Intelligent Design’s Theodicy Problem
Most critics of Intelligent Design focus on ID’s status as a scientific theory, or rather, non-status. Intelligent Design, unlike scientific theories as traditionally understood, is not immediately falsifiable, meaning the validity of the theory cannot be disproved by empirical scientific research. For these critics, ID’s non-theory status is enough to warrant its exclusion from the science classroom, and, at the risk of incurring the wrath of Pat Robertson, rightly so, in my opinion. However, I do think there is room for ID in a college-level “philosophy of science” class that addresses the greater issues of knowledge, empiricism, objectivity, and how we know what we think we know.
For those of us who also possess religious inclinations, Intelligent Design is not just bad science, but bad religion as well. As J. M. Tyree writes in The Revealer, Intelligent Design has a theodicy problem. Theodicy is the study of God’s goodness, power, and knowledge as they relate to issues of suffering and evil in the world. We’re all familiar with the questions: why does an all-powerful and loving God allow evil? Does God cause suffering through disasters, natural or man-made, and if He does, how can he therefore be considered good? Christianity has wrestled with these and similar questions for two millennia and come up with answers ranging from Open Theism (which questions either the all-knowing or all-powerful nature of God) to God controlling everything according to His unknown Will for His Sovereign Glory (the more fatalist fringes of Calvinism). While arguments over theodicy rarely convince outsiders, those of us already possessing belief have a plethora of positions where we can find answers and comfort. Tyree argues that ID is quite deficient in theodicy, and could reveal a creator who is neither intelligent nor a good designer. Or the creator could be malevolent. Tyree’s question “if God [of the Gaps] is willing to meddle with the inner workings of the bombardier beetle, why won’t he put a little extra spin on a hurricane to make sure it doesn’t hit any major cities?” is a classic question of theodicy that ID has not answered–indeed, it cannot.
The theodicy problem is one of the myriad of reasons Christians should not use it as a means to get faith issues into the classroom. The god of Intelligent Design is not even an identifiably Christian god. Any religion that posits the existence of a supernatural creator-being can fit under its umbrella. At best, Intelligent Design represents a warmed-over deism, with an eternal watchmaker god; at worst, as Tyree illustrates, Intelligent Design supports Cthulhu, who would probably do to Dover, PA exactly what Pat Robertson has in mind. Christians who are identifying too closely with the ID movement are making a mistake. Intelligent Design is questionable science and incomplete theology.
24 Responses to “Intelligent Design’s Theodicy Problem”
Tyree’s question “if God [of the Gaps] is willing to meddle with the inner workings of the bombardier beetle, why won’t he put a little extra spin on a hurricane to make sure it doesn’t hit any major cities?” is a classic question of theodicy that ID has not answered–indeed, it cannot.
Well, if a Christian believes that God is not involved in the workings of either a beetle or a hurricane, then I would submit that his beliefs are inconsistant with Scripture.
Kant would have no trouble agreeing with this view, but I think that orthodox Christianity is quite clear in reguard to the ongoing influence of God in creation. To deny this would cause many problems in developing a theology consistant with Scripture.
I agree with your larger point. ID is not a Christian concept in any way. A Muslim or a Satanist would have no problems with it. In this light, it has the same problems as Christians fighting for school prayers. It is questionable to me that an agreeable prayer for a large number of Americans would be useful for any one of their religions.
Oh great thing of moderate to perfect power, please allow us to proceed with our day, if you don’t mind too much. Thanks.
Regardless of the correctness of ID, it in no way “worsens” the theodicy problem (not that there is one in the first place). ID in and of itself cannot be bad theology, for in its purest expression all it does is extend the argument that God created all (a hallmark of theism) to God left behind evidence that he created all, which is quite consistent with scripture (Ps. 19:1 and Rom. 1:20). This is independent of whether or not he used evolution as a secondary cause.
Tyree argues a non sequitur. For example, here is a truly nonsensical statement: “The most devastating objection is that even if you assume the world was designed, it does not appear to be designed by a very nice deity. Bearing in mind that the Christian God must be all-powerful, all-knowing, and all-good, wouldn’t there be some way for God to prevent events like the tsunami, Hurricane Katrina, or Nagasaki?” This is an old question, with well known answers, and it has nothing to do with ID.
Tyree is making a mistake, a common one in fact, that for Christians ID means God designed the universe and all that is in it. No, for Christians, that is a given, independent of ID. ID is only about whether there exists any incontrovertible physical evidence of God’s handiwork. Any problem with a “malevolent” or “suboptimal” design applies to God’s creation (and has been asked for millennia) with or without ID. You don’t need ID to ask “couldn’t God divert a hurricane?” (Answer: yes he could.)
Mr. Heddle,
Your comment is outstanding in representing a Christian perspective on this. Tyree tries to argue from high-minded rationality, but in reality his argument is the same as a child’s: “Why do bad things happen?”
I agree with David. There may be a theodicy problem for Christianity (I think there is, but not the one usually cited; the mere existence of evil is not inconsistent with an omni-benevolent God, but the actual ordering of evil surely is), but ID does nothing to make it any better or worse. There are other, more persuasive theological arguments against ID, however, the most obvious one being the argument made by my colleague Howard Van Till. He points out that the theological problem with ID is that it argues on the one hand that the universe and even the specific position of the earth were created precisely as they are in order to allow life to exist, but on the other hand that the job was done so poorly that God had to continually intervene along the way to insure that existence.
Ed,
but on the other hand that the job was done so poorly that God had to continually intervene along the way to insure that existence.
That argument assumes that a poorly created universe is one that requires constant intervention. I do not see a reason to commit to that idea. What if it was God’s intent to intervene in the world?
Nothing about ID should cause theological problems with a particular belief in God except the belief that God does not exist. God or gods can be good or evil or neutral, but that is independent of the fact that they were accountable for creation.
I don’t think Tyree’s article did what he intended it to do. He basically just made the argument that IDists have been saying all the time – that ID doesn’t say anything about the designer. And this seems to have the same effect on theodicy as, say, the cosmological argument for God would have on theodicy – i.e., none.
Also, the notion of “immediate falsifiability” is pretty much philosophically and historically defunct. There are no “crucial experiments.” I would suggest taking the advice of Van Till’s other colleague – Loren Haarsma – on this matter when he proposes this for ID opponents to consider:
“Don’t play the demarcation game, that is, don’t insist on definitions of science which try to exclude ID. Don’t insist that ID must posit a specific mechanism in order to be ’scientific.’ … Instead: evaluate the scientific parts of ID on their scientific merits; evaluate the philosophical parts of ID on their merits; and evaluate the theological parts of ID on their merits.”
At best, Intelligent Design represents a warmed-over deism, with an eternal watchmaker god; at worst, as Tyree illustrates, Intelligent Design supports Cthulhu, who would probably do to Dover, PA exactly what Pat Robertson has in mind.
Best line of the week. Scared my cats half to the side of the Great Old Ones from my hysterical laughing fit.
Thanks, David. I’ve been needing that in recent days.
I think I need to take issue with your “college-level” description. Questions of whether process==purpose need to be brought up much earlier. We can spare the kids the really hardcore details, but just letting them know that there are alternatives to reductionism would help.
I don’t think the kids need a whole quarter of philosophy, but the lack of it certainly stands out.
Macht wrote (quoting Haarsma):
Which parts are the scientific parts, exactly? The only thing I think could possibly be contrued as such are their criticisms of evolution, but what kind of scientific “part” is that? It’s not a hypothesis or a theory, it is analysis of another hypothesis or theory. And the structure of the argument includes a decidedly illogical conclusion (if not evolution, therefore God).
The primary criticism of the ID movement, irreducible complexity, doesn’t seem to get us very far at all. Behe admitted on the witness stand in Dover that he has read only a small fraction of all of the papers on the evolution of the systems he claims are IC. He also claimed that his ideas are testable, but admitted that he has no intention of performing those tests despite having the opportunity and the expertise. And the closest he has come to doing so, his paper with Snoke (a computer simulation) showed, by his own admission, that IC systems could in fact evolve even under the most restrictive possible formulations.
There is no ID model that can be tested, is there? There is no model of the natural history of the earth that explains the patterns of life’s appearance either temporally, anatomically, or geographically. There is no proposed mechanism. What exactly is there about ID that is open to scientific verification? It appears to me that it relies almost exclusively on personal incredulity, that you have to convince them that evolution is true or they will claim it’s not. And that’s hardly a scientific proposition.
As far as the “demarcation game” goes, I think much is made of it on both sides that is oversimplified. The ID advocates like to quote Larry Laudan, for example, and his criticisms of Ruse and Popper, but they never really finish the argument. Laudan says that there are no set of criteria that are both necessary and sufficent to demarcate science from non-science, but does this really mean that there is no demarcation at all? I don’t think so and I don’t think Laudan does either. At worst, it means there are some cases at the margin that are difficult to formulate, but that doesn’t mean that one can’t make a distinction most of the time. No one in their right mind would claim that a Shakespeare sonnet is a scientific theory, they are distinct categories of ideas, are they not?
Ack. The word “formulate” in the last post should be “formalize”. My apologies.
By the way, I should add that I agree with Macht that requiring a “crucial experiment” as a form of falsification is shoddy thinking. Falsification need not require any such thing, only what is termed “naive falsificationism” does so. The deeper falsification problem for ID, in my view, is that it can be consistent with absolutely any set of data. There is no possible set of data that could be construed as inconsistent with ID because of the unbounded nature of the alleged designer. Much of the cross examination of Michael Behe in the Dover trial dealt indirectly with this issue, as he essentially argued that ID was “consistent” with virtually anything from young earth creationism to common descent. If, for example, that at the genomic level humans differ from other species in exactly the proportion predicted by anatomical phylogenetic trees and branching times, then this is entirely consistent with a designer using similar structures to achieve similar results; if the evidence shows the opposite, then that is consistent with a designer starting from scratch and designing an entirely new structure to achieve a given result. X or ~X, either is perfectly fine in every conceivable area of the dispute. This is what scientists call a “sterile” idea – because it’s consistent with every possible set of data, there is no way to ever know whether it’s true or not because it isn’t testable – no matter the results of the test, they’re consistent as well.
Ed wrote,
“The deeper falsification problem for ID, in my view, is that it can be consistent with absolutely any set of data. There is no possible set of data that could be construed as inconsistent with ID because of the unbounded nature of the alleged designer.”
I disagree. There may be no set of data that is inconsistent with God–but there certainly is for ID. Once again, ID is not that God designed the universe, but that God left behind evidence for his design. For biological ID, demonstrated evolutionary pathways for the icons of irreducible complexity would certainly deal a fatal blow. Oh sure, some people would hang on, but that situation would not be unique to ID. Otherwise reputable scientists held on to the steady-state model of the universe for decades. For cosmological ID, proof of another universe would be fatal, as would proof that the fine-tuning is an illusion. You just can’t say, for example, people would argue that God designed the multiverse. No doubt they would–but it wouldn’t be ID they were arguing, they would just be arguing for God.
It becomes very clear in the theoretical ultimate limit–where science has explained all from a fundamental theory of everything, that in such a day ID would be dead. All we theists would have left is to claim that God is responsible for this universal and comprehensive law. But again, that wouldn’t be ID.
It is not the lack of falsifiability that makes ID “not science,” it is the absence of predictions.
David Heddle wrote:
No it wouldn’t. There are complex biochemical systems that have exactly the same traits that Behe uses to define irreducible complexity – multiple interacting parts that are all required for the system to function – that are accepted as having evolved without intelligent intervention even by Behe himself. Two obvious examples are the hemoglobin system and the antifreeze protein system in arctic fish. If those examples of the evolution of allegedly irreducibly complex systems doesn’t falsify ID, why would explaining two or three more do it? Behe chose those examples precisely because there is still some mystery as to how they evolved, but prior to having a good explanation for the evolution of antifreeze proteins he could have made the exact same argument for those. The ability to evolve an IC system without intervention cannot be evidence against ID; if it was, ID would already be falsified.
But how could that possibly be proven?
But falsifiability and making predictions are essentially the same thing. Making risky predictions is precisely how most theories are tested. If the predictions turn out to be false, the theory has to be revised or thrown out. Because ID is consistent with any possible set of data, it can’t make a risky prediction.
Haarsma’s answer to “which parts are scientific parts” is
(BTW, I’m getting this from these notes of a talk he gave.)
Ed, of course ID isn’t consistent with every set of data. You just argued that it isn’t consistent with hemoglobin! (Whether that is true or not, I have no idea.) And whether Behe gives up his ideas based on this is beside the point. If scientists gave up on their ideas in the face of anomalies, then science would go nowhere. For example, Less than a year after Einstein formulated his theory of special relativity, Walter Kaufmann had an experimental falsification of SR. Einstein thought SR was right even though he could find nothing wrong with Kaufmann’s experiments. Einstein, of course, turned out to be right even thought Kaufmann’s experiment wasn’t shown to be flawed until years later. And most people at the time agreed with Kaufmann, not Einstein. I’m not saying Behe is Einstein, I’m just saying there’s nothing wrong with him not giving up his theories even when the evidence points against them. But for you to say that ID is consistent with all conceivable empirical data on the one hand but is wrong because of the empirical data of hemoglobin and antifreeze protein systems just makes no sense. If all you are saying, however, is that ID can always add ad hoc hypotheses in order to explain away anomalies, then you are correct – but this is pretty much true of any scientific theory.
Macht wrote:
Okay, let’s take these one at a time. The first one is not a part of ID at all, so far as I can tell. I’m not aware of any ID advocate even writing on the subject, much less perform any research on it, and I’ve never seen anyone attempt to relate that to any ID theory or hypothesis.
The second one is, as I noted, appears to be nothing more than a measure of personal incredulity. They look at the hundreds or thousands of complex biochmical systems made up of multiple interacting parts, select the ones that are least understood at this point and declare them “irreducibly complex”. But they could have made the same argument for any other system before evolutionary biologists worked out the evolutionary pathways by which they evolved. Clearly, then, the mere fact that we don’t yet understand how a particular system developed is not a measure of the validity of ID or of evolution.
The third one again appears nowhere in ID work that I am aware of, nor does it entail ID at all. Molecular biologists are at work every day on questions of self-organizing complexity and it certainly doesn’t support ID, nor has it ever been integrated into any testable ID hypothesis.
No, you misunderstood my point. It’s not that ID is inconsistent with hemoglobin, it’s that the fact that we have examples of irreducibly complex systems with well understood evolutionary explanations not requiring intelligent intervention do exist, even by the admission of ID advocates. This should indicate that Behe’s arguments concerning ID aren’t much of a challenge to evolution, but they don’t. Why? Because he simply picks out those systems that aren’t well understood yet and points to them. But history shows that his conclusion does not follow from his premise.
You are conflating falsifying IC with falsifying ID. The IC argument is little more than an argument from personal incredulity, it doesn’t logically follow from the existence of complex biochemical systems. As long as there are systems that aren’t perfectly understood at any given time, this argument remains in play. But because the conclusion (”the system must have been designed”) doesn’t logically follow from the premise (”I don’t see how this system could have evolved”), it’s neither a good argument for ID nor a good argument against evolution.
I think the summary is this: we can’t prove beyond a reasonable doubt that God exists using only physical evidence. We can’t disprove it either. This has been the case for centuries.
God does not have to be a “god of the gaps” in order to exist and have influence on the universe. Even if we were able to perfectly explain the workings of the physical reality around us, we would be unable to use that knowledge to test anything that is not physical. We can also not disprove any non-physical entity having influence in our universe using only physical evidence.
These facts alone should be sufficient to argue that there ought to be at least lip service paid to the possibility that we are unable to explain all that is (”being” in a physical and non-physical sense) using “science” by whatever definition you choose for that word.
Ed, you’re right, I was conflating falsifying IC with falsifying ID. I think my point (”If all you are saying, however, is that ID can always add ad hoc hypotheses in order to explain away anomalies, then you are correct – but this is pretty much true of any scientific theory”) still stands, though. The fact that the broad theory of ID isn’t falsifiable doesn’t really mean much as far as science is concerned. (When I say the “broad theory of ID” I don’t mean what you’ve said above (”I don’t understand, therefore it must be designed”) – very few ID advocates make that argument. Rather, I mean something along the lines of “design is the best explanation for certain features of the universe.” This isn’t really falsifiable and I view it to be something close to Lakatos’s “hard core.” It’s too broad of a claim, not very specific, and not very “close” to the empirical data. This broad theory, by itself, isn’t really scientific, either. But when you combine it with specific theories and hypotheses – e.g., IC – these are what scientists can argue for and against. This isn’t really much different than any other scientific theories, though. The Newtonian “hard core” wasn’t falsifiable and it lived with anomalies for years. It eventually lost out to Einstein’s theories, but it wasn’t really falsified.
As far as Haarsma’s 1rst and 3rd examples, I don’t know. He doesn’t really have a reason to make them up since, last time I checked, he accepts evolution. And as far as the 2nd, he never said they were good scientific arguments – his point was just that they are scientific and should be evaluated as such.
Macht wrote:
Perhaps the source of the misunderstanding is that this is the first time you and I have ever discussed this, but I should make it clear that when I refer to ID I mean biological ID, not cosmological ID. I have no problem with the notion that the universe was created, or that it was created specifically to be capable of sustaining life. My objection is to ID as a biological argument against evolution, which I think fails miserably. At that level, it is little more than a god of the gaps argument, which have never been useful or proved valid in science.
If you want to go Lakatosian, then you need to translate Ed’s complaints into Lakatos-speak. I think this is how Ed would put it: ID isn’t even part of a degenerating research program, since usually that term is reserved for programs that were once progressive, and now are starting to flag. ID has never, ever been part of a progressive research program, and it shows zero signs of being able to do so. And that’s why it should be flat-out rejected as even a candidate for being scientific.
Ed and philospher,
If the issue at hand was whether you accept ID or not or whether it is good science or not or whether it is a progressive or degenrating research program or not, then I probably wouldn’t have chimed in in the first place. Whether it ever was progressive doesn’t matter – according to Lakatos what matters is if it can theoretically be progressive. Even sophisticated falsificationists would say this. He says that research programs should be given sufficient time to become progressive but this doesn’t help anybody right not – it only helps in hindsight.
Ed,
In regards to irreducible complexity, you are making a reasonable but (in my opinion) incorrect assumption that if the flagellum was explained by a pathway (something more than a conjecture) that Behe et. al. would just move on to something else. But when the paradigm collapses, so goes the theory, at least in many cases. Maybe the movement, surely weakened, would move on to another system. But at some point, if the same process recurs, the number of adherents would fall below a critical mass and those holdouts would become insignificant.
Regarding another universe or proving that fine tuning is an illusion, you wrote:
“But how could that possibly be proven? ”
I don’t know about another universe–because of General Relativity it may be impossible to detect one even in principle. For fine-tuning, well this is something people are trying to disprove. For example, if you demonstrate that the synthesis of Carbon and Oxygen in stars really isn’t very sensitive to the fundamental constants then that piece of fine tuning evidence would vanish.
“But falsifiability and making predictions are essentially the same thing.”
No. Detecting another universe would falsify cosmological ID, but the only prediction is a negative one: you will not find another universe.
Whether it ever was progressive doesn’t matter – according to Lakatos what matters is if it can theoretically be progressive. That doesn’t sound right at all, at least not with that “theoretically” modifier in there. The point of appealing to progressive/degeneracy is exactly to do the normative work of determining what should or should not be pursued. It’s also something that depends significantly on not just the would-be researchers, but on the world: no amount of theoretical fiddling can turn a doomed theory into a non-doomed one, if it just has nothing productive left (or never had anything productive) to offer. The whole point of the research program framework is to acknowledge a sense in which, as you said, in principle any theory is infinitely defensible with ad-hoc fixes — but to give a fix on why in actual, normatively-correct practice there is a difference between acceptable and inacceptable forms of such anomaly resistance.
Maybe a more relevant way of addressing your point is to note that the anti-ID crowd is always happy to acknowledge that if ID were to actually start making anything even remotely resembling progress, then there’d be something scientific to talk about. So, in one sense, it can take all the time it wants. But while it’s taking that time, and before it actually turns up anything, it doesn’t get to ‘pre-count’ as doing successful science on spec.
“But falsifiability and making predictions are essentially the same thing.”
No. Detecting another universe would falsify cosmological ID, but the only prediction is a negative one: you will not find another universe.
I do think that it is right that ‘making predictions’ is too weak a standard here. Making only ‘heads I win, tails we flip again’ sorts of predictions is not really putting a theory at risk in the way that can count in a theory’s favor. Only surviving a real, falsification-risking test can be a significant mark of empirical success for a theory. (Of course, sometimes a theory can be preferred even without new empirical successes — the drive to unification in theoretical physics might be like that, where it’s not so much about new predictions as about systematization & simplification.)
David Heddle wrote:
I think there’s something deeper here. It isn’t just that they would move on to a different pathway, it’s that they have chosen the ones they have because their argument is, at its core, disingenuous. As I said, there are other complex biochemical systems that meet the criteria of being irreducibly complex (multiple interacting components that do not function without each component in place) but that they accept as having evolved without intelligent intervention. That alone is enough to show that their argument – that irreducible complexity is evidence of design – is false. That’s why we keep getting into shifting definitions of IC, and now all of a sudden we have Behe arguing that a system could be “more” or “less” IC, where prior to that the argument has simply been that if a system displays IC, it must have been designed.