100 (or so) Arguments for Atheism

A popular view in contemporary analytic philosophy of religion is that while there are many arguments for theism -- cosmological, ontological, and teleological arguments; moral arguments; arguments from consciousness; etc. (by Plantinga's lights, two dozen or so) -- there are only one or two arguments for atheism, viz., the problem of evil and (more recently) the argument from divine hiddenness.

This is a misconception. Here are over a hundred:

10. The Free Will Offense (Schellenberg)
11. Schellenberg's new deductive argument from evil. (Schellenberg) 
12. The argument from the absurdity of life in a Christian (and, arguably, any traditional theistic) universe (Wielenberg)
13. An abductive argument for naturalism (Oppy)
14. The argument from ordinary morality (Maitzen)
15. An ontological disproof of theism (Maitzen)
16. The problem of theistic evidentialist philosophers (Lovering)
17. The argument from autonomy (Kahane, Rachels)
18. The argument from ugliness (Aikin and Jones)
19. The common core/diversity dilemma (Thornhill-Miller and Millican
20. The argument from the philosophy of nature (Cordry)
21. The argument from natural inequalities (Mizrahi)
22. The argument from social evil (Poston)
23. The argument from insect suffering (Crummett)
24. The argument from scale (Everitt)
25. The argument from religious evil (Kodaj)
26. The argument from idolatry (Linford and Megill)
27. The argument from indifference (Linford and Megill)
28. The argument from the requirement of divine interference (Maring)
29. The argument from eternally separated lovers (Hassoun)
30. The argument from peer disagreement
31. The argument from the impropriety of worship (Aikin)
32. The argument from the impropriety of belief (Nagel)
33. The argument from abstract objects (Davidson, Craig, me)
34. The argument from inhospitable environment (me)
35. The argument from teleological evil (me)
36. The argument from material causality (me)
37. The argument from revulsion (me)
38. The argument from the ineffectiveness of prayer (various)
39. The argument from divine evil (Lewis)
40. The argument from hell (Sider)
41. The argument from the meaning of life (Megill and Linford)
42. The argument from the demographics of theism (Maitzen)
43. The problem of no best world (Rowe, others)
44. The problem of incoherent/incompatible properties (various)
45. The problem of mitigated modal skepticism (me)
46. The structure and dynamics argument (me)
47. The argument from Mandevillian intelligence (me)
48. The argument from quantum mechanics (me)
49. The argument from wave function realism (me)
50. The argument from low priors (Draper)
51. The argument from decisive evidence (Draper)
52. Epicurean cosmological arguments for naturalism (me)
53. The argument from cognitive biases (Lucas, me)
54. The argument from the etiology of religious belief (De Cruz, others)
55. The argument from moral psychology (Park)
56. The argument from moral epistemology (Park)
57. The argument from meager moral fruits (Draper)
58. The argument from imperfection (Everitt)
59. Smith's cosmological argument for atheism (Smith)
60. The argument from tragic moral dilemmas (me)
61. The argument from substance dualism (me)
62. A Leibnizian cosmological argument for naturalism (me)
63. Arguments from order and fine-tuning against theism (me)
64. Arguments from sub-optimality (Darwin, Dawes, others)
65. Probabilitistic ontological arguments against theism (me)
66. Arguments from the success of naturalistic explanations (D. Lewis, Dawes, others)
67. The argument from lack of character (me)
68. The problem of divine authority (me)
69. The problem of polytheisms (Lataster and Philipse)
70. The problem of alternative monotheisms (Lataster)
71. An abductive argument for liberal naturalism (me)
72. The problem of demiurgism (me)
73. Sterba's deductive argument from evil
74. Another ontological disproof of classical theism (me)
75. The problem of natural nonbelief (Marsh)
76. The problem of permissivism (me)
77. Pragmatic arguments for atheism (Cockayne & Warman; Lougheed)
82. The argument from anti-religious experience (me, Adams and Robson)

Top 10 Posts of 2018

A Weaker Version of the Principle of Material Causality Gets the Same Result
Wave Function Realism, The Unreliability of Perception, and Theism
Ontological Arguments, Anselmianism, and Irony
Notes on Morriston's "Creation Ex Nihilo and the Big Bang"
Notes on Morriston's "Must the Beginning of the Universe Have a Personal Cause?"
Partial Notes: Morriston's "A Critical Evaluation of the Kalam Cosmological Argument"
Notes on Morriston's "Must There Be a Standard of Moral Goodness Apart from God?"
Notes on Ruloff's Papers on Arguments from Propositions and Intentionality to God
Notes Index: Ehrman's Jesus: Apocalyptic Prophet of the New Millennium
Quote of the Day

and one to ring in the new year:
Three Dozen (or so) Arguments for Atheism

Maitzen on Normative Objections to Theism

Maitzen, Stephen. "Normative Objections to Theism", in Oppy, Graham, ed. The Blackwell Companion to Atheism and Philosophy (Wiley-Blackwell, 2019).

Linford and Megill's New Paper on Two Underexplored Arguments Against Theism

Linford, Dan and Megill, Jason. "Idolatry, indifference, and the scientific study of religion: two new Humean arguments", Religious Studies (2018), doi:10.1017/S0034412518000653.

Here's the abstract:
We utilize contemporary cognitive and social science of religion to defend a controversial thesis: the human cognitive apparatus gratuitously inclines humans to religious activity oriented around entities other than the God of classical theism. Using this thesis, we update and defend two arguments drawn from David Hume: (i) the argument from idolatry, which argues that the God of classical theism does not exist, and (ii) the argument from indifference, which argues that if the God of classical theism exists, God is indifferent to our religious activity.

"The Will Not to Believe"

...is the title of an intriguing new paper (forthcoming in Sophia) by Joshua Cockayne and Jack Warman. Here's the abstract:
Is it permissible to believe that God does not exist if the evidence is inconclusive? In this paper, we give a new argument in support of atheistic belief modelled on William James’s The Will to Believe. According to James, if the evidence for a proposition, p, is ambiguous, and believing that p is a genuine option, then it can be permissible to let your passions decide. Typically, James’s argument has been used as a defence of passionally caused theistic belief. However, in the existing literature, little attention has been given to topic of passionally caused atheistic belief. Here, we give much needed attention to the issue of how areligious passions can justify atheistic belief. Following James, we argue that if atheism is a genuine option for an agent, it is permissible to believe that God does not exist based on her hopes, desires, wishes, or whatever passions incline her to disbelieve. After defending the coherence of passionally caused atheism, we go on to suggest why this position is a tenable one for the atheist to adopt.

McDaniel's New Critique of Grounding-Based Formulations of PSR

McDaniel, Kris. "The principle of sufficient reason and necessitarianism", Analysis (2018).

Here's the abstract:

Peter van Inwagen (1983: 202–4) presented a powerful argument against the Principle of Sufficient Reason, which I henceforth abbreviate as ‘PSR’. For decades, the consensus was that this argument successfully refuted PSR. However, now a growing consensus holds that van Inwagen’s argument is fatally flawed, at least when ‘sufficient reason’ is understood in terms of ground, for on this understanding, an ineliminable premiss of van Inwagen’s argument is demonstrably false and cannot be repaired. I will argue that this growing consensus is mistaken and that a powerful argument relevantly similar to van Inwagen’s should still concern us, even when we understand ‘sufficient reason’ in terms of ground.

The penultimate draft can be found here.

Ontological Arguments, Anselmianism, and Irony

*Quick thought. For kicks (sort of). In draft.*

Plantinga's modal ontological argument aims to demonstrate the existence of a modally ultimate being, upon which all else depends for its existence. But the argument seems to require an ontologically prior Platonic modal space, of which God's existence is a function (of its existence and structure).[1] To see this, just think about the way in which Plantinga's ontological argument proceeds, and its ontological underpinnings. On such a view, abstract objects are fundamental, and God is an intermediary, derivative layer of reality -- sandwiched between the layers of necessary abstracta and contingent concreta -- contrary to the aims of the argument (and to the Aseity-Sovereignty doctrine).

The theist might try to avoid this implication via theistic activism. But if the theist goes this route, then the bootstrapping problem rears its ugly head, in addition to other problems (e.g., the Benacerraf problem for God's knowledge of abstracta, etc.).  On the other hand, if the theist takes the theistic conceptualist route, then this also leads to the bootstrapping problem and other problems (e.g.,  concepts are concrete, non-repeatable entities, and therefore not suitable to play the role of abstracta).  On yet a third hand, if the theist goes the nominalist or fictionalist route, then the ontological argument fails. But this exhausts the relevant possibilities for the theist regarding abstracta: straight platonism, theistic activism, theistic conceptualism, nominalism, and fictionalism/pretense theory.[2], [3]

There is thus non-trivial pressure to think that Plantinga's ontological argument succeeds only if the being it proves to exist is not ontologically ultimate, but rather one that supervenes on an ontological foundation of platonic modal space.

[1] Rather like a modal analogue of the way in which (some argue that) our universe is a function of the nature and structure of relativistic quantum fields.

[2] Hybrid views tend to take on board the problems of pure views.

[3] Strictly speaking, other views are epistemic possibilities (e.g., structuralism) -- but seem to succumb to similar problems.

[4] Such a being would be what I have elsewhere called a necessary dependent being, and the platonic modal space on which it depends would be what I have elsewhere called a necessary independent being.

Wave Function Realism, The Unreliability of Perception, and Theism

According to a growing chorus of voices (see, e.g., a number of papers in this volume), (i) there are strong grounds for accepting a realist construal of quantum mechanics (QM), and (ii) the most plausible interpretations of QM entail that the wave function is a real entity that exists in a configuration space of very, very many dimensions (about 3x10 to the 80th, according to current estimates). But it's not at all clear how the ordinary three dimensions of our experience can be accounted for via this configuration space. In fact, some (e.g., Alyssa Ney) have argued that they probably can't, in which case there is non-trivial epistemic pressure to think the three dimensions of ordinary experience are, in an important sense, mirage-like (L.A. Paul has similar worries, but doesn't explicitly come down on the matter one way or the other. Jill North and Jenann Ismael have stronger suspicions). But if so, then there is non-trivial epistemic pressure to think that ordinary perceptual experience is massively unreliable.

Now according to many theists, if God exists, then God designed us in such a way as to ensure that our perceptual faculties reliably track the truth about the world, where this includes beliefs about the ordinary objects of experience having extension in length, width, and depth. But if the above worry is at all on track, then ordinary perceptual experience is massively unreliable. And if that's right, then something has to give: either (i) theism is false, or (ii) the existence of the God of theism doesn't make it likely that our perceptual faculties are reliable. The first disjunct is of course fatal to theism. The second might well be equally so, given a few more premises concerning God's omni-attributes and a bridge principle from those attributes to the reliability of sense perception. However, even if this can't be done, the second disjunct still causes trouble for a number of theistic apologetic strategies. For example, it would be devastating to Plantinga's theory of warranted Christian belief. Either way, then, if the growing chorus is onto something, troubling epistemic consequences for theism follow.

Notes on Ruloff's Papers on Arguments from Propositions and Intentionality to God

“Divine Thoughts and Fregean Propositional Realism”, IJPoR 76:2 (2014): 41-51.

In this paper, Ruloff critiques Anderson and Welty’s argument from intentionality to God. Ruloff expresses their argument as follows:
P1: The laws of logic are propositions.
P2: Propositions bear intrinsic intentionality.
P3: Therefore, the laws of logic are propositions that bear intrinsic intentionality. (From P1 and P2)
P4: x is intrinsically intentional only if x is mental (or mind-dependent).
P5: Therefore, the laws of logic are propositions that are mental (or mind-dependent). (From P3 and P4)
P6: The laws of logic exist necessarily.
P7: If the laws of logic exist necessarily and are mental, then the laws of logic are the contents of a necessarily existent mind.
P8: Therefore, the laws of logic are the contents of a necessarily existent mind. (From P5, P6, and P7)
P9: Therefore, a necessarily existent mind exists. (From P8)
Ruloff offers a defeater for P4, i.e., against the thesis that intentional entities must depend upon a mind. Toward that end, he sketches a standard account of propositions: Fregean Propositional Realism (FPR), according to which propositions are mind-independent abstract objects that are intrinsically and essentially intentional, and yet wholly independent of any and all minds. Officially:
Fregean propositional realism (FPR): Propositions are abstract, mind and language-independent, truth-bearing, representational entities, that function as the referents of propositional attitude reports and the meanings of declarative sentence-tokens.
Crucially, it is widely held that there is a straightforward argument for their intrinsic and essential intentionality, viz., propositions are the primary bearers of truth-values, and their truth-values are simply a function of their correspondence with the world (or lack thereof). Indeed, on FPR, the intentionality of thought is derivative of/dependent upon the intentionality of propositions, and not the other way around. 

Ruloff offers two main arguments for FPR that are commonly given and widely accepted. The first is the “singular term” argument (STA). According to STA, in sentences containing “that”-clauses e.g., “Joe believes that the ball is red.” – the “that”-clause is most naturally taken to be a singular term that is a referring expression. Prima facie, then, there is something to which “that”-clauses refer. But the most plausible candidates for such referents are mind-independent, abstract objects. 

The second argument is that the robust theoretical utility of FPR confers justification on the view. For example, it explains how (a) “the same semantic content can be expressed by different people uttering different sentence-tokens in different languages”; (b) “how the same semantic content can be believed by different people”; (c) “how mental states gain their representational content”; and (d) “makes intuitive sense of our ascriptions of truth and falsity”; etc.

Given that FPR is a well-motivated view and that it entails that propositions are essentially intentional entities and yet mind-independent, P4 is undercut.

“Against Mind-Dependence”, Philo 17:1 (2014):92-98.

In this paper, Ruloff evaluates Gould’s argument that propositions are caused by/grounded in a (divine) mind:
1. Propositions bear intrinsic intentionality.
2. X is intentional only if x is mental.
3. Therefore, propositions are mental (mental entities or mental states).
4. If propositions are mental, then they are thoughts.
5. Therefore, propositions are thoughts.
6. If propositions are thoughts, then they are effects of some mind.
7. Therefore, propositions are the effects of some mind.
Ruloff’s core claim is that (3) is subject to a reductio, entailing four assumptions that are widely rejected on grounds of implausibility: (i) had there never been any mental states, then there would be no true or false propositions; (ii) the total number of true and false propositions must equal the total number of mental states; (iii) it’s impossible for two people to believe the same proposition or share the same thought; and (iv) some propositions will fail to have contradictories. But if so, then there are grounds for rejecting (1) or (2) (or both).

(i) is implausible because it’s a prima facie false counterfactual or counterpossible: propositions would be true or false even if no one entertained them. (cf. McGrath (2012), Soames (1999), Jubien (1997), Swartz & Dowden (2004), etc.). The following counterfactual/counterpossible is even more salient: If there were no beings capable of mental states, “There are no beings capable of mental states” would’ve still been true ; (ii) is implausible because prima facie, there are many – infinitely many – more propositions than mental states; (iii) is implausible because propositions can be entertained by -- be “in” -- more than one mind at once, while mental states are concrete particulars, located in discrete regions of space and time, and no concrete particular can be in more than one region of space and time at once (cf. Frege); and (iv) is implausible because there seem to be propositions that at most one person has ever thought of. But if so, then if propositions are mental states, then no one will have thought of the contradictories of such propositions. But it’s absurd to think that some propositions don’t have contradictories. Therefore, propositions can’t be mental states. Until Gould addresses these implausibilities, his theistic conceptualism has undefeated defeaters.

“On Propositional Platonism, Representation, and Divine Conceptualism”, EJPR 8:4 (Winter 2016): 195-212.

In this paper, Ruloff critiques Gould & Davis’ critique of propositional platonism in favor of theistic conceptualism. In particular, Ruloff argues that their argument relies upon at least five implausible assumptions: (i) propositions must be representational in order to be the bearers of truth-values; (ii) propositions are abstract entities whose constituent components are properties, relations, and concrete individuals (i.e., that propositions are to be given a Russellian analysis); (iii) propositions are structured abstract entities; (iv) a proposition’s truth-conditions must be explained solely in terms of the representational properties of its constituent components; and (v) if the propositional platonist isn’t able to explain the representational properties of a proposition in terms of its constituents, a wholesale rejection of propositional platonism is justified. 

Against (i): Jeff Speaks’ account of propositions analyzes them in terms of properties. On Speaks’ account, “The ball is red” is true just in case the property being such that the ball is red is instantiated. But if so, then propositions can be bearers of truth-values without being inherently representational entities. Absent a successful critique of Speaks' account, (i) is undercut.

Against (ii): (a) Fregean accounts of propositions take the constituents to be structured senses, or modes of presentation, and not properties, relations, and individuals; and (b) Moorean accounts of propositions take them to be structured concepts/open sentences plus gap-filling expressions. Absent a successful critique of these accounts, (ii) is undercut.

Against (iii): (a) George Bealer has a well-developed and defended account of whole propositions as unstructured, metaphysically simple, ontologically primitive, sui generis abstracta. These simple entities are associated with a decomposition tree, and thus structure can be attributed to propositions in this derivative way; (b) Robert Stalnaker has a well-developed and defended account of propositions as sets of possible worlds, or functions from possible worlds to truth-values. Example: the proposition expressed by “Jen is a philosopher” is the set of worlds in which the referent of 'Jen'   is a member of the set of things that are philosophers. Equivalently, it’s the function f that maps a possible worlds w to the value True just in case Jen is a philosopher in w; (c) Keller & Keller have recently argued that the principle of compositionality is false, as it admits of counterexamples or rests on several very controversial assumptions. Absent a successful critique of these accounts and K & K's apparent counterexamples, (iii) is undercut.

Against (iv):  we've seen above that (a) Jeff Speaks has a well-developed and defended account of propositions as properties that entails that propositions are not representational at all, in which case they lack representational components; (b) Bealer’s and Stalnaker’s accounts of propositions entail that their truth-conditions don’t depend on simpler and structured parts; and (c) Keller & Keller’s arguments provide an undercutting defeater for the claim that propositions have internal components as constituents (cf. their arguments against the principle of compositionality). Absent a successful critique of these, (iv) is undercut.

Against (v): Even if the propositional platonist can’t explain how an abstract proposition can get its representational features from its constituents, propositional platonism would still be rationally justified. This is because of the widely endorsed theoretical utility argument for propositional platonism. According to the argument, because the view “elegantly and powerfully simplifies, unifies, and systematizes our thinking about language and communication, a commitment to propositional platonism is warranted.” (p. 209). For example, it explains (a) how the same semantic content can be expressed by different people uttering the different sentence tokens of different languages; (b) how the same semantic content can be believed by different people; (c) how mental states gain their representational content; (d) alethic modality (possibility, necessity, contingency, etc.); and (e) our ascriptions of truth and falsity. As Michael Jubien argues, given the impressively strong theoretical utility argument, we should “try to get used to the mystery” of how propositions can be intrinsically representational entities (Jubien 1997, p. 103). She is therefore warranted in taking the representational properties of propositions to be a theoretical primitive. Absent a successful critique of this argument, (v) is undercut.

Checking In

Hi Gang,
Just wanted to check in and say that I hope your summer is going well. 

Also, this. (H/T: K.) You're welcome.

A Weaker Version of the Principle of Material Causality Yields the Same Conclusion

Recent versions of the cosmological argument deploy weaker causal and explanatory principles to make them more plausible: Perhaps the evidence isn't strong enough to support the claim that, necessarily (or at least, actually), every object or state of affairs (or some proper subclass thereof) has a sufficient reason or efficient cause of their existence or obtaining. And perhaps such principles beg the question against the atheist by assuming the causal or explanatory structure of a theistic universe. But surely (so the thought goes) the atheist's intuitive evidence supports the weaker claim that everything can have an efficient cause or sufficient reason. Such arguments go on from there to argue that a theistic conclusion can be gotten from such weaker principles.

In the same spirit, I think the principle of material causality can be weakened so as to be adequately supported by even the theist's intuitive evidence:
Weak PMC: Possibly, every concrete object (and aggregate of such) that has an originating or sustaining efficient cause has an originating or sustaining material cause, respectively. 
In simple terms, Weak PMC says that it is possible that all made things are made from or out of other  things. A bit more carefully, it says that there is at least one possible world in which all concrete individuals and stuffs that are made are made from or out of other concrete individuals or stuffs.

Now my own view is that a much stronger version of PMC is true -- viz., that it holds of metaphysical necessity. However, I'd be willing to bet that most people believe that PMC is non-vacuously true in at least the actual world, and for good reason: it's intuitive, it has no uncontroversial exceptions, and it's encoded in the well-confirmed conservation laws of physics. A fortiori, then, I think that even the theist has enough intuitive evidence to warrant the claim that there is at least one possible world W in which such a principle is non-vacuously true. But if so, then in W, there are concrete objects that are made, and all concrete objects that are made are made out of other things or stuff. And if so, then no concrete objects in W that are made are made ex nihilo, in which case no god or gods made them ex nihilo in W. But on classical Anselmian theism, for any world that contains concrete objects or stuffs distinct from God, at least some of those objects or stuffs were made ex nihilo.  It follows that the god of classical Anselmian theism doesn't exist in W. But if so, then by (i) the fact that classical Anselmian theism entails that God is a metaphysically necessary being, and (ii) Axiom S5 of S5 modal logic, it follows that such a God doesn't exist in the actual world, either.

Therefore, one can generate a successful argument from material causality against theism with even  a very weak version of PMC. How powerful of an argument is it?

Very powerful, I'd say. To see this, consider the following line of reasoning. Leaving aside formal and final causes, there appear to be four possible scenarios for causal principles that govern a universe:

(i) Both an efficient cause principle and a material cause principle
(ii) An efficient cause principle without a material cause principle
(iii) A material cause principle without an efficient cause principle
(iv) Neither an efficient cause principle nor a material cause principle

Here's the rub: our grounds for a material cause principle are at least as strong as our grounds for an efficient cause principle (I would say they're stronger for a material cause, as there appear to be counterexamples to the efficient cause principle in cases of certain quantum phenomena). But if so, then starting from the neutral standpoint of agnosticism, there is a default presumption in favor of (i) as one's epistemic starting point. Rejecting it in favor of one of the other three options therefore requires adequate grounds for doing so. But it's unproblematic for the atheist to start at (i).  By contrast, the theist has their work cut out for them: They must find grounds to reject (i) in favor of (ii).  But prima facie, it looks like no such grounds can be forthcoming.  For prima facie, any argument for theism will run afoul of Weak PMC, as Weak PMC blocks cosmological, ontological, and design arguments, and arguments from substance dualism as arguments for theism (as opposed to arguments for, at best, pantheism, panentheism, and demiurgism). It therefore looks as though the argument from material causality holds out the promise (or peril) of constituting an intrinsic defeater-defeater against classical Anselmian theism.

By contrast, it's not obvious to me that the same can be said of the standard arguments for atheism. For such arguments seem to be resistible if the theist can provide other grounds for theism, such as those mentioned above. But the problem is that, unlike the argument from material causality, the arguments from evil and hiddenness can't hamstring those arguments. It therefore seems to me that the argument from material causality holds out the prospect of being the most powerful argument against theism.

The argument from material causality also has other epistemic functions that are slightly weaker, yet still important. First, it can show that the theist incurs a theoretical cost by rejecting Weak PMC, as it seems to be supported by just about anyone's intuitive evidence (at least prior to reflecting on its implications and revising one's web of beliefs). Second, it can function as an undercutting defeater for the newer versions of cosmological arguments mentioned at the beginning of our discussion. 

In at least these ways, then, it seems to me that the argument from material causality is very powerful indeed.

Notes on Morriston's "Creation Ex Nihilo and the Big Bang"

Notes on Morriston’s “Creation Ex Nihilo and the Big Bang”, Philo 5:1 (2002), pp. 23-33.

0. Introduction (fill in later)

1. Craig’s First Argument: Infinite Density = Nothing
1.1 According to the Big Bang theory, the universe began with a great explosion from an infinitely dense point-particle.
1.2 There can be no object having infinite density.
1.3 So, “infinite density” is synonymous with “nothing”.
1.4 Therefore, the Big Bang theory requires that the universe had a beginning and was created out of nothing.

2. Criticisms of Craig’s First Argument
2.1 First, an infinitely dense point-universe is not nothing
2.1.1 the initial singularity is not nothing Nothingness can’t begin expanding, since there is nothing there to expand. By contrast, the “point-universe” began expanding. Even if the point-universe lacks spatial and temporal spread, it yet has other properties, e.g., being a point, being infinitely dense, being capable of expanding, etc.
2.2 Second, 'infinitely dense entity' is not synonymous with 'nothing'
2.2.1 Compare: (a) There can be no round squares; therefore, (b) 'round square'; is synonymous with 'nothing'. 
2.2.2 In general, from the fact that there can be no entity E, it doesn’t follow that ‘E’ is synonymous with ‘nothing’.
2.2.3 Therefore, even if there can be no infinitely dense point-universe, it doesn’t follow that ‘infinitely dense point-universe' is synonymous with ‘nothing’
2.3 Third, if nothing can be infinitely dense, then the universe was never infinitely dense
2.3.1 But if so, then premise 1 is false as stated
2.3.3 Craig’s argument is therefore subject to a dilemma: Either we change Craig’s description of what the Big Bang theory says, so that it doesn’t involve a state of infinite density, or we don’t if we do, then the argument loses its basis for inferring that the universe was created out of nothing If we don’t, then the criticism above (viz., that if (2) is true, then (1) is false) goes through either way, the argument is unsound
2.4 Fourth, few Big Bang cosmologists today think the universe was ever infinitely dense and point-sized
2.4.1 It’s true that, on the standard Big Bang model, the geometry of the universe’s expansion, when traced backward in time, continually decreases toward a limit of a diameter of zero
2.4.2 However, the limit diameter zero is thought of as an ideal, and not an actual, limit
2.4.3 Furthermore, we have no theory that allows us to infer the universe’s behavior  as we approach this limit Relativity theory breaks down prior to Planck time (i.e., 10-43seconds), and quantum effects become dominant prior to that point we need a quantum theory of gravity (which we don’t yet have) to accurately infer what happened prior to Planck time Until then, any claim about the earliest stage of our universe’s history is “sheer speculation”

3. Craig’s Second Argument: No Time Prior to the Singularity Entails Creation Ex Nihilo
3.1 The initial singularity exists at the earliest point of space-time.
3.2 There is no time prior to the earliest point of space-time.
3.3 Therefore, there was nothing temporally prior to the initial singularity.
3.4 So, the initial singularity must have come into existence out of nothing.
3.5 If, therefore, the initial singularity was created, it must have been created out of nothing

4. Criticisms of Craig’s Second Argument
4.1 First, the Big Bang theory doesn’t entail that there was no time “prior” the singularity
4.1.1 Craig himself has argued that it’s possible that there is a more fundamental “metaphysical time” that can exist independently of the physical time of our universe.
4.1.2 Craig’s thought experiment: Suppose God led up to creation by counting “1, 2, 3, … fiat lux!” 
4.1.3 In this scenario, time is elapsing, and yet no physical objects exist.  Its moments are individuated by the succession of contents in God’s mind.
4.1.4 Craig thinks this thought experiment shows that a time “prior” to physical time is metaphysically possible.
4.1.5 In fact, given Craig’s theism, he thinks this metaphysical time is actual
4.1.6 According to Craig, metaphysical time is absolute, tensed, and dynamic
4.1.7 But if so, then Craig’s view of metaphysical time entails the metaphysical possibility of time prior to the Big Bang singularity
4.1.8 The epistemic possibility of metaphysical time prior to the singularity is thus an undercutting defeater for premise 2
4.2 Second, there being nothing temporally prior to the singularity doesn’t entail there being nothing causally or ontologically prior to the singularity
4.2.1 What follows from (3) is not (4), but rather the weaker claim that the universe didn’t come from something that existed  at an earlier time.  It's therefore compatible with the possibility of the universe created from timeless stuff
4.2.2 To close the logical gap between (3) and (4), then, we need another premise, viz.,
(3 ½) If there was nothing temporally prior to the initial singularity, then it must have come into existence out of nothing.
4.2.2 But it’s not clear that (3 ½) is true Craig is already committed to saying that, ontologically prior to the singularity, God exists timelessly It’s therefore not clear what principled grounds he could have to rule out the epistemic possibility that, causally and/or ontologically prior to the singularity, other things besides God exist timelessly (e.g., a timeless stuff from which the universe was made) And if that’s right, it’s not clear how Craig can rule out the epistemic possibility that God created the universe out of a timeless stuff Therefore, even if there was no time prior to the singularity, it doesn’t follow that God created the universe out of nothing
4.2.3 Anticipated reply: Craig thinks he can rule out the epistemic possibility of timeless stuff because he thinks: (a) the only possible stuff from which God could make the universe is matter/energy; (b) timeless stuff is quiescent; and (c) matter/energy is never quiescent.
4.2.4 Rejoinder: it’s not clear that (a) is true: it’s epistemically possible that God created the universe out of some timeless stuff that’s distinct from matter/energy
4.2.5 Third, the grounds for a requirement of a timeless efficient cause of the universe is on an epistemic par with the grounds for a requirement of a timeless material cause The evidence for both causal principles is the same Both are intuitive Both enjoy strong empirical confirmation [N.B., actually, the case for material causes is stronger, given apparent counterexamples to the need for efficient causes in quantum mechanical phenomena. --EA] The theoretical costs of both is the same We’ve never observed timeless stuff, but we’ve never observed a timeless person, either. It’s odd to think that the material cause of the universe was timeless sans creation, and then entered time with its creation, but then Craig thinks the same is true of the efficient cause of the universe: God is timeless sans creation, but entered time at the moment of creation Given epistemic parity, we have a dilemma: Either our commonsense intuitions about ordinary cases of causation can reasonably applied to the beginning of the universe or they can’t. If they can, then creation of the universe out of timeless stuff is more plausible than creation ex nihilo. If they can’t, then we can’t draw any conclusion whatever about the existence and nature of the cause of the universe. Either way, Craig’s argument fails

Important New Book

Here. Not quite on point for a philosophy of religion blog, but important nonetheless, and no doubt of interest to some readers of this blog. A nice review of the book can be found here.

Index: Notes on Ehrman's Jesus: Apocalyptic Prophet of the New Millennium

Below is a list of links to my recent series of posts on Ehrman's Jesus: Apocalyptic Prophet of the New Millennium. The book provides a nice, clear summary of the case for the mainstream view among New Testament scholars that Jesus was fundamentally an apocalyptic prophet heralding an imminent eschaton, and not the Son of God.

(ch. 1 is omitted, as it can be summarized quickly as follows: Christians from the present all the way through the past have believed that the end of the world would occur in their generation. There is a good reason for this: Jesus thought so, too.)

Notes: Assessing The Case for an Apocalyptic Jesus in Ehrman’s Jesus: Apocalyptic Prophet of the New Millennium

Notes: Assessing The Case for an Apocalyptic Jesus in Ehrman’s Jesus: Apocalyptic Prophet of the New Millennium


-Scholars use criteria of authenticity to sift the earliest sources with eyewitness testimony to reconstruct the historical Jesus

-Applying these results yields seven general lines of evidence that Jesus was an apocalyptic prophet:
  • #1: The earliest sources portray Jesus as an apocalyptic prophet
  • #2 The later sources tone down the apocalyptic language in the earlier sources
  • #3: Those connected to Jesus and his message before and after his earthly ministry were apocalypticists
  • #4: The apocalyptic prophet hypothesis makes best sense of his core teachings
  • #5 The apocalyptic prophet hypothesis makes best sense of his ethical teachings
  • #6: The apocalyptic prophet hypothesis makes best sense of his actions
  • #7: The apocalyptic prophet hypothesis makes best sense of his last days 
On the Nature of Ehrman’s Argument

-Ehrman’s argument is best construed as an inference to the best explanation
  • Inferences to the best explanation proceed by listing a range of data, and arguing that one competing hypothesis best explains that data
  • A hypothesis is the best explanation of the data to the extent that it exemplifies the theoretical virtues (simplicity, scope, conservatism, predictive power, etc.) better than any competing theory
Reconstructing Ehrman's Argument
-Call the hypothesis that Jesus was an apocalyptic prophet, Apocalyptic Prophet
-Ehrman’s argument can then be stated as follows: 

1. If Apocalyptic Prophet is the best explanation of the relevant data, then Jesus was probably an apocalyptic prophet.
2. Apocalyptic Prophet is the best explanation of the relevant data (cf. 1-7 above)
3. Therefore, Jesus was probably an apocalyptic prophet

Evaluating Ehrman’s Argument

-The key premise is (2): the Apocalyptic Prophet hypothesis is the best explanation of the relevant data.

-But is it?

-To answer that, we’ll need to answer two other questions:
  • What standards does an explanation have to meet to be the best? 
  • Does the Apocalyptic Jesus hypothesis meet those standards better than any competing hypothesis?
Step 1: What Makes an Explanation the Best?
-In the type of case at hand, predictive power and related theoretical virtues aren't all that relevant, as we're evaluating hypotheses that explain past facts, and not future phenomena.  Determining which hypothesis is the best explanation therefore largely boils down to the following three:
  • 1. Simplicity: One hypothesis is simpler than another if the former adds fewer new assumptions than the latter to explain the data.
  • 2. Scope: One hypothesis has wider explanatory scope than another if the former explains more data than the latter.
  • 3. Conservatism: One hypothesis is more conservative than another if the former requires throwing out fewer of our prior beliefs than the other that are already well-justified.
-In short, then: the simplest, most conservative hypothesis that has the widest explanatory scope is the best explanation of the historical Jesus, and is thus the most probable one.

-The consensus view: The Apocalyptic Jesus hypothesis best explains the relevant data, viz., at least the seven discussed in Ehrman’s book (cf. Sanders, Vermes, Fredriksen, et al.)

-However, some scholars deny this and reject premise 2

-If they are better explainations, then they must meet the criteria above (and perhaps others) better than The Apocalyptic Prophet hypothesis.

-The main obstacle with such approaches is that they have trouble explaining all the early, multiply-attested data for this hypothesis discussed in Ehrman’s book. Let’s briefly look at how they aim to do so:

1. Responses from Conservatives:
A. Divide and Conquer (Craig Blomberg, et al.): the strategy is to take the group of passages that seem to have Jesus saying the apocalypse is imminent (i.e., that it would occur within his generation), and provide an alternative interpretation for each one (e.g., when Jesus said that his generation wouldn't pass away before the apocalypse happened, what he really meant was that the Jews wouldn't pass away before it happened. When Jesus said that his earliest disciples wouldn't finish preaching to the surrounding cities of Israel before the end happened, he was really referring to the perennially incomplete task of evangelism to the Jews; etc., etc..)

B. Destruction of the Temple (NT Wright, et al.): the strategy is to argue that, yes, Jesus really did predict the end within his generation, i.e., we should take those passages at face-value. However, all the end-time passages were fulfilled with the destruction of Jerusalem in 70 A.D.

C. Possible End Only (Ben Witherington, et al.): the strategy is to assert that what Jesus really meant in all those passages is that the End *might* be at hand, not that it *is* at hand.

2. Responses from Moderates
A. Jesus Didn't Say That: (Raymond Brown et al.): the strategy is to say that all the passages that have Jesus preach the imminent coming of the apocalypse are inauthentic -- i.e., Jesus didn't say those things; they were attributed to Jesus by the early church. However, Jesus is the Son of God.

B. Jesus Said it, and He Was Wrong. But So What! (Dale Allison, C.S. Lewis et al.): the strategy is to admit that the passages where Jesus predicts an imminent end are authentic, and to admit that he was wrong -- the End didn't happen -- but to say that it has no serious implications for the truth of Christianity. Jesus is still the Son of God.

3. Responses from Liberals: 
Jesus Wasn't an Apocalypticist (Crossan, Mack, et al.): the strategy is a variation on the Jesus Didn't Say That response of the Moderates. The strategy is to argue that all those passages that have Jesus preach the imminent coming of the apocalypse are inauthentic -- i.e., Jesus didn't say those things; they were attributed to Jesus by the early church. However, Jesus wasn't the Son of God, either -- at least not in any literal sense. Rather, he was a social reformer, a revolutionary, a sage philosopher, or some variant of thereof.

Evaluation: Which Hypothesis Is the Best Explanation?

-To evaluate Ehrman’s Apocalyptic Jesus argument, then, we need to:
  • Examine each hypothesis
  • Determine which hypothesis is the simplest, most conservative hypothesis with the widest explanatory scope
-The consensus view: hypothesis that Jesus was primarily an apocalyptic prophet (cf. Sanders, Vermes, Fredriksen, Ehrman, etc.)

-Why? The other proposals mentioned above seem to be more complex (i.e., they add new assumptions to explain the same data), have less explanatory scope (i.e., they can’t explain all seven lines of data as well, and sometimes not at all), and/or less conservative (i.e., they require that we throw out some of the seven lines of data discussed in Ehrman’s book).

Conclusion: It therefore looks as though Ehrman is correct: the most probable hypothesis is that Jesus was primarily an apocalyptic prophet of an imminent eschaton.

Review of Ekstrom's <i>God, Suffering, and the Value of Free Will</i>

  Kevin Timpe reviews the book for NDPR .