Tuesday, November 14, 2006

Oppy on atheism

Among secular philosophers, Graham Oppy is one of the most astute. His recent book on Arguing About Gods (Cambridge 2006) is a carefully reasoned and generally fair-mined analysis many theistic proofs.
It is, of course, slanted to atheism, so I don’t agree with him at many points. But it’s one of the best entries in contemporary atheology.

In the same year the book as published, he also published an essay in which he reviews and updates Bertrand Russell’s celebrated essay.


In a way, Dawkins and Harris are easy targets. They are worth targeting due to their media penetration.
Daniel Dennett is more sophisticated, but in defending his secularism and attacking the faith, he substitutes science fiction for hard science.

In fairness to atheism, let’s move up the food chain and see how a more responsible and competent unbeliever lays out a case for atheism. As is my wont, I’ll offer a running commentary on his major arguments.

“Apart from identifying failings in some theistic arguments, Russell also makes some observations about the true wellsprings of belief in God. On his view, the main reason why people believe in God is simply because this belief is inculcated in them from early infancy. Furthermore, according to Russell, the other important motives for belief in God are fear of the unknown (including, in particular, fear of death), and a desire for safety (including, especially, a desire for a big brother who will look after your interests and guarantee the satisfaction of your most deeply held wants). While I agree with Russell that nonbelievers are committed to merely causal stories about the prevalence and strength of belief in God, I think that any such story will be much more complicated than Russell (or Freud, Marx, Engels, or Durkheim) allows. There has been some interesting recent work in this area--see, for example, Lawson and McCauley (1990), and Atran (2002)--but I do not think that we are yet close to a fully satisfying account.”

This is a candid and quite welcome admission.

“While I think that there is some truth in each of these observations, there is also a considerable amount of overstatement. For example, that organized Christian churches have often been enemies of moral progress is uncontroversial, but it is unclear that they have been worse than other religious and social institutions in this regard. As we all know, it is very easy to mistake mere correlations for causal relations; we should not be too quick to suppose that widespread correlation of Christian belief with ignorance, misery, and suffering is evidence that Christian belief has been a major causal agent in the production of that ignorance, misery, and suffering.”

Another candid and long overdue admission.

“Reality is described by science; there are no metaphysical spooks hiding behind the scenes.”
There are a number of problems with this claim:

i) How does he deal with varieties of scientific antirealism?
ii) What is the relation between appearance and reality? Is the world the way we perceive it?
iii) What is his theory of abstract objects?
iv) Does evolutionary psychology cut the nerve of rationality?
v) His claim is self-refuting, for science must postulate a number of metascientific assumptions. As Del Ratzsch has pointed out:


Empirical data by themselves can neither generate, identify, drive us to, nor conclusively confirm some single theory from among all possible competitors. Underdetermination, then, presents us with a forced choice between empirical purity (at a cost of the theoretical) and theoretical legitimacy (at the cost of empirical rigor). Thus, when we do single out a particular theory, whatever selection process we employ will of necessity involve 'extra-empirical' factors - factors beyond the purely empirical.[5] A genuine realist science thus cannot survive on just empirical observation and logic, but requires richer conceptual resources and is of necessity integrally embedded in a deeper conceptual matrix from which, I shall argue, it cannot be cleanly detached intact. What are those resources?

Among the operative presuppositions upon which science unavoidably depends, many are conceptually unproblematic, and various of them are so familiar as to ordinarily escape special attention. The following list is not presented as exhaustive.

1. General metaphysics.

a. Intelligibility of nature. For any robust idea of science itself to make sense, the cosmos must be assumed to be to some degree understandable by the minds doing the science. Historically science has assumed that nature embodies an inherent intelligibility - that properly conceived, nature's mechanisms and structures make sense.

b. Basic character of nature. Relevant presuppositions concerning nature include e.g.: that there really is a real world, that it is largely 'out there', that it exists largely independent of us, that it persists, that change is real, that events in that world have effects, etc. Although change is real, there are also stabilities and uniformities extending through time and space. There is a unity to reality - we live in a cosmos, rather than a random and arbitrary patchwork of different jurisdictions. And whatever their ultimate character, there are 'laws' of nature, that seem to inhabit a region somewhere between logical necessity and accidental generalization. The laws of nature, and nature itself, manifest a logical contingency, but an ordered - not a chaotic - contingency. There are other sorts of presuppositions as well, which I will not pursue here.[6]



vi) I’ll introduce another difficulty below.

“If there is anything contingent in the world, then there is brute--i.e., inexplicable--contingency in the world.”

On the face of it, this claim is straightforwardly fallacious. How does he reason for “anything contingent” in the world to brute factuality?

In principle, you could reason from brute factuality to pure contingency, but you can’t reason from local (or global) instances of contingency to brute factuality.

The existence of contingent facts is not itself an evidence of brute factuality, for a contingent mode of existence may well be ontologically dependent on something necessary.

That’s what makes it contingent. It’s contingent on something else.

Of course, Oppy can try to deny this connection, but that will require an argument. He cannot infer brute factuality from contingency without a supporting argument.

“Hence, if there is anything contingent in the world, then there are things--events, facts--that simply have no explanation.”
Several problems:

i) Even if valid, this piggybacks on a false premise (see above).
ii) He is admitting that, at rock bottom, reality is irrational. What does that say about the secular worldview?
iii) In addition, what does it say about his commitment to scientific explanation?

“In particular, then, there is no justification for supposing that belief in God is justified simply because the truth of that belief would account for otherwise inexplicable contingency in the world. If we are puzzled by why the world is one way rather than some other way that it might have been, our puzzlement cannot be removed by supposing that the world is the way it is because God chose to make it that way. If we are worried by unexplained contingency, we shall want to know why God chose to make the world that way: postulating God does not remove the unexplained contingency, but it does land us with a whole new raft of explanatory burdens and commitments. This is not progress.”

A couple of problems:

i) To say that positing an agent fails to provide a complete explanation is far from saying that such a posit an agent is devoid of explanatory value.

If a homicide detective fingers the butler as the killer, this has a lot more explanatory value than treating the murder as a brute fact. It’s a sufficient explanation to account for the immediate circumstances.

In principle, this raises a whole new set of questions. How do we explain the existence of the butler? When was he born? Where was he born? Where did he go to school? Who were his parents? Who is his best friend? Does he have a girlfriend? Does he prefer cheddar cheese to Swiss cheese? What were his motives?

But it isn’t necessary to answer all of these incidental questions for the “butler-hypothesis” to be an adequate explanation of the crime scene.

ii) Oppy is assuming that we don’t know why God chose this world over another. But suppose we do have a source of information in answer to that question?

Oppy doesn’t even consider the possibility that God may have revealed his reasons to one degree or another.

“If we suppose that there is a concept of cause that has proper application to our world, then there are events and occurrences that simply lack causes, including, in particular, events and occurrences where entities come into existence and processes commence.”

Assuming, for the sake of argument, that this is true, where does it leave his commitment to scientific explanation as our only window onto reality?

“Consequently, I do not think that the thought that God might be the initial cause for what would otherwise be uncaused initial events and occurrences justifies belief in God--for God's actions and decisions are also events and occurrences that, on the intuition in question, stand in need of causes.”

This claim is simply inept. According to classical Christian theism, God is a timeless agent. Hence, his actions and decisions are not another set of events or occurrences, but rather, a timeless state of being instead of a temporal process of becoming.
To be sure, there are Christians who regard God as in some sense temporal. Process theology is an extreme example.
But Oppy also needs to adapt his objection to the traditional conception of God’s relation to time and eternity.

“If we suppose that there is some intelligible sense to an objective conception of ‘perfection,’ then it is highly plausible that, if there were a perfect being, it would simply be unable to create an imperfect universe: i.e., a universe whose history was less than optimal on any dimension of evaluation.”

i) This assumes, without benefit of argument, that there is such a thing as an optimal world. But why make that assumption?
Why not assume that different possible worlds may exemplify alternative or incommensurable goods?
So there may be no such thing as a best possible world.
ii) Apropos (i), he is also assuming, without benefit of argument, that a suboptimal world is equivalent to an imperfect world. But what does that mean?

If one possible world exemplifies lesser goods, while another possible world exemplifies greater goods, how is the lesser world imperfect? One possible world may be better than another, but as long as a possible world is good, how is such a world imperfect?

“In particular, it is highly plausible that if there were a perfect being, it would simply be unable to create a universe in which there are departures from moral perfection.”

A departure from moral perfection would not involve a choice between lesser and greater goods, but a choice between good and evil, or the lesser of two evils.

“But it is very plausible--if not utterly obvious--that the universe that we inhabit is populated with events and occurrences that do mark departures from moral perfection. So it is highly plausible to suppose that, on our initial assumption about objective perfection, there is no perfect being.”

A couple of problems:

i) He is equating moral perfection with a static state of affairs. But what if the present imperfection of the world is a means to a greater good or second-order good?
ii) Where does he get his ideal standard over against which he judges our world to be morally imperfect? If this world is the only world there is, then how is he in any objective position to contrast the actual world with a more perfect or optimal state of affairs?

“While this claim hardly cuts against the claim that the Christian God exists--since the Christian God is continuous with the God of the Old Testament, and it is hardly contentious that the God of the Old Testament is less than morally perfect--it is equally clear that this claim does cut against the views of many contemporary Christian philosophers of religion.”

Is it “hardly contentious” to an observant Jew that the God of the OT is less than morally perfect?
The Jewish race is not extinct. There are contemporary Jewish intellectuals to believe in the OT and look to the OT for moral guidance.

“If we suppose that claims about human free actions are intelligible, then it seems to me that it is a mistake to think that human beings have what philosophers call 'libertarian' freedom, as opposed to 'compatibilist' freedom. To act freely is simply to act on one's normally acquired beliefs and desires in the absence of certain kinds of constraints, and there is no inconsistency in the thought that actions that possess this kind of freedom have physical causes. On the libertarian conception of freedom, one acts freely only if, in the very circumstances in which one acted, it was within one's power to do otherwise--which is incompatible with efficient causation of action. But as I see it, the only alternative to efficient causation is absence of causation; and, if one's actions are only 'free' because they have no causes, then this is not a kind of 'freedom' worth wanting. Among the consequences of this view, two are particularly important. First, it is a mistake to suppose that the free choices of a supernatural agent might be 'ultimate' explainers: for if 'freedom' is libertarian, then what is appealed to ultimately has no explanation; and if 'freedom' is compatibilist, then free choices are no less in need of explanation than any other kinds of events. Second, given that freedom is 'compatibilist,' it is very hard to see how the presence of evil in the world might be explained in terms of the value that freedom possesses; for there surely are possible worlds in which agents always freely choose the good, and it is very hard to see why a perfect creator either could not or would not make one of those worlds if it made any world at all.”

I agree.

“If we suppose that our ordinary mentalistic vocabulary should be given a realistic construal, then I take it that our 'mental' states are nothing other than certain kinds of states of our brains.”

Are they? Suppose we apply his statement to the statement itself. Is his statement true? Can I entertain a true belief about his statement?

But if my belief is a physical state, then to what does it correspond? The physical state of my brain isn’t like the letters and words composing his sentence. It isn’t descriptive or symbolic of his sentence, or the propositional content thereof.

So, in the absence of any resemblance between the two, how does my brain state represent or refer to his sentence about brain states? What accounts for the alethic relation between a brain state and a sentence about a brain state?

“I do not deny that there are imperfections in our current understanding of consciousness; but I do not see how the postulation of spooky mental stuff promises to give any additional insight.”

i) One wonders how much reading he’s done in the philosophy of mind.
ii) Notice, too, his appeal to the argumentum ad ignorantiam. This is a godless-of-the-gaps appeal. He fills in the lacunae of his physicalism by a leap of blind faith in the inexorable progress of science.

“Moreover, I take it that the welter of information that we possess concerning neural deficits, and the nature of various kinds of physical impacts on our 'minds,' provides very strong reason for denying that we are essentially nonphysical spooks who are only contingently wired up to our bodies.”

i) Dualism is well-aware of the very same data. No surprises here.
ii) In theistic dualism, the body, via the senses, is a source of information by which the soul acquires some of its knowledge of the external world.

“Of course, there is some alleged data that supports the distinct existences hypothesis; but I take it that, after due consideration, we should find that reports of out-of-body experiences, astral travel, and the like are not reflections of how the world really is.”

Has he given “due consideration” to the standard literature? Cf.:

R.W.K. Paterson, Philosophy and the Belief in a Life After Death (New York: Saint Martin’s Press, 1995.

Michael Stoeber and Hugo Meynell, Eds., Critical Reflections on the Paranormal (New York: State University of New York Press, 1996).

James Houran and Rense Lange, Eds., Hauntings and Poltergeists: Multidisciplinary Perspectives (McFarland and Company, Inc., 2001).

David Lester, Is there Life After Death? (McFarland and Company, Inc., 2005).

David Fontana, Is There An Afterlife: A Comprehensive Overview of the Evidence (O Books, 2005).

Lance Storm and Michael A. Thalbourne, eds., The Survival of Human Consciousness: Essays on the Possibilities of Life After Death, ed (McFarland and Company, Inc., 2006).

“If asked to engage in fundamental metaphysical speculation about the nature of our world, I would give greatest credence to a kind of supervenient naturalism. Alas, it is no easy matter to give an exact characterization of supervenient naturalism.”

So supervenient naturalism is not an actual alternative, but a secular faith-commitment. A voucher in place of a detailed, working model.

“If asked to provide a broad outline of the history of the universe since the Big Bang, I would--to the best of my ability--outline the history that is delivered to us by our best sciences. In particular, I would insist that life appeared on Earth some billions of years ago, and that human beings are directly descended from those ancestral life forms.”

i) I disagree with him, but I’ll ignore than for now.
ii) How is his faith in science consistent with his denial of the principle of sufficient reason? If, by his own admission, the world is ultimately unintelligible, given the occurrence of surd events, then what happens to the principle of induction?

“Moreover, I would insist that there is no need to appeal to the hypothesis of intelligent design to explain any of the features that our universe has possessed over the course of its history.”

ID theorists would demur, but I’ll delegate that debate to the respective parties.

“Furthermore, I would say that any impulse to postulate intelligent designers to explain structural or organizational features of the universe should be matched by an impulse to postulate further intelligent designers for those intelligent designers--to explain the structural or organizational features of the beliefs, desires, and intentions of the initial intelligent designers.”

i) This is inept, for it compares the incomparable. It assumes tht causes resemble their effects.
ii) It also fails to distinguish between abstract complexity (e.g. the Mandelbrot set) and cumulative, composite complexity.

“If asked to provide a broad outline of recent human history--say, the last 10,000 years--I would provide an outline in which there is no mention of the actions of supernatural agents, and no mention of the occurrence of miraculous events. Of course, I might mention well-known reports of the actions of supernatural agents, and well-known reports of the occurrence of miraculous events; but I would insist that all such reports are simply mistaken.”

“All such reports are simply mistaken.” That’s a sorry excuse for an argument.

“In particular, I would insist that none of the religions that have a place in recent human history have supernatural origins; and I would insist that there have been no miraculous events that provide support for particular religious beliefs.”

This assumes that most religions lay claim to a supernatural origin and miraculous attestation.

“All of the religions that have a place in recent human history--and all of the writings that are associated with those religions--can be happily accommodated within the supervenient naturalism outlined earlier.”

How “can they be happily accommodated within the supervenient naturalism” when, by his own admission, “it is no easy matter to give an exact characterization of supervenient naturalism.”

Oppy is resorting to secular fideism to overthrow religious claims.

“Moreover, in principle, there are purely naturalistic explanations for all religions, religious beliefs, religious writings, religious experiences (and reports thereof), and so forth--though, in practice, we may not always be able to fix on those explanations, owing to lack of data and theoretical sophistication.”

Yet another appeal to secular fideism. We lack the supporting evidence, but we’ll keep on believing in “purely naturalistic explanations for all religions,” &c. no matter.

“If asked to say how much of the Christian Gospels is historical truth, I would say that I don't know. It might be that they are entirely works of the imagination; or it might be that there was a historical figure who became the focus for what are, in significant part, works of the imagination. I am not impressed by the claim that if we suppose that the Gospels are largely works of the imagination, then we shall need to suppose that the rest of ancient history is so as well. For we have the best of reasons for thinking that much in the Gospels does spring from the imagination--namely, the fantastic nature of the events that are described therein--but we have no such reason for thinking that everything else that is recorded in ancient history must also spring from that source.”

The Gospels are “imaginary” because they are “fantastic.” This objection assumes that miracles never happen. Where’s the supporting argument?

In his section (7.4) on the argument from miracles (Arguing About Gods, 376-81), he admits that Humean objections are a failure.

“Moreover, while there are ancient histories that bear marks of methodological sophistication, including explicit discussion of the principles that apply when weighing the reliability of sources and the like, the Gospels are plainly not amongst those histories.”

This depends on the distance between the timing of the event and the timing of the record. Where contemporaries are reporting on contemporaneous events, we wouldn’t expect the same methodological caveats as apply to a historical reconstruction of events from the distant past.

“In particular, if there are basic truths about values, then those truths are necessary, and hence not capable of having a 'grounding' in anything.”

A couple of elementary problems:

i) There’s no reason why moral norms must be necessary truths. Especially where Christian ethics is concerned, moral norms are, in some measure, preadapted to natural kinds. To the way in which God has constituted the human race. So moral norms could well be contingent facts.
ii) Even if we assume that moral norms are necessary truths, does naturalism enjoy the metaphysical resources to underwrite necessary truths? What is the status of abstract objects (of which necessary truths are a subset) in Oppy’s ontology?

“Many Christians suppose that there can be no virtuous behavior in the absence of Christian belief.”

That’s a simplistic statement of the contention. The real position is more like this:

i) The primary question is not whether unbelievers can be virtuous, or whether, in fact, they are virtuous, but why they should be virtuous, given their outlook.

It’s a de jure question rather than a de facto question.

ii) Due to common grace, unbelievers are often better than their unbelieving creed.
iii) But the more consistently and self-consciously secular their outlook, the less virtuous they will be, if they are free to act with impunity.


  1. Very interesting post, Steve.

    I thought the bit near the beginning about Del Ratzsh with regards to rank empiricism to be of note. If we deflate rank empiricism for the unbeliever, does the believer suggest that God is knowable through natural means? In other words, suppose the believer is convinced by the Christian that the senses can't account for everything. Then what? I think an unbeliever might be exasperated at the thought of trying to obtain contact with a timeless, immaterial being without use of the senses.

    Also, you make mention of dualism. You've touched on this in some of your posts in the past...does your position adhere to the Cartesian version, or is there more to it than that?

  2. Steve,

    “Reality is described by science; there are no metaphysical spooks hiding behind the scenes.”

    I have seen another take on this by an atheist here that rationalized non-material entities.

  3. The link:


  4. I try really hard to follow some of this stuff and its over my head many times, are there any recommendations for books to help me out with a good understanding of philosophy.

  5. I asked Steve once and he recommended the following:

    Donald Palmer, Looking at Philosophy (McGraw-Hill 2001)



    http://www.ditext.com/runes/index.html (dated, but useful for older references)

  6. Mathetes said:

    "Also, you make mention of dualism. You've touched on this in some of your posts in the past...does your position adhere to the Cartesian version, or is there more to it than that?"

    I'm basically a Cartesian dualist, although I'm also an interactionist, and I don't assume, pace Descartes, that animals are merely automata.

  7. RE: “Reality is described by science; there are no metaphysical spooks hiding behind the scenes.”

    Basically all I think he is saying is that science observes, tests and measures, and theorises about the natural world.

    The natural world is what constitutes reality. Scientific processes do not endorse nor disendorse the supernatural; they merely has no way to examine it supposing it exists.

    "Methodological naturalism is the belief that observable events in nature are explained only by natural causes without assuming the existence or non-existence of the supernatural."