whether an inference is rational, and the degree of rational confidence that I can have in it, not whether it gets taught in the right class.
That is, we can’t even apply a demarcation criterion if we have one. There’s no “there” there. I sense the phrase “pathetic level of detail” will lurk behind any response from IDers.
* Pluralism, Logical Empiricism, and the Problem of Pseudoscience * George A. Reisch * Philosophy of Science, Vol. 65, No. 2 (Jun., 1998), pp. 333-348 (article consists of 16 pages) * Published by: The University of Chicago Press on behalf of the Philosophy of Science Association * Stable URL: http://www.jstor.org/stable/188264
Noumena/BdN: Many thanks. An interesting paper, one of several lately that have pushed me towards Neurath, who I suspect I would have liked as a person as well as a thinker. However, neither Neurath nor Reisch are profferring a demarcation criterion (apart from “it is something that may be incorporated into science”, which is true but uninformative). What they are offering if I read that paper rightly is that we know when it is science, and on that account can incorporate it. Suppose I propose theory X. It is not empirically disconfirmed, but has almost no connection with the rest of the discipline. Is it possible science? Now suppose I propose the notion of momentum in physics in the 14thC… I still think that we identify things on an exemplary basis – not exactly Dupré’s “family resemblance” account, but something similar. In this case we can tolerably distinguish science from nonscience, as Burke would have said. It is not that there are a set of necessary or sufficient conditions to be scientific. It is not that there is some majority of conditions either. It is that overall the proposal satisfies the conditions that things we do know are science satisfy; it may be a majority, it may be all of those criteria the exemplar satisfies, it may be within spitting distance in some logical space. Whatever, we know when it is science because it closely resembles other science. Hence, any proposal (say, Madam Blavatsky’s theosophy) that fails to be like other science isn’t science. The problem only arises when things get ambiguous (like phlogiston theory or string theory). Creationism and ID aren’t even in the ballpark, demarcation criterion or no. It is not that ID posits the supernatural designer. It posits no designer or motive cause what so ever; it just uses a word. If the supernatural designer could be specified in properties so that it was predictable, then it would do fine as a hypothesis (for then it would have a nature and would become part of the natural order). It is that it has no specified theory whatsoever. As Wesley and I wrote in our paper, you may as well say that things are what they are because they are gazuply mufnordled. Every technical term of art used by ID is either arbitrary, contrary to fact, or simply vacuous. And that is why it ain’t science.
Snowflake: [quote]It is not that ID posits the supernatural designer. It posits no designer or motive cause what so ever; it just uses a word. If the supernatural designer could be specified in properties so that it was predictable, then it would do fine as a hypothesis (for then it would have a nature and would become part of the natural order). It is that it has no specified theory whatsoever. As Wesley and I wrote in our paper, you may as well say that things are what they are because they are gazuply mufnordled.[/quote] I disagree. I think the claim that something is “intelligently designed” does have some meaning. Intelligent design implies, at the very least, a purpose and a selection between alternatives in pursuit of that purpose. The claim that species were intelligently (purposefully) designed is a major departure from current evolutionary theory. Moreover, IDists also make the stronger claim that some structures observed in organisms [i]could not[/i] have evolved by the mechanisms described by evolutionary theory. I agree that these claims have no predictive consequences, and therefore are not falsifiable. But if you’re making that your basis for saying that ID is not science, then you are applying a demarcation criterion.
And that doesn’t preclude the possibility in principle that there could be a rational, scientific case for ID. I actually have a hard time imagining what a scientific case for ID would look like. What would be the test that would falsify it? What test could be done where outcome A is proof and the lack of A is disproof? I guess in theory there is a possibility that a test can be designed for anything. But I just can’t imagine how you would test for ID.
I actually have a hard time imagining what a scientific case for ID would look like. What would be the test that would falsify it? What test could be done where outcome A is proof and the lack of A is disproof? Suppose we found an ancient extraterrestrial city with records of how the ETs had conducted genetic engineering of life on Earth at various times during our evolutionary history to produce a desired result. That would be pretty compelling evidence for ID.
I even think that a certain degree of imprecision is usually necessary for a hypothesis in its early days. Amen!
I still think that we identify things on an exemplary basis – not exactly Dupré’s “family resemblance” account, but something similar. In this case we can tolerably distinguish science from nonscience, as Burke would have said. It is not that there are a set of necessary or sufficient conditions to be scientific. It is not that there is some majority of conditions either. It is that overall the proposal satisfies the conditions that things we do know are science satisfy; it may be a majority, it may be all of those criteria the exemplar satisfies, it may be within spitting distance in some logical space. Whatever, we know when it is science because it closely resembles other science. Hence, any proposal (say, Madam Blavatsky’s theosophy) that fails to be like other science isn’t science. The problem only arises when things get ambiguous (like phlogiston theory or string theory). Creationism and ID aren’t even in the ballpark, demarcation criterion or no. I can imagine some latter day logical positivist pulling his hair out in handfuls as he reads the above and screaming incoherently, “that’s not philosophy of science that’s sociology!!” Oh and by the way, I totally agree with the quoted paragraph.
There is a reason why logical positivism is no longer accepted; as practised in the heyday, it simply fails to describe science. Whatever it is that it does describe isn’t science but something else by the same name. More recent logical empiricists, on the other hand, would have a hard time disputing the above.
You do of course realise that the passage of yours that I quoted and with which I am in total agreement is dangerously Kuhnian in concept and would evoke cries of ‘elitism’ and ‘thought police’ from Lakatos.
One thing that people often fail to get about my take on science is that it is not normative. I don’t know what should be done in science, I only know what is. So if someone needs a definition of what “true” science should be, they will not find it in my ideas. At best we can do only a descriptive epistemology. Success in science is not guaranteed by method or by desire. So Lakatos, who still harbored hopes of a normative epistemology, is in trouble in that respect (well, he’s not, as he’s dead, Jim, but anyone who takes his line is).
At best we can do only a descriptive epistemology. Success in science is not guaranteed by method or by desire. I couldn’t have expressed it better myself.
I confess to having been influenced by Lakatos and Laudan, a pair of normativists. And if I had hair to spare, I’d be pulling at it. A normative epistemology of natural science doesn’t need to guarantee success… I suppose I should also confess to having been influenced by Pierce. Seems to me that description of scientific practices is not the same thing as an epistemology of science.
There’s a strong flavor (or flavour) here of “I may not be able to tell exactly when day ends and night begins, but I sure as Hell can tell the difference between them.”. I like that, having taken philosophy of science from one of the last of the Vienna Circle folks (Herbert Feigl) who left the fold. “In the end,” Feigl said in class one day, “there’s the philosophy of ‘nothing but’ and the philosophy of ‘something more,’ but I teach the philosophy of what’s what.”
Pretty much. Readers will recall that I quoted Burke on night and day (day and night) a little while back. It’s something I am working on. Descriptive epistemology is not, in itself, the whole of epistemology, of course, but I think on more general grounds (Humean inductivism) that the best we can ever do is predict based on the past, and if things are imponderable then we simply can’t tell what will work in the future. Any normative epistemology is, in my view, a summary of what is working best at the moment.
normativist: AAAAAGHHH!!! as Calvin might say to Hobbes. There are some neologisms that should be instantly banned on grounds of unspeakable ugliness.
Born prematurely on April 5, 1588, when his mother heard of the coming invasion of the Spanish Armada, Thomas Hobbes later reported that “my mother gave birth to twins: myself and fear.” [Wikipedia]
I confess to having been influenced by Lakatos and Laudan, a pair of normativists. Imre Lakatos influenced me more than any other single individual both in the philosophy and the history of science. However consequent following of the direction that I perceived as having been set by Lakatos in the history of science led me to the inescapable conclusion that all normative or prescriptive epistemologies of science are totally inadequate as attempts to capture the reality of science. Science evolves and its methodologies and standards evolve too meaning that any normative or prescriptive epistemology of science is at very best a snap shot of ‘one area’ of science at ‘one point in time’ and will prove false or inadequate when applied elsewhere.
I particularly enjoyed the arguments raised in “The Demarcation Problem…Again.” A think I’ve found the perfect image from 1641 to accompany the debate. http://www.britishmuseum.org/collectionimages/AN00047/AN00047532_001_l.jpg Henry Peacham’s “The World is Ruled and Governed by Opinion.” Dame Opinion is held in a tree watered by a fool. The tree is spontaneously generating a collection of very strange fruit about to take flight into the world. The fruit of the tree is a collection of cheap pamphlets and broadside ballads that inflame the humours of public opinion. Dame opinion is blinded by the weight of the Tower of Babel that sits on her head. She has a globe on her belly, a rod in one hand and a chameleon in the other. The chameleon can assume all colours apart from white which would seem to reflect Dame Opinion’s ability to transform into any form other than truth. Elisabeth Neumann suggested that in the print the fool waters and imbues Public Opinion with life and that it is up to us to imagine the guise that present-day fools who water public opinion my take.
If one’s epistemology is descriptive, not prescriptive, does that mean that even if, by following the rules, one comes up with wrong answers, that’s OK? How about applying that to deductive logic? The Axiom of Choice – as mathematicians generally use it, that’s all there is to be said about it? (Given that we now feel confident that it will never be disproved.)
What counts as a “wrong” answer if there is no general epistemological prescription? There are answers that satisfy or fail to satisfy local standards, and for science, that is enough. The standards, too, are evolving. And yes, I think that logic is also evolving, but that doesn’t mean we won’t eventually end up in stasis or converge upon agreed solutions. But don’t misundertake me, as Cheech and Chong said. I’m not proposing that all there is to science is what scientists do. Scientists also wear jeans and play sports. What I am suggesting is that science has exemplary cases, and we measure novel ones against that, because science has overall value to society, culture and human needs.
If one’s epistemology is descriptive, not prescriptive, does that mean that even if, by following the rules, one comes up with wrong answers, that’s OK? Every answer that natural science has ever produced has in the long run turned out to be wrong in some way or some aspect and the assumption is that it will continue so in the future.
Ideas from natural science also influence the wider culture and ethnology of the societies in which they are such an object of discussion and have done so for a long period of time. In this sphere it is not important wither the rules have been followed or if the answer is correct. As it is not a scientific or philisophical answer that is being sought.
Thony C – “…any normative or prescriptive epistemology of science is at very best a snap shot of ‘one area’ of science at ‘one point in time’ and will prove false or inadequate when applied elsewhere.” Similarly for any descriptive epistemology of science, right? And note that a normative epistemology, doesn’t have to aspire to eternal truths. As Lakatos argued, normative standards, even in the formal sciences, evolve in tandem with substantive developments/discoveries. “Every answer that natural science has ever produced has in the long run turned out to be wrong in some way or some aspect and the assumption is that it will continue so in the future.” Shades of Laudan! Though it didn’t drive him to descriptivism (AAAAAGHHH!!! ).
I will one day blog about the Pessimistic Metainduction. Let’s just say that experience teaches me that not all science eventually ends up being wrong, and I make a generalisation from that… 🙂
I’m not proposing that all there is to science is what scientists do. Scientists also wear jeans and play sports. I wonder whether it would help to separate Science from The Scientific Method. I love Feyerabend’s argument that organised crime fell the right side of the demarcation tracks, but if you’re demarcating The Scientific Method this is fine. But the reason organised crime isn’t science is because it’s not studying the right thing. Perhaps Science consists of what questions scientists ask when they are Doing Science, and also The Scientific Method that they use to answer it. I will be most disappointed if the next time I look in here there isn’t a list of N philosophers who have made the point, in greater detail, more subtly and with much longer words. And probably explaining why this approach is naïve and wrong.
Snowflake: What counts as a “wrong” answer if there is no general epistemological prescription? There are answers that satisfy or fail to satisfy local standards, and for science, that is enough. The standards, too, are evolving. But to remain science they must evolve in some logical way, and not just change arbitrarily. Suppose that the vast majority of scientists turned fundamentalist and adopted the standard that a literal reading of the Bible trumps all other evidence. Would that make that a valid scientific standard, and make YEC a valid scientific inference? Surely not. (If this example sounds too sudden to be considered “evolving”, imagine that the change happened gradually. Scientists might start by giving only a negligible weight to Biblical evidence, and gradually increase that weight over time.) If scientific standards are not to be just arbitrary (as postmodernists might claim) there must be some meta-standard for judging them, even if that standard is applied intuitively and not explicitly. It seems to me there are two possibilities: – Scientific standards are judged by their results, e.g. they lead to theories that allow us to predict observations better than do alternative standards. – Scientific standards are logically derivable from more basic standards, which in turn were judged by their results. (I would say that the most basic principles of induction are probably innate in us. It was natural selection that judged them to be effective.)
I am not sure if Snowy’s comment about pessimistic meta-induction was aimed at me or at the people with whom I am debating but I will give my take on scientific realism anyway. I am a realist, I am a scientific realist, I am to some extent a naïve realist; I believe that there is a real world out there that we can perceive and observe with our senses and that science is the only available system to analyse and explain this real world. I think science delivers explanations for phenomena that we observe in this real world. I also think that with time these explanation have gotten better and better and this is what scientific realists mean when they say that our scientific explanation get closer and closer to the truth, whatever that may be. I personally don’t think that one needs to invoke the concept of truth; one only needs to consider the concept of utility or usefulness. Our scientific explanations of the real natural world have gotten better with time simply because we can do more with them. When we apply our explanatory theories in order to control, change or influence the world in which we live we can observe whether our explanations are useful or not. Our scientific theories when utilised in applied science and science based or aided technology are effective and work; we can land men on the moon and bring them back again; we can effectively eradicate diseases; we can construct various types of motorised vehicles; we can… Our theories work when applied to the real world and over the millennia our ability to utilise the products of science have expanded immensely. I am not sure if one can quantise this progress in any sort of effective manner but anybody who would deny it is either blind, dumb or a liar. John is fond of quoting Ian Hacking on the ontology of physical particle who says (paraphrasing) ‘if I can utilise it, it exists’; I can spray electrons therefore they exist. I argue that we can utilise the results of science therefore science has an objective reality but I am still convinced that a ‘science truth machine’ that is a normative epistemology will never exist.
It would seem to me that what is at issue are sciences relationships with a wider world on the issue of i.d. and its relationship with wider cultural history. Once you have a full understanding of the basics it would seem to my very philosophical untutored eye that this is the time to start gazing in philosophical reflection. I’ve certainly put a lot of the philosophical questions I need to engage with on hold as they can wait until I’ve stood the material on its feet and let it run around rather than sitting on the floor and analyzing the script. But then my only experience of philosophy comes from the dancing master of the Old Vic Theatre School where I first trained before University.