INTRODUCTION
According to Greg Dawes, a theory that can be saved from refutation only by adding one or more auxiliary hypotheses for which there are no independent reasons to believe is true, does not possess the virtue of simplicity. Not only this, but a theory that is preserved from refutation in this manner also loses empirical content, and becomes less falsifiable. Unless such a theory exhibits a range of other virtues, it will be unworthy of our acceptance.[1]
For example, theists often employ defenses to try to mitigate the evidential force of various facts about evil in the world. Unlike a theodicy, a defense is merely true for all we know. There isn’t any independent reason to think that it is likely true. The more auxiliary hypotheses of this sort a theory exhibits, the greater will be the corresponding decrease in simplicity and empirical content. Unless such a theory exhibits a range of other virtues, this will significantly count against it.
[1] Greg Dawes, Theism and Explanation, 133, 135.
According to Greg Dawes, a theory that can be saved from refutation only by adding one or more auxiliary hypotheses for which there are no independent reasons to believe is true, does not possess the virtue of simplicity. Not only this, but a theory that is preserved from refutation in this manner also loses empirical content, and becomes less falsifiable. Unless such a theory exhibits a range of other virtues, it will be unworthy of our acceptance.[1]
For example, theists often employ defenses to try to mitigate the evidential force of various facts about evil in the world. Unlike a theodicy, a defense is merely true for all we know. There isn’t any independent reason to think that it is likely true. The more auxiliary hypotheses of this sort a theory exhibits, the greater will be the corresponding decrease in simplicity and empirical content. Unless such a theory exhibits a range of other virtues, this will significantly count against it.
[1] Greg Dawes, Theism and Explanation, 133, 135.
ASSESSMENT
While we should agree with Dawes that theistic explanations of evil that appeal to defenses lack the virtue of simplicity, on the other hand, we cannot agree with the wholly unsupported implication he draws from this one example. That implication is that, in general, theistic explanations are less simple than atheistic explanations. Not only is it the case that a theist has many theodicies at her disposal (which are auxiliary hypotheses with independent support), but the implication Dawes wants to make commits two highly significant fallacies of inductive logic. To demonstrate this, it will be helpful to take a brief excursus on different types of inductive arguments.
C-,P-, AND F-INDUCTIVE ARGUMENTS
Drawing on the work of Richard Swinburne, and Paul Draper, we can define and explain the differences between three types of inductive arguments. Let B be our background information or evidence; E be the evidence to be explained; and H be an explanatory hypothesis:
1) “C-inductive argument”-- An argument in which the premises confirm or add to the probability of the conclusion, i.e., P(H | E & B) > P(H | B).
2) “P-inductive argument”--An argument in which the premises make the conclusion probable, i.e., P(H | E & B) > 1/2.
3) “F-inductive argument”--An argument in which the evidence to be explained favors one explanatory hypothesis over one or more of its rivals, i.e., P(E | H1 & B) > P(E | H2 & B). Explanatory arguments are F-inductive arguments and have the following structure:
1. E is known to be true, i.e., Pr(E) is close to 1.
2. H1 is not intrinsically much more probable than H2, i.e., Pr(H1 | B) is not much more probable than Pr(H2 | B).
3. Pr(E | H2) > Pr(E | H1).
4. Other evidence held equal, H1 is probably false, i.e., Pr(H1 | B & E) < 0.5.
Good F-inductive arguments show that E is prima facie evidence — that is why (4) begins with the phrase, “Other evidence held equal.” They leave open the possibility that there may be other evidence which favors H1 over H2; indeed, they are compatible with the situation where the total evidence favors H1 over H2, or even that there is understated evidence that weakens, cancels, or outweighs the prima facie weight that E confers on H2.
F-inductive arguments are “stronger” than C-inductive arguments insofar as they show E not only adds to the probability of H2, but that E is more probable on the assumption that H2 is true than on the assumption that H1 is true. They are weaker than P-inductive arguments, however, because they don’t show that E is ultima facie evidence — they don’t show that E makes H2 probable.[1]
Keeping in mind the ways in which an F-inductive argument differs from a C- and P-inductive argument, we are now positioned to explain the two fallacies I have claimed Dawes has committed.
THEISM AND SIMPLICITY
The two fallacies that Dawe’s commits are the fallacy of understated evidence, and the violation of the rule of total evidence:
THE FALLACY OF UNDERSTATED EVIDENCE--Even if certain facts about evil constitute evidence in a good F-inductive argument for atheism, there could be more specific facts about evil in favor of theism that weaken, cancel, or outweigh this same evidence in favor of atheism. Thus, wholly apart from considerations of unsupported auxiliary hypotheses, theism may have a high degree of simplicity, even with respect to the evil in the world.
VIOLATION OF THE TOTAL EVIDENCE REQUIREMENT—Even if certain facts about evil constitute evidence in a good F-inductive argument for atheism, and even if there aren’t any items of understated evidence with respect to evil that would favor theism over naturalism, theism can still be much simpler than atheistic explanations all things considered. On our total evidence, theism will be simpler then atheism so long as theistic explanations possess other explanatory virtues better than atheistic explanations with respect to all the other available data we know.
DEMONSTRATING THE SIMPLICITY OF THEISM
Unlike Dawes, I have actually attempted to show that the general facts which are antecendently more probable on naturalism than theism (e.g. evil, pain and pleasure), have more specific facts which strongly favor theism over naturalism. Moreover, I have done the hard work of looking at all the other available data to show that the total evidence strongly favors theism over naturalism. The upshot for theistic explanations is that none of the aforementioned items of evidence are auxiliary hypotheses lacking independent support, and hence, demonstrate that theism far surpasses naturalism with respect to the virtue of simplicity as Dawes defines it (please see my post: Scholarly Naturalism/Paul Draper/The Fallacy of Understated Evidence).
CONCLUSION
Simple theories have few or zero auxiliary hypotheses which lack independent support. Showing that atheistic explanations are simpler than theistic explanations is a daunting task that we can forgive Dawes for not trying to undertake in any detailed measure. However, unless Dawes at least gestures in the direction of someone who has tried to demonstrate such a feat, his one illustration from the problem of evil will not accomplish this task either. On the contrary, as I have indicated above, I have tried to undertake at least the beginnings of such a task and my conclusion is that theism is far simpler than atheism.
[1] Jeffery Jay Lowder, http://www.patheos.com/blogs/secularoutpost/2014/03/21/f-inductive-arguments-a-new-type-of-inductive-argument/#sthash.Ft0prtn4.dpuf
While we should agree with Dawes that theistic explanations of evil that appeal to defenses lack the virtue of simplicity, on the other hand, we cannot agree with the wholly unsupported implication he draws from this one example. That implication is that, in general, theistic explanations are less simple than atheistic explanations. Not only is it the case that a theist has many theodicies at her disposal (which are auxiliary hypotheses with independent support), but the implication Dawes wants to make commits two highly significant fallacies of inductive logic. To demonstrate this, it will be helpful to take a brief excursus on different types of inductive arguments.
C-,P-, AND F-INDUCTIVE ARGUMENTS
Drawing on the work of Richard Swinburne, and Paul Draper, we can define and explain the differences between three types of inductive arguments. Let B be our background information or evidence; E be the evidence to be explained; and H be an explanatory hypothesis:
1) “C-inductive argument”-- An argument in which the premises confirm or add to the probability of the conclusion, i.e., P(H | E & B) > P(H | B).
2) “P-inductive argument”--An argument in which the premises make the conclusion probable, i.e., P(H | E & B) > 1/2.
3) “F-inductive argument”--An argument in which the evidence to be explained favors one explanatory hypothesis over one or more of its rivals, i.e., P(E | H1 & B) > P(E | H2 & B). Explanatory arguments are F-inductive arguments and have the following structure:
1. E is known to be true, i.e., Pr(E) is close to 1.
2. H1 is not intrinsically much more probable than H2, i.e., Pr(H1 | B) is not much more probable than Pr(H2 | B).
3. Pr(E | H2) > Pr(E | H1).
4. Other evidence held equal, H1 is probably false, i.e., Pr(H1 | B & E) < 0.5.
Good F-inductive arguments show that E is prima facie evidence — that is why (4) begins with the phrase, “Other evidence held equal.” They leave open the possibility that there may be other evidence which favors H1 over H2; indeed, they are compatible with the situation where the total evidence favors H1 over H2, or even that there is understated evidence that weakens, cancels, or outweighs the prima facie weight that E confers on H2.
F-inductive arguments are “stronger” than C-inductive arguments insofar as they show E not only adds to the probability of H2, but that E is more probable on the assumption that H2 is true than on the assumption that H1 is true. They are weaker than P-inductive arguments, however, because they don’t show that E is ultima facie evidence — they don’t show that E makes H2 probable.[1]
Keeping in mind the ways in which an F-inductive argument differs from a C- and P-inductive argument, we are now positioned to explain the two fallacies I have claimed Dawes has committed.
THEISM AND SIMPLICITY
The two fallacies that Dawe’s commits are the fallacy of understated evidence, and the violation of the rule of total evidence:
THE FALLACY OF UNDERSTATED EVIDENCE--Even if certain facts about evil constitute evidence in a good F-inductive argument for atheism, there could be more specific facts about evil in favor of theism that weaken, cancel, or outweigh this same evidence in favor of atheism. Thus, wholly apart from considerations of unsupported auxiliary hypotheses, theism may have a high degree of simplicity, even with respect to the evil in the world.
VIOLATION OF THE TOTAL EVIDENCE REQUIREMENT—Even if certain facts about evil constitute evidence in a good F-inductive argument for atheism, and even if there aren’t any items of understated evidence with respect to evil that would favor theism over naturalism, theism can still be much simpler than atheistic explanations all things considered. On our total evidence, theism will be simpler then atheism so long as theistic explanations possess other explanatory virtues better than atheistic explanations with respect to all the other available data we know.
DEMONSTRATING THE SIMPLICITY OF THEISM
Unlike Dawes, I have actually attempted to show that the general facts which are antecendently more probable on naturalism than theism (e.g. evil, pain and pleasure), have more specific facts which strongly favor theism over naturalism. Moreover, I have done the hard work of looking at all the other available data to show that the total evidence strongly favors theism over naturalism. The upshot for theistic explanations is that none of the aforementioned items of evidence are auxiliary hypotheses lacking independent support, and hence, demonstrate that theism far surpasses naturalism with respect to the virtue of simplicity as Dawes defines it (please see my post: Scholarly Naturalism/Paul Draper/The Fallacy of Understated Evidence).
CONCLUSION
Simple theories have few or zero auxiliary hypotheses which lack independent support. Showing that atheistic explanations are simpler than theistic explanations is a daunting task that we can forgive Dawes for not trying to undertake in any detailed measure. However, unless Dawes at least gestures in the direction of someone who has tried to demonstrate such a feat, his one illustration from the problem of evil will not accomplish this task either. On the contrary, as I have indicated above, I have tried to undertake at least the beginnings of such a task and my conclusion is that theism is far simpler than atheism.
[1] Jeffery Jay Lowder, http://www.patheos.com/blogs/secularoutpost/2014/03/21/f-inductive-arguments-a-new-type-of-inductive-argument/#sthash.Ft0prtn4.dpuf
RSS Feed
