Friday, February 12, 2010

What method do we use to make determinations about God?

We have been having a discussion with Ten Minas Ministries regarding methodologies, specifically scientific method. It appears to boil down to this: If the theist agrees the scientific method is not able to make supernatural determinations—what method does the theist propose to use instead?

Ten Minas Ministries responded with a comment that needs some unpacking:

I propose that we test a theory. I observed in the past that when Mr. X was exposed to phenomenon A, he responded with reaction 1. So I want to perform an experiment using the scientific method. I expose Mr. X to phenomenon A again, testing to see if yet again he will respond with reaction 1. But low and behold, this time he responds with reaction 2. When I do the experiment again I observe reaction 3.

The scientific method cannot test intelligent agents because an intelligent agent can choose to react differently. That is why the scientific method is ill equipped to test claims of supernatural intelligence like the Christian God. That is why there are other disciplines like sociology, anthropology, psychology, etc. It isn't a cop out. It is a common sense limitation of the scientific method. It was never intended to measure intelligent reactions.

Not a single one of you relies upon the scientific method for every bit of knowledge in your life. Contrary to what the Barefoot Bum said, the scientific method is not simply "the method of drawing conclusions from evidence." Yes, it is one method of drawing conclusions from evidence, but so is philosophy. So is sociology. So is historiography. The scientific method employs experimentation. Therefore, if the question you are investigating is not subject to experimentation the scientific method is useless.

You do not conduct an experiment to decide whether or not it is probable that you will be struck by a car before you cross the street. Yet you certainly "know" that if you step into oncoming traffic you will end up in the hospital or in the grave.

Dagoods says the theist does not propose a method. Yet I just proposed an alternative method in our discussion on my blog. I specifically mentioned utilizing the methodology of philosophy, which involves applying inductive or deductive reasoning to agreed premises.

Sometimes the theist feels like we are banging our heads against the wall. We are told we are not providing a methodology, so we provide a methodology only to hear the atheist repeat the same accusation. We are told that we do not provide any reasons for inserting God but simply arbitrarily assign a "God of the gaps." So we provide our reasons only to have the same accusation repeated over again. We see the atheist repeatedly refusing to accept any methodology other than the scientific method all the while refusing to admit that for the vast majority of knowledge they have arrived at in their lives they did not use the scientific method. Then if we get frustrated by the virtual ignoring of everything we say, the atheist declares triumph by claiming that the theist has no answer for their arguments.

Repeating the same thing over and over again does not make it true. There is little else I can say to respond to your points other than to say that theists have proposed alternative methodologies over and over again. Ignoring them does not make this any less true. The theist repeatedly provides reasons why they fill these gaps with God. Again, ignoring these reasons does not make your argument true.


If we are proposing an “alternate” to the scientific method; I think it important to first know what we mean by “scientific method.” How would you know you took an “alternate route” if you didn’t know what the original route was?

The Barefoot Bum recognized this issue and specifically asked, “What do you think scientists mean by “the scientific method.”?

Wikipedia defines it as: “To be termed scientific, a method of inquiry must be based on gathering observable, empirical and measurable evidence subject to specific principles of reasoning. A scientific method consists of the collection of data through observation and experimentation, and the formulation and testing of hypotheses .“ or “A method of discovering knowledge about the natural world based in making falsifiable predictions (hypotheses), testing them empirically, and developing peer-reviewed theories that best explain the known data. “

Another ‘net resource states, “A process that is the basis for scientific inquiry. The scientific method follows a series of steps: (1) identify a problem you would like to solve, (2) formulate a hypothesis, (3) test the hypothesis, (4) collect and analyze the data, (5) make conclusions.“ Merriam-Webster defines it as, “principles and procedures for the systematic pursuit of knowledge involving the recognition and formulation of a problem, the collection of data through observation and experiment, and the formulation and testing of hypotheses “ Ten Minas Ministries gave similar definitions here.

We see a common theme running through these descriptions: Testable hypothesis, data observed (at times through experimentation,) and a theory best explaining the known data. It should be noted this is an ongoing process in that the resulting theory becomes the new hypothesis, and new data may require an additional revision (or even complete elimination) of the new hypothesis.

We quickly realize why the scientific method is not utilized by the theist, as by definition supernatural is outside the natural realm, and is therefore untestable, unobservable, and unverifiable by the only means at our disposal—the natural world.

Ten Minas Ministries appears to propose two solutions to the theists’ problems in this regard:

1) Scientific method is insufficient to make determinations when intelligent agents are involved; and

2) Philosophy is an alternative method;

Dealing with the first he uses the example of phenomenon acting on an individual, who may exhibit different reactions. Because of these different reactions, Ten Minas Ministries claims scientific method is unusable. This is not true at all. In fact we use the varying reactions to make statistical determinations, through the scientific method.

Take the phenomenon of my cutting off another car in traffic. Now look at a number of proposed reactions from the driver I cut off:

1. She does nothing.
2. She honks her horn.
3. She gives me the finger.
4. She strikes my car with her own.
5. She gives me a three-week paid vacation to Tahiti.
6. She develops pancreatic cancer.
7. Her car turns into a DeLorean Time machine.
8. She obtains the ability to shoot laser beams from her eyes.

Every single person, in reading this list, makes an assessment as to each reaction’s viability. Why do we immediately eliminate DeLorean’s and laser beams? Because our own observations, other observations we read, any test we perform all results in no DeLorean’s and no laser beams. We may not consciously be using the scientific method—but that is exactly how you immediately eliminate those possibilities.

We can go further. By observing and collecting data, even though horn-honking, finger-giving and car smashing are possibilities, we can make statistical determinations as to the likelihood of each. Yes, an individual CAN decide to smash my car, but my experience, one could say my numerous experiments in this field, indicate this possibility to be so low, I decide to do it anyway.

Curiously, Ten Minas Ministries uses the example of stepping into oncoming traffic, and how we do not experiment this, yet we know we will be seriously injured if we do so. I am not certain what he means that this hasn’t been experimented. We may not have personally done so, but we read and see pictures and video over and over and over demonstrating in the fight between pedestrian and automobile—automobile wins. Every time. I don’t have to personally experiment every claim—I can use the experiments of others.

I am seeing a confusion over the term “experiment”—it is being narrowly defined to include replication of the exact theory proposed. In other words, to experiment whether I will be injured if I walk in front of a speeding car, I must personally do so, rather than view the experiments of others.

I would disagree the scientific method is so narrowly defined to a certain person in a certain situation MUST obtain a certain result, only through experimenting that exact situation. If this is what Ten Minas Ministries is proposing, we are certainly coming at it from two very different foundations.

Continuing in my thread, he uses a few examples that further demonstrate this possible confusion. For example he indicates historical determination is outside scientific method, stating, “We cannot conduct an experiment to determine whether or not Nero truly was emperor of Rome...” I am uncertain what he means by “experimentation.”

Using the scientific method, we look first at the hypothesis: “Nero was Emperor.” We look at the data—historical books, coins, statutes, opposing opinions, world results, etc. We can “experiment” (loosely) by comparing other leaders and other results and see the same consistent results. Other leaders have coins. Other leaders have statutes. Etc. And again, even under this method, new data could cause a change in hypothesis.

Look at this another way. We cannot replicate the Big Bang. Does this mean we are not using the scientific method to determine its existence, because we cannot “experiment” by creating another Big Bang? We cannot experiment with a non-heliocentric solar system by making another solar system. We cannot experiment with making stars go super-nova—does this render such determinations outside scientific method?

Vinny made a similar observation when it comes to history.

I was surprised Ten Minas Ministries differentiated certain fields as being outside scientific method, such as psychology. Actually, there is experimentation and data observation—any person familiar with such things as the Minnesota Multiphasic Personality Inventory (MMPI) or the whole battery of neuropsychological testing understands the use of scientific method in this field. Granted, it is not as precise or studied as other fields such as chemistry—but that does not remove it entirely from scientific method. Sociology can use experimentation.

There is debate whether Anthropology uses scientific method. Notice though, in that debate is the proposed alternative method is defined and addressed.

What is the proposed alternative method given here? “Philosophy.” Philosophy is the search for knowledge. It may inspect how one goes about that search, or what one utilizes. I am uncertain, though, what is meant by “Philosophy” as a method. I know you may feel as if you are beating your head against a wall, but I am truly stumped by what you mean as “philosophy” as an alternative method.

We seem to be in agreement in definition and understanding regarding “scientific method.” Both your definition and the others I found lay out certain steps that appear pretty clear.

What “steps” are proposed in the method of “philosophy”? It’s like saying, “we use the method of communication” which is really no method at all, merely the only way we can relate an idea to another human. What is the method of “philosophy”—can you describe it in alternative terms to scientific method?

Ten Minas Ministries, I agree whole-heartedly none of us use the scientific method for every bit of information in our lives. We are influenced by culture, upbringing, unique situations, etc.

This is exactly the point of this whole exercise! We recognize our own tendency to, even unknowingly, insert our own bias in determining. We want--we really, really want--that person to be a wonderful spouse, so we overlook and rationalize away the fact he is a convicted wife-beating deadbeat. “He’s changed.” We ignore the statistical evidence against it.

This is why I harp on method. To remove the very problem you identify. If we stick with the method, we can both recognize such bias and remove them from our determination.

I may want the canonical gospels to be historically accurate. I may really, really want it. In recognizing that desire, I realize the bias will impact my study. They may still be historical; they may not. I look for a way to root out such bias by creating a consistent method whereby the possible outcomes are: historical, partly historical, not at all historical or unknown. Only by sticking with the method can I hope to remove my bias—even if (and ESPECIALLY if) the results are not my desire.

I know you feel as if you are beating your head on a wall, but I truly do not see what method you are proposing, in lieu of scientific method. What steps are you saying we take in a “philosophical” method?

11 comments:

  1. Because there appears to be some confusion about the difference between deductive and evidentiary reasoning, I've written a short essay on the subject: Evidentiary and deductive reasoning:

    "Although related, deductive and evidentiary reasoning work in "opposite" directions. Deduction asks the question: what interesting statements are theorems of this formal system? Evidentiary reasoning asks the opposite question: in what formal system are these interesting statements theorems?"

    ReplyDelete
  2. A couple of other notes...

    First, skeptics typically use science (at worst) as a synecodche, to refer to the general case of any evidentiary methodology with a foundation in observation and experience. Quibbling over categories of science is unhelpful.

    We quickly realize why the scientific method is not utilized by the theist, as by definition supernatural is outside the natural realm, and is therefore untestable, unobservable, and unverifiable by the only means at our disposal—the natural world.

    This statement is not strictly true. First of all, it's not really coherent to talk about the "outside" of the natural world, because -- at least to science -- the "natural" world is just the world of everything that's knowable. The construction of "supernatural" as "outside the natural world" assumes incorrectly that "natural" is an ontological designation; it is, rather, an epistemic designation.

    Suppose, however, that we were to charitably grant the theist the definitions necessary to make the supernatural "coherent". Suppose we take "natural world" as an ontological designation, perhaps as synonymous with the physical world. Suppose that it is coherent in this sense to talk about a supernatural deity "outside" the physical world. Suppose even that science does in fact restrict itself to only the physical world. (None of the above are actual metaphysical principles of science, but let's suppose.)

    Even then, this supernatural deity would have to affect the physical world, and therefore its effects would be physical and ipso facto part of the domain of science, and susceptible to study. Furthermore, if the theist's position had any validity, these effects should be impossible or very difficult to explain in terms of a few simple, mechanical causal processes, in just the same sense that the actions of human beings are impossible or difficult to determine in terms of a few simple, mechanical causal processes. To explain human behavior we have to assume not just the laws of physics, but that an enormous amount of state information has been physically instantiated in the real world, and the state itself has causal efficacy. (In contrast, thermodynamics, while dealing with the interaction of many individual molecules, does not account for how the individual molecules are arranged, i.e. the state information, to make predictions.)

    Therefore, even granting the most charitable assumptions to the theist, the argument that "science" cannot consider the supernatural because it is outside the domain of science fails. Either the supernatural has no effect whatsoever on the physical world, or science can consider its effects, and we should be able to infer its existence from those effects.

    Note that Ken Miller's suggestion that a supernatural deity operates by "unobservable" intervention (e.g. by some form of the "butterfly effect") doesn't hold water. If such a deity's interventions were nontrivial in overall effect, we should be able to detect its interventions statistically. We should see that we're "getting lucky" far more often than we should be. Four billion years to develop multicellular life and another five hundred million years to develop intelligence does not seem at first glance to constitute extraordinary luck, especially when it looks like human beings will be able to do the same trick in a couple of centuries.

    ReplyDelete
  3. You are of course correct that even in the most restricted sense, we can acquire knowledge about psychology and human behavior using double-blind experiments. However, you're taking Ken's "bait" here. He wants to take the focus off of his methodology and onto scientific methodology, especially in its most restricted, experimental sense. On the one hand, your argument is sound, but you don't hammer home the conclusion: we can indeed at least know something about human behavior using experimental science; the argument that science is useless because a deity is intelligent fails: we might not be able to know everything, but we should be able to know something, and we do not.

    Even so, Ken can overcome this objection: the deity is intelligent; there are some things that we can't know about intelligence using experimental science; perhaps the deity is just that sort of intelligence that has only those properties that we cannot know using experimental science but we can know by other means.

    I became discouraged by Ken's confusion between deductive and evidentiary reasoning and didn't keep reading, so I don't know if he introduced yet another methodology subsequently. You yourself are a careful reader, however, so I will conclude that Ken hasn't introduced anything but the "philosophical" (scare quotes), which he describes as "applying inductive or deductive reasoning to agreed premises."

    (It's unclear precisely what sense of "inductive" he means here; induction has a specific meaning in logic, and is typically defined as a specific premise or inference rule in a formal systems. I strongly suspect that Ken doesn't really understand the difference between mathematical induction and evidentiary reasoning. The distinction is nontrivial to understand, and is probably due to lack of education, not willful ignorance or intentional obtusity.)

    However, regardless of the limitations and restrictions of experimental science, Ken's choice of alternative method, the philosophical method, is odd because it doesn't help him in the least. Philosophers have been using the philosophical method for millennia to attempt to establish any kind of knowledge, and they have failed so spectacularly that we had to invent a new method — evidentiary reasoning — to make any kind of progress in acquiring knowledge.

    Not only has the philosophical method failed to acquire knowledge, philosophical arguments for the existence of a deity fail according to the standards of the philosophical method itself! One cannot apply deductive or inductive (?) reasoning to agreed upon premises and conclude the existence of a deity. To conclude the existence of a deity, one must introduce controversial premises, or use something other than deductive or inductive (?) reasoning.

    So I don't think Ken's complaint is legitimate. The theist has not introduced an alternative methodology, or at least not any methodology that actually helps them. (Or if they have, they have buried it in unjustified or irrelevant quibbles about some supposed "a priori" bias of the skeptics' preferred scientific method.)

    ReplyDelete
  4. The Barefoot Bum,

    Thank you for the essay on deductive and evidentiary reasoning.

    Yes, I am guilty of “taking Ken’s bait” here, because I am confused as to how he is differentiating between this “alternate method” being proposed and what he means by “scientific method.”

    This confusion comes from the examples, such as experimentation needed to cross the street. I get the impression Ten Minas Ministries is limiting what he means by “scientific method” to those situations where precise conditions produce precise results. As if “scientific method” means predicting unique results. 1 gram of this substance added to 1 gram of that substance with specific atmospheric conditions will always produce a specific result. Since intelligent agents are not precisely predictable, he seems to exclude anything utilizing such agents.

    What I was attempting to point out (and see if he could explain how his alternate method is different) is that we can use the same method of postulating, data observation and prediction even when intelligent agents are involved. We can calculate statistical probability, albeit not 100% precision. We cut off cars in traffic based upon such statistical probability, even then there is a remote (less than 1%) chance the other driver will pull out a gun and start firing bullets.

    The problem comes, of course, how to determine statistical probability without the ability to observe. When I was differing “natural” between “supernatural” I was (apparently ineffectually) attempting to demonstrate “supernatural” is deliberately placed outside our realm of knowledge—therefore eliminating the ability to observe or make statistical probability. Once the supernatural crosses the plane (as it were) and enters the natural, as it must do for us to obtain knowledge, we could then generate statistical probability.

    The problem remaining as to how to determine when the supernatural has crossed the plane.

    This is where I remain unclear as to what method Ten Minas Ministries proposes, in lieu of data observation, to make such determinations. Simply put—how can we tell the difference between:

    1) Supernatural intervention; and
    2) Natural intervention not known.

    As to what he means by “philosophy”—I will use his own words and let you decide:

    As for the proper methodology, I have already mentioned one on several occasions: philosophy. In philosophical terms, this is a matter of seeing if we can agree on certain starting premises. Much debate can be had on which premises are appropriate and which are not. Sometimes these premises may be things that are derived by using the scientific method. However, if we agree on certain premises (because they are scientifically derived or because we find that existentially we cannot live life without them), then the discussion turns to whether or not the existence of God can be logically derived from those premises.

    Therefore, observation, necessary presuppositions and logic can be an alternate methodology.

    ReplyDelete
  5. Yes, I am guilty of “taking Ken’s bait” here, because I am confused as to how he is differentiating between this “alternate method” being proposed and what he means by “scientific method.”

    He's essentially trying to muddy the waters so we will not look too closely at the details of his own methodology. If his preferred "methodology" is just one among a panoply of others, then the details fade into triviality. Who cares if it's nothing like science, or nothing like the study of history, or even nothing like psychology? They're all nothing like each other anyway. If we're swallowing the camel of ninety-nine completely different methodologies, why strain at the gnat of a hundredth?

    ReplyDelete
  6. [The philosophical method] is a matter of seeing if we can agree on certain starting premises... [I]f we agree on certain premises... then the discussion turns to whether or not the existence of God can be logically derived from those premises.

    Mere agreement on premises a very weak foundation for drawing conclusions. We might agree mistakenly, and we can undermine the process by just by disagreeing on the premises. And, in truth, philosophers don't agree on the premises. Ever. Heck, even mathematicians have trouble agreeing on premises, such as the axiom of choice.

    But, more importantly, Ken glosses over precisely what we want a methodology for! how precisely do we agree on the premises?

    Much debate can be had on which premises are appropriate and which are not. Sometimes these premises may be things that are derived by using the scientific method... because we find that existentially we cannot live life without them.

    This is exactly what you're talking about here, Dagood. Ken introduces his method only after the interesting work has been done, i.e. agreeing on the premises, which he leaves vague. That's like describing peer review and calling it the scientific method.

    ReplyDelete
  7. I've thoroughly enjoyed the exchange among the few posts and at the theist site discussing the problem of their refusal to appreciate, or confusion over, the limits of scientific exploration when it comes to supernatural explanations.

    I especially enjoyed The Barefoot Bum's concise treatment of the natural/supernatural subject to weed through the gobbledygook and show why the theist assumptions about science being too limited to deal with a personal/interactive god being somehow beyond the scope of a naturalist methodology.

    Either their god has some impact on reality that would put it within the realm of scientific discovery and scrutiny, or it's purely conceptual and ends up being little more than fun thought experiments in philosophy classes.

    They seemed to have successfully backed themselves into this corner on their own accord. And where mathematics is overwhelmingly conceptual in nature, it makes a handy tool for building models for real concepts.

    God concepts merely allow explanatory place holders that cannot themselves be explained. Worse, it is generally argued that we cannot presume to even understand them as human beings... which seems fairly useless to a human being who seeks answers.

    They may not want to arbitrarily fill in any and all gaps with god, and may be perfectly open to scientific discovery on some matters... but from my skeptical viewpoint, it seems indistinguishable from merely attempting to avoid the negative connotations of the term, while still employing the tactic where it suits them.

    They may feel they are gaining a better understanding of the universe by lowering the burden of proof for their religious convictions, but they simply cannot show how they can prove this to themselves, let alone a skeptic. Endless requests for some sort of methodology to differentiate between wishful thinking and actual useful insight have gone generally unanswered.

    The best they seem to have to offer are things along the lines of philosophical thought experiments on par with asking if I'm just a figure in your dream, or you in mine, did the idea of the chair exist in the ether before there were physical chairs, am I in the matrix, etc? But the most they can do with this is to deal with whether their assertions can be logically sound within the conceptual framework they establish, while having little relation to showing any real, provable effects on the universe human beings inhabit.

    Thanks for the discussion, both sides. It was an informative and civil exchange. I hope the theists gain a better understanding of the skeptic perspective and the skeptics can find useful points and better understanding of their own positions on top of highlighting what we see as weaknesses in the opposition tactics/arguments.

    TBB... I'm going to lift your explanation of natural/supernatural for a blog post. It seems to cut through a lot of the obfuscation on the issue I see a lot, and it's worth sharing.

    ReplyDelete
  8. TBB... I'm going to lift your explanation of natural/supernatural for a blog post. It seems to cut through a lot of the obfuscation on the issue I see a lot, and it's worth sharing.

    You're more than welcome to do so. I've written on the topic at greater length on my own blog as well:

    Naturalism and supernaturalism (6/14/09)
    Naturalism and Supernaturalism (4/5/08)
    What is Naturalism?

    ReplyDelete
  9. The theist commenting here needs to get outside of his Evangelical bubble so as not to embarass himself. We scientifically test the reactions of intelligent agents all the time, it is called survey research and it is widely used to assess the ability for communications to impact behavior based on historical normative scores within certain categories. Regression towards the mean anyone?

    No, what the theist wants to do is to create methodology out of special pleading and act like he is operating with humility and intellectual democracy.

    In short, the theist is an a liar who lacks a consistent ethic.

    ReplyDelete
  10. I just feel like tossing my two cents in here...

    Curiously, Ten Minas Ministries uses the example of stepping into oncoming traffic, and how we do not experiment this, yet we know we will be seriously injured if we do so. I am not certain what he means that this hasn’t been experimented.

    This sounds like a bastardization of the high school humanities class approach to David Hume. He wrote of the need for experience and empiricism to fully understand things and the idea that we can't know without evidence. This lead the smart-asses in my class at the time (:cough: this guy :cough:) to say, "So is he saying that he wouldn't know it's bad to get hit by a car without actually getting hit by a car?"

    That's not what Hume said at all. But you need an understanding that's more than an inch deep to get that.

    Using the scientific method, we look first at the hypothesis: “Nero was Emperor.” We look at the data—historical books, coins, statutes, opposing opinions, world results, etc. We can “experiment” (loosely) by comparing other leaders and other results and see the same consistent results. Other leaders have coins. Other leaders have statutes. Etc. And again, even under this method, new data could cause a change in hypothesis.

    This is technically accurate, but not really right. Modern historians use a system based off of the thinking of the scientific method but are incapable of using the scientific method. My favorite explanation for that is, "We can't replicate Napoleon in a lab."

    What you have actually described, though, is the process of historiography. From studying Roman records, histories, numismatics, and archaeology we know that an Emperor named Nero existed. There is absolutely no doubt in the mind of the historian. We don't have to replicate Nero in the lab any more than you have to replicate the meal you ate last night. The memory of it is there, the evidence of the preparation of your food can be found if you know where to look.

    Think about it this way. You tell me you had spaghetti for dinner last night. First, you are a primary source for your dinner, so I can say that you're a reliable eyewitness (as long as you're, y'know, sane). Second, I can go look in your sink and see if there is a pot with a noodle still stuck in the bottom or a plate that hasn't been scraped off and washed that still has some sauce on it. Then I can check your trash for evidence of an empty Ragu can, seasoning packets, or bits of the sort of vegetables that usually go in to home made sauce, depending on how you make spaghetti. I can then put all of my evidence together and say, "Yes, you ate spaghetti last night." I don't need a plate of spaghetti to know that.

    Also, for what it's worth, I wrote about this exact thing last night. It's the second half of the post, but it's in there...

    ReplyDelete
  11. Much depends on how narrowly or broadly we define the "scientific method". The arbitrary labels are irrelevant, though.

    All "good" epistemic methods share a common base: constructing the simplest (ontological) explanation to account for the (phenomenological) facts in evidence. In science, we're looking primarily for the derivation rules, so we take pains to introduce prior and subsequent states into evidence, and our "simplest account" consists of the natural laws that connect those states. In historiography and law, we are trying to determine the prior state (it is not already in evidence), so we introduce natural laws into evidence via more-or-less "expert" testimony of physical scientists (and a healthy dose of common sense).

    But the commonality is still there; the only difference is which part of the ontological explanation we're focusing on.

    As I noted earlier, the problem with the "philosophical method" is that Ken leaves the interesting part — coming up with the agreed-upon premises — undefined or at best vaguely-defined.

    ReplyDelete