I'm pleased to announce the completion and posting in draft of an article written by Tim McGrew and me that I hope will be of help and interest to a variety of readers. "A Cumulative Case for the Resurrection of Jesus of Nazareth" is available on my web site here. It has been commissioned for the Blackwell Companion to Natural Theology, edited by William Lane Craig and J. P. Moreland. Blackwell has given us permission to post a draft version of our own, with none of their pagination or typesetting, to a personal web site. If you refer others to the on-line version, please make it clear that it is a draft and that the "real" final version will be published on paper with Blackwell.
The main thing missing from this version is the bibliography, which is presently being formated. The article is in MLA style, so if you should want full reference information for a given book, please feel free to e-mail me and ask.
Since the piece is very long, I'm giving here a few highlights and features, with some page numbers in the PDF, in case you want to zero in on particular sections rather than either browsing it or trying to read the whole thing.
Update The bibliography is now available here.
Although to a large extent we set aside the many textual debates, we do discuss some of these on pp. 4-12 and there give the reader an idea of where to go for further reading.
Though I will be talking a lot about confirmation theory in this post, I want to stress that there is plenty of the article that can be read quite easily without knowing such theory--for example, not only the discussion of textual assumptions but the laying out of what we call the salient facts on pp. 15ff and even our rationale on pp. 29ff for giving those facts the weight we do give them. Though it's a professional article, we also try to explain what we are doing with the formalism.
A major feature of the article is its wedding of historical, evidential apologetics with a detailed use of Bayesian confirmation theory. The advantages of such an approach are too numerous to mention. One advantage is that using the odds form of Bayes's Theorem allows us to isolate the impact of the direct evidence for the resurrection and to see that evidence as clearly as possible without being distracted by indirectly relevant issues such as other arguments for or against the existence of God.
This approach also allows us to address a confusion that comes up frequently in the literature as to whether we are obligated to treat the existence of God as a premise for the occurrence of a miracle like the resurrection or whether we may treat the resurrection as evidence for the existence of God. The sections later in the paper on "Worldview Worries" (pp. 47-55) and on Alvin Plantinga's so-called "principle of dwindling probabilities" (pp. 55-60) deal with this issue in some detail.
A Bayesian approach has the nice feature that it allows us to demonstrate the relative weakness of theories that "explain" the evidence but that have a low prior probability even on the assumption that the resurrection did not occur. We use a Bayes factor analysis to describe the relative power of R (the proposition that the resurrection occurred) and ~R (the proposition that it didn't) to explain various bits of evidence. If a sub-hypothesis--say, the theory that the disciples all had hallucinations--has a low prior probability given ~R, it does little to help ~R in what one might call its "Bayes factor competition" with R to explain the evidence in question. (See pp. 27-28 and 46-47.)
The question of independence is of particular interest in arguments about miracles. Much of the force of arguments for a miracle comes from the existence of multiple, independent lines of evidence--for example, multiple, independent witnesses. But it is easy enough to point out that Jesus' disciples knew one another and were interacting with one another during the time when they testified that Jesus was risen. It's not as though they were all locked up in separate rooms and didn't know what each other were saying. Doesn't this make it unreasonable to treat their testimonies on the matter as independent? The surprising answer is that it doesn't. We discuss that issue on pp. 41-47.
One final feature of this paper that I must mention is its references to works of historical apologetics from the 18th and 19th centuries. This aspect of the paper is entirely the work of my co-author; I am fairly ignorant of the history of apologetics, beyond what I have recently learned from him. But it has emerged very strongly in the course of our research for this article that the last several centuries are full of now-unsung heroes of the faith, men who came forward and defended Christianity against the assaults of, e.g., the deists. And, speaking in terms of the actual strength of their case, they won. That men like David Hume should be thought, now, to have put paid to the possibility of defending a religious miracle by testimony is an historical scandal. References to Chalmers, Butler, Paley, and many more are scattered throughout the paper but are probably heaviest in the final section, pp. 60-69, which discusses the second part of Hume's "On Miracles." Hopefully this paper can be in a sense a down payment on what is owed to these men in the way of unearthing them from an undeserved obscurity.