The Evolutionary Argument Against Naturalism (Pt. 2)

[...continued from here.]

Last time, I introduced Alvin Plantinga's argument that believing in both evolution and naturalism results in a general belief-reliability crisis. Since I do hold both of these beliefs, I'm motivated to reflect on his argument and figure out whether I need to make an adjustment.

Bold vs. Cautious

I would characterize Plantinga's argument as bold because he reaches for the conclusion that actual, human believers in both evolution and naturalism have a general defeater for their beliefs. At least, they do after grasping his argument.

He is suggesting that I, Garren, have better reason to believe my beliefs are mostly false than to believe my beliefs are mostly true. I can't take this seriously. It's like telling Philosophy freshmen about Descartes' demon that systematically deceives one's senses, and expecting the students to really doubt everything.

But to have a defeater for [the belief that my cognitive faculties are reliable] it isn't necessary that I believe that in fact I have been created by a Cartesian demon or been captured by those Alpha-Centaurian superscientists. It suffices for me to have such a defeater if I have considered those scenarios, and the probability that one of those scenarios is true, is inscrutable for me. It suffices if I have considered those scenarios, and for all I know or believe one of them is true. In these cases too I have a reason for doubting, a reason for withholding my natural belief that my cognitive faculties are in fact reliable.¹

Plantinga's alternative is to accept another story that an external Agent wanted me to mostly believe true things, so He tinkered with evolution to give me reliable belief-forming mechanisms...except when it comes to the stunningly important belief that He exists.

At this point, I must admit that naturalistic evolution does have a major disadvantage: there is less room to simply make up convenient stories about it.

I propose a toned down, cautious version of the argument which doesn't deal with skeptical scenarios. There's no need to claim actual, human believers in evolution and naturalism are involved in a belief-destroying vortex. Instead, Plantinga could argue that theistic evolution provides a better explanation than naturalistic evolution when it comes to the unquestioned premise that we do have (more or less) reliable belief-forming mechanisms, i.e:

If true, theistic evolution would neatly explain why we have reliable belief-forming mechanisms.

If true, naturalistic evolution would provide a very poor explanation of why we have reliable belief-forming mechanisms.

By inference to the better explanation, theism beats out naturalism.

Can Naturalistic Evolution Offer A Decent Explanation?

Suppose evolution is true and naturalism provably can't ever provide a decent explanation for our (more or less) reliable beliefs. Naturalists might still resist the notion of divine intervention because of other considerations that count against theism or for naturalism, but I grant that the consideration we're considering would be a strong point against naturalism.

Now, the important question is whether Plantinga is offering a reason to think naturalism can't ever provide an explanation, or is he merely pointing out the current lack of such an explanation? It would be easy to dismiss him, if his argument were another "How could evolution design an eye?" or some similar structural mystery. These questions keep turning up reasonable answers.² But it's clear that the kind of "beliefs" he is concerned about are not just a matter of physical structure.

To paraphrase:

Naturalists these days all seem to be materialists. What sorts of things are beliefs under materialism? Neural events or states hooked into the overall operation of the brain. "So considered, beliefs will of course be able to enter the causal process that leads to behavior."³

But, any properly-so-called belief must also have the property of being associated with a proposition, e.g: that Frank Herbert wrote Dune. Otherwise, the neurophysiological event wouldn't really be about anything. "How does a neural event somehow get assigned a certain proposition as its content? It is hard to think of any scenarios that are as much as decently plausible."³ And, once assigned, the propositional content itself would be an irrelevant bystander to the physical operation of the brain.

So you see, the core of Plantinga's argument has to do with the nature of propositions. The naturalistic evolution of neurophysiological states (or events) which are generated in response to sensory input and which inform behavioral output goes completely unchallenged.

This makes the Evolutionary Argument Against Naturalism a fairly niche philosophical concern, perhaps one that is only conceivable for those who take proposition or property talk too seriously. Consider a maze-learning robot that adds information to its internal state, engages in some reasoning to infer when it has probably been dropped into a different point of a past maze, and makes goal-oriented choices based on its internal map. If the robot has reasoned that the goal is probably three feet forward, ninety degrees to the right and six inches forward again, then I would say the robot believes the goal is located there.

We could say the relevant bits of computer memory are "associated" with the propositional content that the goal is three feet forward and six inches to the right. Then we could worry how this separate content-bearing property can reach back into the robot's cybernetic brain and causally influence the bits and volts. Or, just maybe, we could question the philosopher's analysis that took the propositional content out of the realm of causally informed (and informing) bits and volts in the first place.

I admit unfamiliarity with the philosophy of propositions, but it seems plausible that they are just linguistic descriptions of possible world states. A belief may be associated with a world state without involving a linguistic description of that state (though some Postmodernists may disagree). The robot internally represents a possible state of the maze, then we use language to describe that state, and feel the description is both integral and external to the robot's electronic belief. I suspect something like this underlies Plantinga's dualistic intuition.

At any rate, his argument doesn't cause me a lot of concern about the rationality of believing in both evolution and naturalism. I'm much more inclined to think analytic philosophers sometimes generate their own problems, and this is one of those times.


1. Plantinga, A. (2002). Introduction. In Beilby, J. (Ed.), Naturalism defeated? (1-12). Ithica, New York: Cornell University Press. p 11.
2. Dawkins did a fun visual demonstration of eye evolution in Growing Up in the Universe.
3. Plantinga, A. (2002). Reply to Beilby's Cohorts. In Beilby, J. (Ed.), Naturalism defeated? (204-275). Ithica, New York: Cornell University Press. p. 212-213.

The Evolutionary Argument Against Naturalism (Pt. 1)

Since Alvin Plantinga has been in the news lately promoting his book Where the Conflict Really Lies: Science, Religion and Naturalism, I thought it would be a good time to review his famous argument against naturalism from the early 90s.¹

(Note: For his more recent moral argument against naturalism, see this post.)

Taking Evolution Away From Atheists and Beating Them With It

...that's the plan, anyway! Plantinga argues that — contrary to expectations — evolutionary theory is friendly to the belief that humanity was intentionally created by God in His image, and fatally undermines the belief that humanity evolved without divine assistance.

Basically:

• Natural selection only cares (so to speak) about survival.
• Survival is not significantly affected by the truth of beliefs.
• It's unlikely, therefore, that purely natural design would produce minds adept at forming true beliefs.

Our ability to form mostly true beliefs makes sense if God intervened in our development; but if anyone believes evolution occurred without such intervention, she is left with a story of how she probably does not have a mind adept at forming true beliefs. Evolution + Naturalism = Self Defeating Belief.

Natural selection only cares (so to speak) about survival.

Pretty much, yeah.

Survival is not significantly affected by the truth of beliefs.

Here's where things get interesting. You might think it's better — in terms of survival — to have mostly true beliefs than mostly false beliefs. Plantinga questions this assumption by examining five possibilities of how beliefs and behavior relate to each other.

(Note: Early on, he focused on Possibility #5; over time, his focus has shifted to Possibility #1 as the core challenge to naturalistic evolution.)

Possibility #1 — Epiphenomenal Beliefs

He starts by drawing a distinction between beliefs as the neural structures that combine with desires to produce behavior and beliefs as carriers-of-propositional-content. The latter definition of belief is the sort that is true or false and it "can't be a matter of definition that there are neural structures or processes displaying both propositional content and causal efficacy with respect to behavior".²

Neural structures which guide behavior.
vs.
Mental stances which are true or false.

Natural selection could have designed neural structures to work in ways that produce survival-enhancing behavior, but mental stances are just extra. "[Mental stances] are not causally connected with behavior, then they would be, so to speak, invisible to evolution; and then the fact that they arose during the evolutionary history of these beings would confer no probability on the idea that they are mostly true, or mostly nearly true, rather than wildly false."²

To put it simply: true/false mental stances run free from the neural structures which affect behavior.

Possibility #2 — Same Thing

...except here Plantinga highlights the possibility that mental stances are caused by behavior or that both mental-stances and behavior are caused by a third thing. So there is a connection, but it's still invisible to natural selection.

Possibility #3 — The Form (Not Content) Affects Behavior

I read a poem very loudly, so loudly as to break a glass; the sounds I utter have meaning, but their meaning is causally irrelevant to the breaking of the glass. In the same way it might be that these creatures' beliefs have causal efficacy, but not by way of the content of those beliefs.³

Is this a real party trick he does?

Possibility #4 — Belief Content Affects Behavior, Negatively

Behavior-affecting belief content might be like sickle-cell anemia: not helpful, but genetically attached to other traits which are helpful.

Possibility #5 — Belief Content Positively Affects Behavior

The true/false mental stances we take actually do help us survive. But...true beliefs don't necessarily help more than false beliefs!

You see, beliefs have to be combined with desires to produce behavior. As Plantinga points out, "there will be any number of different patterns of belief and desire that would issue in the same action; and among those there will be many in which the beliefs are wildly false."⁴

True Belief + Desire A -> The Helpful Action
False Belief 1 + Desire B -> The Helpful Action
False Belief 2 + Desire C -> The Helpful Action
False Belief 3 + Desire D -> The Helpful Action
[...and so on]

Since natural selection only cares about the helpful action, it's more likely that one of the many possible false beliefs will be paired up with a matching desire, than that the single true belief will be paired up with its matching desire. Plantinga gives the example of a pre-historic man, Paul, who has the adaptive behavior of running away from tigers, but not because of the belief you might expect:

Perhaps Paul very much likes the idea of being eaten, but whenever he sees a tiger, always runs off looking for a better prospect, because he thinks it unlikely that the tiger he sees will eat him. This will get his body parts in the right place so far as survival is concerned, without involving much by way of true belief.

[...] Or perhaps he thinks the tiger is a large, friendly, cuddly pussycat and wants to pet it; but he also believes that the best way to pet it is to run away from it."⁴

He goes on to give other possibilities of false beliefs being matched up with desires that result in Paul running away from tigers. The point is: natural selection has no reason (so to speak) to favor one scenario over another, so long as Paul's body survives to reproduce.

It's unlikely, therefore, that purely natural design would produce minds adept at forming true beliefs.

If this conclusion is successful, then a person who believes in naturalism + evolution has a defeater for the belief that her belief-forming faculties are reliable. This, in turn, constitutes a defeater for all of her beliefs...including the belief that naturalism is true.

Meanwhile, a person who believes in a creator God + evolution isn't stuck with an across-the-board defeater for his beliefs. In other words, evolution mixed with this kind of theism is stable, but evolution mixed with naturalism catastrophically destructs.

[continued here...]



1. See Chapter 12 of his book Warrant and Proper Function. For a shorter and more recent restatement, see the Introduction of Naturalism Defeated? which is a collection of essays on this very topic, edited by James Beilby.
2. Plantinga, P. (1993). Warrant and proper function. New York, New York: Oxford University Press. p. 223. 
3. Ibid. p. 224. 
4. Ibid. p. 225.

Lingo: Poisoning the Well

I am at war with him; but there is such a thing as legitimate warfare: war has its laws; there are things which may fairly be done, and things which may not be done. I say it with shame and with stern sorrow;—he has attempted a great transgression; he has attempted (as I may call it) to poison the wells.¹

To poison the well (or wells) is to use some preliminary tactic which has the effect of greatly impeding any fair and reasonable discussion. As you can see above, the term comes from an analogy to poisoning a city before attempting to take it by force. Poison may succeed where force of argument would otherwise fail.

In Federalist No. 83, Alexander Hamilton argued against making trial by jury a Constitutional requirement in civil cases. In addition to his actual arguments, he wrote: "It is conceded by all reasonable men, that it ought not to obtain in all cases." This is poisoning the well since anyone who tries to argue for trial by jury for all civil cases has already been labeled "unreasonable."

I was inspired to write this post because I ran into a rather extreme case of well poisoning in Francis J. Beckwith and Gregory Koukl's book Relativism: Feet Firmly Planted in Mid-Air which reminded me of William Lane Craig's approach in Reasonable Faith: Christian Truth and Apologetics. Early on, both books tell horror stories about what life would be like if the positions they defend were rejected. This strongly influences readers to accept the authors' arguments uncritically and reject opposing arguments offhandedly.

Craig on the Horrors of Non-Christianity

Before examining "the question of God's existence," Craig explores "the disastrous consequences for human existence, society, and culture if Christianity should be false."² He spends page after depressing page claiming that our lives are totally without significance unless we will live forever and there is a God. Then Craig tells a story of Nazi doctors performing vivisection on pregnant women, saying this is consistent with atheism and a story about a man giving his life to save others is inconsistent with atheism.³

The choice is clear: cruel, pointless existence if there is no God vs. fulfilling, meaningful life if "biblical Christianity" is true.

(Yes, I did happen to notice that he left out many alternatives besides atheism and his brand of Christianity.)

Beckwith and Koukl on the Horrors of Moral Relativism

Chapter Two of their book is titled "What Is Moral Relativism?" But before trying to define the position they're attacking, the authors explain in Chapter One that moral relativism is about living for personal pleasure without any concern for how others are affected. To make their point, they tell a story about a group of nurses lounging in their break room, "smoking and drinking coffee," while coldly choosing to let a premature child die on the metal counter rather than try to save it or comfort it. When one nurse arrived and tried to hold it, another snatched it away and put it live into a jar of formaldehyde.⁴

This, they claim, is what moral relativism looks like. In fact, it's hardly necessary to argue against it at all!

...

I try not to use such tactics even if it would help me win rhetorically. But sometimes I wonder if choosing not to poison wells in debates is as quaint and self-restricting as many might view the choice to not poison the wells of a city before sending soldiers out to die on its walls. Am I more willing to loose respectably than win underhandedly? Maybe so.


1. From John Henry Newman's Apologia Pro Vita Sua.
2. Craig, W.L. (2008). Reasonable faith: Christian truth and apologetics (3rd ed.). Wheaton, IL: Crossway Books. p. 65
3. Ibid. p. 80-82
4. Beckwith, F.J. and Koukl, G. (1998). Relativism: Feet firmly planted in mid-air. Grand rapids, MI: Baker Books. p. 21

Reading the ACLU Policy Guide (Pt. 8)

Series explanation and overview here.

Note: These are my summaries of the 1995 version of the guide, not the policies themselves.

Loyalty and Security

Policy 103 — Clear and Present Danger Test

Personal opinions are constitutionally protected unless they qualify as creating a “clear and present danger” by either being “an integral part of conduct violating a valid law” or “a direct incitement to specific and immediate violation of law” or they “threaten a danger of unlawful acts so great and so immediate that time is lacking for answer, or if need be, for other protective measures against the threats and acts.”

Protected speech includes speech against democracy, so long as it does not constitute a clear and present danger. I take this to mean that speech against free speech rights is still protected by free speech rights.

Policy 104 — ACLU and Totalitarianism

The ACLU is opposed to “any governmental or economic system which denies fundamental civil liberties and human rights.” (This is consistent with the ACLU supporting the right of other groups and individuals to express support for totalitarian systems.)

Policy 105 — Smith Act and Subversive Activities Control Act of 1950

This policy actually concerns three federal acts:

• The Alien Registration Act of 1940 (aka the Smith Act)
• The Internal Security Act of 1950 (aka the Subversive Activities Control Act or the McCarran Act)
• The Communist Control Act of 1954

Besides requiring legal, adult aliens to register with the government, the Smith Act made it a crime to advocate for the violent overthrow of the US government. At first glance, this might not sound like such a bad thing to criminalize, but citizens who don't have the slightest chance of putting such ideas into action could have been imprisoned for twenty years for expressing the opinion. Furthermore, entire organizations could have been deemed in violation and all members and tangential supporters would be made criminals, without even expressing the condemned opinion personally.

Further acts specifically hounded anyone with communist associations, or who supported broadly communist ideas whether those ideas were related to violent political change or not.

The ACLU's position is that "there should be no governmental restriction on advocacy of any sort, unless the adovcacy [sic] shall cause, in the existing circumstances of its utterances, a clear and present danger of illegal action." Merely increasing the probability that listeners may choose to commit a crime later is insufficient. Regarding organizational guilt: "Guilt is personal; it may not be attributed by association." The ACLU also opposed the McCarran Act because it required members of communist organizations to report themselves, a violation of the Fifth Amendment.

Policy 106 —Wartime Sedition Act

I'm a little confused about this policy. It references the "Wartime Sedition Act of 1917", but I'm finding information on the Espionage Act of 1917 and the Sedition Act of 1918 (a set of amendments to the first). I agree that the Sedition Act could have been "used against anyone, at virtually any time, to suppress criticism of the government in the name of national security," but I'm seeing multiple sources claiming those parts were repealed a long time ago. Recent ACLU concerns appear to be about last year's WikiLeaks fiasco, as it relates to the original espionage sections.

Policy 107 — Emergency Measures in Peacetime

The ACLU is opposed to indefinitely continuing civil rights-related measures put into play during "genuine war emergencies." This is applied to the Cold War and — I suspect — would apply to our perpetual War on Terror.

Policy 108 — House Internal Security Committee

This policy explains the ACLU's opposition to the House Un-American Activities Committee (later renamed to Internal Security Committee). Instead of doing much in the way of starting legislation for the House, this Committee made a big public show of investigating citizens for expressing "un-American" ideas by a standard "so vague that a citizen cannot know whether the citizen's political activity runs afoul of the Committee's private definition of Americanism." These non-trial trials took away due process rights and chilled free speech. The Committee was dissolved by the mid 70s.

Policy 109 — State and Local Legislation

The ACLU opposes non-federal laws "dealing with advocacy of political doctrine" because there is (or was) already federal legislation in the area, and adding further restrictions just makes the civil liberties situation worse.

Policy 110 — Federal Employee Security

While the ACLU recognizes that some federal jobs directly related to national security warrant security screening, they oppose extending such screening to the many federal "non-sensitive" jobs which only require candidate fitness for the work.

The ACLU specifically opposes investigation into candidates' sexual orientation as a security matter. Apparently, it was claimed that homosexuals were a security risk because they could be blackmailed into betraying their country out of fear of expose. (Notice anything self-fulfilling here? Such investigations create or greatly increase the very motive they seek to exclude.)

Due process must be given to candidates denied positions for security concerns.

Policy 111 — Private Employment Security

The ACLU opposes the practice of any private employers investigating the national security risk of employees or requiring loyalty oaths. If a private employer is contracting with the government in a sensitive area, then the government itself should be the party to conduct security investigations of employees involved in such work.

Policy 112— Professional Associations' Membership Qualifications

Mere association with other organizations should not be grounds for removing a member from a professional association (particularly bar associations). It must be shown that a lawyer, for example, is actually not doing his or her duty as a legal professional because of "external obedience" to another organization.

However, the ACLU does recognize some latitude for not admitting a new member to a professional organization if there is thought to be a high probability he or she will not be able to conform to professional duties. This risk still needs to be evaluated on a person by person basis, not as a sweeping prejudice against the membership of other organizations.

Policy 113 — Federal Benefits and Loyalty Tests

"Loyalty oaths or disclaimers of membership in certain organizations violate the First Amendment rights of freedom of belief and association, and may never be required for participation in government-funded programs such as public housing or subsidies, welfare benefits, veteran's benefits, Social Security or Medicate."

Yes, that's the whole policy.

Policy 114 — Military Discharges

Members of the military should be treated as other federal employees, i.e. screening for security should only be done if that individual's job directly relates to national security.

Discharges must be based on job performance, not merely for exercising constitutional rights before or during service.

Policy 115 — Loyalty Oaths

Swearing non-membership in certain organizations has been required "for employment of government workers and teachers, for students seeking government aid, and for Social Security and Medicare recipients and the like." The ACLU opposes such oaths because they suppress free speech and association rights. Plus, they penalize any citizens with a conscientious objection to swearing these kind of oaths, whether they have had the associations in question or not.

Policy 116 — Governmental Surveillance

Police infiltration of organizations is contrary to the Fourth Amendment's protection against "unreasonable searches and seizures," i.e. those not carried out through the use of a targeted warrant issued on probable cause. Otherwise, "[e]verything that is said and everything that is done over an unlimited period of time comes into the hands of the government, no matter how private, how unconnected with a legitimate state interest."

Perhaps this would be less of an issue if there weren't a history of the government collecting files on individuals, and this to their detriment either by denying them government employment or by publicly condemning them without due process in Congressional hearings.

The ACLU does recognize the proper use of "informers" placed or recruited from organizations, if there is probable cause shown that the organization is involved — or is planning to be involved — in "serious criminal acts." A judge must issue a warrant specifying which part of the organization is to be surveilled and for how long. Warrants should be renewed by the original judge and a limit should be set on the number of renewals.

This policy also speaks against using the military to spy on citizens, against building files on citizens merely for protesting government actions, and against keeping attendance lists of lawful gatherings.

Policy 117 — Controlling the Intelligence Agencies

Bill of Rights violations on the excuse of "national security" need to end. The ACLU has a list of specific measures to remedy the situation. Some highlights:

• Implement only three categories of classified information. (1) Details of defense tech which would help other nations. (2) Tactical military details during declared war. (3) Defensive contingency plans.
• Explicitly state that any information about the US government engaging in illegal behavior is ipso facto declassified.
• Allow Congress to "unilaterally" release Executive branch information.
• Make intelligence agency budgets public.
• Create detailed charters for government agencies, and prohibit any agency activity beyond these limits.
• Rename the CIA to the FIA: Foreign Intelligence Agency. Keep it out of the domestic intelligence business entirely.
• Prohibit all CIA operations intended to secretly manipulate (or just plain overthrow) foreign governments.
• Prohibit the NSA from monitoring US-to-foreign communications.
• Destroy all current files kept on citizens for exercising their First Amendment rights.

Policy 118 — Secret Government Aid to Private Agencies

I'm curious what circumstances prompted this 1967 policy. Apparently a government agency was secretly funding a citizen advocacy group...or something like that.  This endangers free discussion "by the clandestine introduction of ulterior motives of government policy into a supposedly open debate."

Policy 119 — Prior Restraint in National Security Situations

"The true test of our devotion to that principle [i.e. freedom of thought and expression] comes in times of stress or alarm, when those who would suppress the interchange of ideas can appeal for supposed justification to some imminent menace threatening the public welfare."

Citing United States v. The Progressive and the Pentagon Papers as instances when prior restraint "safeguarded nothing more than governmental overreaction, embarrassment, and the desire for secrecy," the ACLU believes prior restraint cannot be tolerated at the very times it is most critical that citizens have access to information.