Monday, January 29, 2007

The ODoR of Howard Van Till

Over at The Secular Outpost, Taner Edis passed along a bit of information about Howard Van Till, formerly a physics professor at Calvin College. He links to the written version of a talk given by Van Till last May to the Freethought Association of West Michigan in which he describes his journey from Calvanist dogmatism to (still Christian?) free-thought. Central to his talk/paper is the concept of one's Operative Depiction of Reality (ODoR), somewhat similar to a worldview but perhaps even larger in scope.

He touches on a number of points that reflect my own experience and observations. For instance, I have asserted that beliefs can produce positive effects without the subject of the belief being true. Van Till writes "The fact that one’s ODoR exhibits many utilitarian values does not ensure that any particular portion of it is true."

The paper is only ten pages long, and well worth reading.

Friday, January 26, 2007

What's Anger Got to Do With It?

This post is part of an ongoing dialog between my friend Ernie and me about the validity of Christian belief. It is a response to Of Anger, Hatred, and Love.


I can appreciate that you are trying to introduce a new subtopic in our discussion that might help to illuminate the difference between the moral obligations we might have under your deistic hypothesis (or even Christianity specifically) versus any non-deistic or non-theistic alternative I might offer. As you know, just last night I wrote that the propositions under consideration were insufficient to provide this illumination. I had already seen your later post by the time I finished writing, but I was not prepared to respond to it yet, for reasons that I will now attempt to make clear.

As I read your post, I was a bit surprised to see you propose anger as a key topic that might open the way to greater understanding. Anger is primarily an emotional response to various kinds of provocation, and while I agree that (some) emotions are relevant to discussions of ethics and morality, I think this is primarily by way of the emotional states that are provoked by one's actions and whether we desire those states or not. Your focus on anger, on the other hand, appears to be aimed at how we express our anger or what we do in response to feelings of anger.

Since anger can be provoked and expressed in a variety of ways, and since anger can result from false beliefs as well as true ones, it seems like a pretty slippery concept to be an anchor for this discussion.

Now, you do propose a different definition of anger to use for purposes of this discussion: "the desire to punish those we believe did (or will) mistreat those we love" -- including, but not limited to, ourselves." This definition is a little better suited as a basis for discussion, but I dislike adopting specialized definitions that differ significantly from normal usage because we consciously have to substitute an unusual meaning for the word whenever we encounter it, and this creates "friction" that inhibits understanding as well as the possibility of (unintentional) equivocation.

In this case, I am confused by the conflation of "anger" (even in your limited sense) with "our moral obligation towards those who mistreat others". Certainly we often feel angry when somebody mistreats us or somebody else, and we might discuss whether we "ought" to feel anger in such circumstances. But it seems to me that the more central issue is how we should treat those that mistreat us or others, and how we deal with feelings of anger is just one part of the larger issue. This larger issue is dealt with (in various ways, with various degrees of success) by various "atheistic" systems of ethics. While the subject of anger specifically may not be addressed, I am not sure that it warrants specific treatment. This is not denial so much as comparative irrelevance.

Your focus, on the other hand, really does seem to be on anger -- except when it's not. Your sections on denial and dispassion focus almost entirely on anger and very little on the target of that anger and our related ethical obligations. The section on hatred, on the one hand, and forgiveness and love on the other, focus more on other mental states that might follow anger, with secondary consideration to how that plays out in action.

But there is another level at which I find your approach confusing. In your conclusion you say:

As far as I know, Christianity is the only moral system that commands us to love and forgive our enemies...

and then,

That said, I firmly believe that (III) is the option best able to satisfy our UU metric of "maximizing happiness and minimizing suffering."

There is something of a contradiction here. UU is (or defines) a moral system, and it does not depend on or equate to Christianity. If (III) were the option best able to satisfy the UU metric, then UU would demand (III). The best you could say (if this is even true) is that Christianity was the first to recognize or promote (III). If so, it was a valuable contribution of Christianity, but not otherwise significant. (Let me be careful to state that I am not sufficiently familiar with the various religious and philosophical movements to state that Christianity was in fact the first or only moral system to command forgiveness and love towards our enemies.)

On the other hand, I am not so convinced as you that (III) is the best possible alternative, or that there are only four alternatives. But that depends on how much you are focusing on "anger" versus "moral obligation towards those who mistreat others", and whether you are viewing (III) as sufficient or only necessary.

You wanted to know how I would deal with anger, hatred and enemies, if I do at all. Please do not be angry, but I will defer my answer to those questions until a later post. ;-)


Thursday, January 25, 2007

Obviously Not A Reply

This post is part of an ongoing dialog between my friend Ernie and me about the validity of Christian belief. It is a response to Obviously (Not).


I apologize if I have confused the discussion by qualifying my support for the statement "The goal is maximize actual and potential Happiness while minimizing actual and potential Suffering." I was concerned about the effect that would have, and it was for that reason that I volunteered to continue using the statement as is, despite my reservations. As I hope to make clear (or at least clearer), I see that statement as a simplified, intuitively straightforward approximation to a more deeply grounded theory of ethics.

I had referred you to a particular post by Alonzo Fyfe, the Atheist Ethicist, in which he provided a criticism of happiness as the sole criterion by which to assign moral value. He does this by describing a hypothetical situation in which perfect happiness is guaranteed, but one which at least some people would intuitively regard as immoral. His claim is that Desire Utilitarianism can account for these cases where happiness is outweighed by other, well, desires. This in no way contradicts the suggestion that in very many cases, happiness is the strongest desire, which is why I could still view total happiness as an approximation of a more complete theory of ethics.

Desire utilitarianism, as described by Alonzo, is much more subtle and powerful than theories built on total happiness alone. (I should note, by the way, that Alonzo's critique was not of utilitarianism generally, but only a specific measure of value that is frequently advanced by atheists, just as I had done.) While I am not sure if I can explain why that is succinctly, I think its explanatory power derives primarily from recognizing that desires are the sole reasons for action, and from the positive and negative feedbacks applied to good and bad desires.

You said were confused because Alonzo appeared to you to be "specifically critiquing Utilitarian metrics for failing to adequately explain our desire" for love and truth, and further, that he would disagree that statements II and III from our MSSB "are potentially testable truth claims given the [Utilitarian] definition of moral value." First, let me repeat that Alonzo was addressing only a particular definition of moral value, not all types of utilitarianism. He is defending a different definition of moral value, but still under the broader umbrella of utilitarianism. Second, let me repeat that the point of the counter-example was not to show that happiness is not a factor in determining moral value, only that it cannot be the sole factor.

I suspect that Alonzo would agree that statements (II) and (III) are potentially testable truth claims given (X) as a metric. His claim, I believe, would be that (X) is not the best metric, and that the definition incorporated into desire utilitarianism is superior. I am almost entirely certain that Alonzo would view (II) and (III) as potentially testable truth claims under desire utilitarianism. (By the way, I am taking your version of (III), "Truth is better than Falsehood" to mean "Belief in truth is better than belief in falsehood".)

Does that help?

Moving on to my criticism of your deistic hypothesis, I apologize as well for creating more confusion there, though in this case, the seeds were planted when you wrote:

I believe that:

  1. We need a set of assumptions comparable to the MSSB in order to support meaningful "social inquiry."
  2. Using a deistic hypothesis -- that "the various systems encompassing humanity are the result of a benevolent Purpose" (one sympathetic to human Reason, Virtue, and Happiness) --- we can derive the MSSB assumptions as theorems, rather than needing to state them as axioms
  3. In science, a discrete list of facts is less powerful than a theory which explains them

[Emphasis mine.] I should have recognized that you did not mean "assumptions" so much as simply "statements" or "facts" or "propositions", since in (B) you spoke of deriving them. In fact, I described them in my response as "statements", but I probably put too much weight on your usage of the term "assumptions". Still, I stand by my claim that only (X) is necessary to support "meaningful social inquiry" (so that I disagree with (A)) and that the statements (I) - (IX) do not need to be stated as axioms (so that I disagree with (B), specifically the final clause, but read on).

Now in your last post you said:

I get that you "do not believe we need to treat (I) through (IX) as assumptions or axioms that could instead by derivable from a deistic hypothesis." [emphasis mine] My question was instead whether you agreed that we could derive something comparable to the MSSB starting from the deistic hypothesis; the last paragraph of your comment hints that you might, but I'd like to know for sure.

Let me explain how what you wrote gave a different impression. In your statement (B) you not only said "we can derive the MSSB assumptions as theorems" [emphasis mine] but also "rather than needing to state them as axioms". To me, this established a (false) dichotomy. If you had simply left off the final clause, I think your intention would have been more clear. I do see that you could have been merely contrasting one approach with another without claiming that the two together exhausted the possibilities. But then again, as I stated in the comments after my last post, the version of the goalpost statement that you were defending was that "belief in a transcendent moral purpose for the universe is ... essential ...". So again, I hope you can understand why I might have understood you to be making an exlusivist claim.

So now we return to the goalpost statement itself. You have suggested replacing the original statement with this one:

"Does the deistic hypothesis B provide a better explanation for the MSSB than an ethical metric such as X?"

The problem I have with this is that I do not think the statements in the MSSB are that difficult to explain under either the deistic hypothesis or an ethical metric like X. (I am not quite sure why the statements specifically about Christianity are particularly relevant to the deciding between a deistic hypothesis and X.) Additionally, it seems possible to me that a deistic hypothesis could in some sense include X, so that anything X explains, DHX would also explain. (Since you originally included X as a statement that DH could explain, this possibility appears somewhat likely.) In short, stated this way, I think the goalpost statement lacks discriminatory power.

I am of course open to reasons why you would disagree with this assessment. My basic position, though, is that there just is not enough meat on that bone.

Here is another perspective. Suppose we agreed on an ethical metric like X, where you consider it as part of a larger deistic hypothesis, and I considerate it the basis of a "standalone" theory of ethics. We also both agree that belief in truth is important to ethical behavior. The major relevant difference between us is what we believe to be true.

Way back in September, you advanced two propositions you would be willing to defend. We have spent the past four months on the first one, and in my view we have more or less agreed that your deistic hypothesis is consistent some core agreed-upon statements, but not necessary. Whether the DH is "more necessary" is as yet an open question, but in my view not a particularly interesting one, not unless the basis for discrimination can be strengthened.

I have always been more interested in the second proposition:

Belief in the Biblical narrative regarding God's role in shaping religious faith is as central and well-justified as belief in the scientific narrative regarding evolution's role in shaping anatomically modern humans.

Or, as you later stated it:

Epistemic dependence on received Scripture as a reliable indicator of divine will.

So, can I suggest we put the subject of ethics on the back burner for awhile and move on? And, while we have ethics simmering, can I suggest that you read Alonzo Fyfe's book, "A Better Place: Essays on Desire Utilitarianism"? As I said, I think DU provides a more sophisticated and well-developed theory of ethics than UU, and so might address some of the issues you find lacking in other systems you have looked into. I would be happy to send you a copy.

What do you think?


Saturday, January 13, 2007

Axioms, Hypotheses and Facts, Oh My!

This post is part of an ongoing dialog between my friend Ernie and me about the validity of Christian belief. It is a response to A Minimal Set of Shared Beliefs.


With perhaps one exception, which I will discuss momentarily, the statements you list as being a Minimal Set of Shared Beliefs (MSSB) are agreeable to me, though as I mentioned before, agreement on the first few is not that earthshaking. Let me repeat them here for easy reference:

  1. Happiness is better than Suffering
  2. Love is better than Hate
  3. Truth is better than Falsehood
  4. Truth is compatible with both Love and Happiness
  5. Love and Happiness are compatible with each other
  6. There are some good things about Christianity
  7. There are many things wrong with Christianity
  8. It is both necessary and possible to improve upon Christianity
  9. Reason and Empirical Observation are both key to improving upon Christianity
  10. The goal is maximize actual and potential Happiness while minimizing actual and potential Suffering

The one item that I need to qualify, surprisingly enough, is the last one. As you note, this reflects what I have described as the metric of Universal Utilitarianism, which I brought into this discussion as being very near to my own thinking on the nature of morality. Participating in this discussion has, quite naturally, led me to further reflection and study. I have, for instance, described universal utilitarianism (or at least my understanding of it) as being incomplete since it "does not reflect our intuitions about how to treat those who mistreat others."

For the past month or so, I have been reading the blog Atheist Ethicist. The author, Alonzo Fyfe, is a philosopher who advocates "Desire Utilitarianism". By this he does not mean simply that whatever you desire is good or right, or that we should do whatever anyone else desires of us. Instead, desires and actions are judged to be morally good or evil based on whether they promote or thwart the desires of others. According to Alonzo (and I tend to agree so far) this provides a better foundation for making moral judgements than happiness and suffering alone. For instance, he argues that sometimes we desire other things more than happiness, so that merely bringing about happiness would produce a morally inferior result.

I may write more about Desire Utilitarianism later after I have better digested how it all plays out. My purpose in bringing it up here is that statement (X) represents an ethical theory that reflected well my thoughts at the time I brought it up, but which may be superceded by a better theory (one with more explanatory power) later. For now, I remain happy to continue using it. As you said:

For purposes of this discussion, let us both stipulate that this moral imperative is at least "proximately true" -- that is, any ethical theory we propose has to either incorporate or address this truth to be considered valid.

However, there is another reason that this statement needs attention, a reason more directly relevent to your argument.

You suggest that the deistic hypothesis allows us to derive all of the statements in the MSSB as theorems, rather than stating them as axioms, and further that a theory that explains a set of facts is more powerful than the set of facts themselves. From this you conclude that your goalpost statement is supported.

I think there are several flaws in this argument, and the most important problem is that (X) is a very different statement from (I) through (IX). The first nine statements, possibly excluding (IV) and (V), are all statements about the value of various things: happiness, suffering, love, hate, truth, falsehood, Christianity, reason and empirical observation. Statement (X) is different because it establishes the grounds for valuing one thing above another.

I submit that (X) is sufficient as a theory of ethical value (subject to the caveat above) and further that by itself it is sufficient to support meaningful social inquiry. The remaining statements are not assumptions nor axioms. They are statements that can be evaluated logically and empirically. Statement (I) follows trivially from statement (X). Statements (II) through (V) can be supported on both theoretical and empirical grounds, once (X) establishes the basis of comparison. Likewise, statements (VI) through (VIII) can be evaluated on empirical grounds. Statement (IX) is a bit out-of-place: reason and empirical observation are part of the methodology we adopt in evaluating not just Christianity but other purported statements of fact. Still, we can use meta-reasoning and meta-observations to evaluate the usefulness of reason and empirical observation in improving Christianity (or other evaluations of value) and so place value on them directly.

In this view, we do not need "a set of assumptions comparable to the MSSB" in order to support meaningful social inquiry. We need only the single (perhaps axiomatic) definition of moral value provided by (X). The remaining statements are not assumptions or axioms; they are potentially testable truth claims given the definition of moral value. So I disagree with your statements A and B because I do not believe we need to treat (I) through (IX) as assumptions or axioms that could instead by derivable from a deistic hypothesis.

The discussion above also hints at another problem. While I agree that a discrete list of facts is less powerful than a theory which explains them, we do not have just these two alternatives. That is, it is possible that other theories could explain those same facts. In such cases, we must look for other facts over which the proposed theories differ or apply heuristics like Occam's Razor and prefer simpler theories to more complex ones. I would suggest that introducing a deity or other Benevolent Purpose to explain the value of love and truth in promoting happiness is unnecessary, and so should be discarded unless we have other compelling reasons for retaining such an element in our theories.

I hope that gives you something to work with.


Friday, January 05, 2007

Responding to Provocation

This post is part of an ongoing dialog between my friend Ernie and me about the validity of Christian belief.


I apologize for dropping my end of the conversation for over a month now. As I said in my post several days ago, there have been a number of factors contributing to my delay. It is not that I have not been thinking about it. In fact, I have been reviewing and clarifying my thinking on various issues including some that we have discussed over the past fourteen months. No earth-shattering changes, but some changes nonetheless.

Let me first respond to the post in which you enumerated a number of things we seem to agree on. I do not have any serious disagreement with your list. I was, however, a little disappointed that agreement on some of the items, particularly (a) - (c) and even (d) was unusual enough to warrant mentioning. I imagine you did so more for completeness and emphasis than because agreement on those items was surprising; at least, I hope so. It did leave me wondering if I have been communicating my views so poorly that there had ever been a question about my stance on those statements.

In your latest oblique and provocative post, you wonder if my primary goal is "truth" or "love" and if I really believe the two are compatible. Let me state categorically that they are compatible. Truth is instrumental (to use your word, thank you very much) to love. Knowledge of the truth helps us to align the consequences of our actions with our intentions.

That does not mean that knowledge leads inexorably to love. Knowledge of the truth can be used for love or for hate, for good or for evil. I do believe, however, that knowledge can lead to a recognition of the interconnectedness of human communities and that we can all lead happier lives by working with each other rather than against; that is, that love is better than hate.

One of the things that confused me about your last post was how you linked to my post on Wisconsin's Marriage Protection Amendment. Now I suppose you might read that and get a fairly negative vibe from me about the authors of the flyer. Was my post lovingly written? Well, I think it was, in this sense: I wrote out of a genuine and strongly felt concern (love) for people that will be impacted by this amendment. Was it unloving for people to denounce racial bigotry? Was it unloving for people to speak out against the debacle that has unfolded in Iraq? Is it unloving for a parent to speak against the destructive behavior or a child? Was it unloving for Jesus to denounce the hypocrisy of the Pharisees?

So having stated clearly that love and truth are compatible, let me try to answer your question about which is my primary goal. I am unsure if you mean my goal in this dialog specifically or in life generally. Generally, I do not see either of those as being the goal: both play a role in reaching other goals, especially the welfare and happiness of myself and others, both present and future. In this dialog specifically, it seems to me there is greater disagreement about what is true than about love. As well, I believe that some of the harm done by Christianity is caused by people who want to do good but their actions are based on false beliefs, and this disconnect between belief and truth causes unintended harm. In these cases, it is the beliefs that should change. So, as a tactical choice, I think it would be fair to say that my writing has been more concerned with truth.

Finally, you asked about my basis for throwing the first stone and what I could offer in place of Christianity. Why am I the one throwing the first stone? From my point of view, the course of my life has been substantially affected by the beliefs that I held for thirty-plus years of my life, including a variety of (what I perceive to be) harmful effects. Those were the first stones. Perhaps my responses have been less than ideal. Certainly they have. But even so, my concern is for the truth, for the harm caused by Christian beliefs, and for the benefits to mankind that I believe will become possible when the obstacles raised by false beliefs are removed. (Christian and other religious beliefs are not the only false beliefs to be found, but due to personal history and circumstance, they are some of the most significant to me.)

I am sorry if you have been expecting me to be presenting an alternative to Christianity. I do believe better alternatives exist and can be described. For instance, I have previously referred to A Secular Humanist Declaration seven months ago. A recent version of the Humanist Manifesto, Humanism and Its Aspirations is also on the right track. I will try to include more content describing and advocating such alternatives.


Wednesday, January 03, 2007

A Belated Examination of the Christmas Accounts

Blogging has been on hold for a month, partly for lack of inspiration and partly for busy-ness. Worse, the subject of this post is over a week past (or over fifty weeks in the future). Better late than never, I guess.

Have you ever noticed how different are the stories of Jesus' birth as told in the books of Matthew and Luke? Last year I examined one important difference, that Joseph, Mary and Jesus travel to Egypt in Matthew but to Jerusalem in Luke, but this is not the only difference. In fact, the points of commonality are far fewer than the differences, some of which are contradictory.

What are the common elements in each account? King Herod, Joseph, a virgin Mary, Jesus, Bethlehem, Nazareth, and part of Jesus' genealogy. That is all.

Elements that are unique to each story but not actually contradictory include the census (Luke 2:1-3), an angel appearing to Joseph in a dream (Matthew 1:20), an angel appearing to Mary (Luke 1:26-38), the magi (Matthew 2:1-12), the shepherds (Luke 2:8-20), stable (Luke 2:7,12,16), a house (Matthew 2:11), slaughter of male children (Matthew 2:13-18)), Zacharias and Elizabeth (Luke 1:5-25,57-80), and Simeon and Anna (Luke 2:25-38). There are some interesting things that could be said about these events, and I will mention two of them again in a moment.

First, though, let us examine some of the contradictions.

Where did Joseph and Mary live before Jesus was born? Luke makes it pretty clear that their home was Nazareth because he tells us that they had to travel to Bethlehem for the census. In Matthew, on the other hand, there is no mention of Nazareth until the family is returning from Egypt. In Matthew 2:21-23 it is fairly clear that when they went to Nazareth, they were not returning to their prior home but that they chose a new home where they would be safer from the reign of Archelaus in Israel.

I have already mentioned that in Luke, the family travels from Bethlehem to Jerusalem to Nazareth, while in Matthew they travel from Bethlehem to Egypt to Nazareth. When did they leave Bethlehem? In Luke it is clear that leave as soon as the days of Mary's purification were complete (about forty days, apparently). The timing in Matthew is far less certain. We do know that Herod asks the magi when the star appeared and that he subsequently orders all male children under the age of two killed "according to the time which he had determined from the magi." Also, as mentioned above, Matthew says the magi visit Jesus and Mary in "the house", without further qualification. Without Luke's account to tell anyone differently, anyone reading that phrase would understand it to be Joseph and Mary's house where they normally lived. So, reading Matthew by itself seems to suggest that Jesus and his family remained in Bethlehem for as long as two years, since in this account Bethlehem was their home and they had no reason to leave until the threat from Herod materialized. But while this is the plainest reading, one could attempt to reconcile the timing in several ways. First, one might say that the star appeared two years before Jesus was born in order to give the magi time to travel. Second, one might say Herod may simply "playing it safe" by killing children up to two years old. In either case, while I believe a reading of Matthew alone argues for a longer delay than Luke alone, I also suspect very few Christians will be persuaded on this point.

A third contradiction can be found in the genealogies that Matthew and Luke give for Jesus. Matthew lists the genealogy from Abraham forward to Jesus, while Luke lists the genealogy from Jesus backwards all the way to Adam (and God before him). These differences are clearly not important. What is more significant is that the genealogies are different between David and Joseph. Matthew lists 26 men between David and Jesus, while Luke lists 40. Of those, only two names are shared, Shealtiel and Zerubbabel, near the middle of each list. (Interestingly, a footnote in my study Bible says that these names do not refer to same people!)

How can the lists be reconciled? I am aware of two strategies used by apologists, though perhaps there are others; this not an issue that I have examined carefully. One suggestion is that one of the genealogies records Joseph's ancestry (since Joseph was his legal but not natural father) and the other Mary (who was his only natural parent). This appears to violate the plain reading of both Luke and Matthew, both of which explicitly list Joseph as the penultimate person in the genealogy.

The other approach has been to claim that both genealogies are Joseph's, but that Matthew skipped some generations. I am not sure how those who hold this view resolve the fact that each list of names is so different. It is not enough to suggest that in each list two different names are used to refer to the same person since, particularly near David's end of each list, the names listed are known to refer to different people. Allowing Matthew to skip generations creates other problems, because Matthew is careful to point out that there were fourteen generations from Abraham to David, fourteen from David to the Babylonian captivity, and fourteen from the captivity to Jesus. What meaning can this symmetry have if it is attained simply by skipping over generations whenever it is convenient?

(It is also not feasible to assume that Luke also skipped generations so that between the two we get a complete list. Not only would such a careful mismatch strain credibility by itself, it would require 66 generations between David and Jesus, so that, on average, each son would need to be born when his father was fifteen years old. Let me be clear that I am not aware of anyone who actually claims this is the resolution to the contradiction. Just sayin'.)

So there are as many as four contradictions between the two stories, some more certain than others. Now, let's return to the story elements that are neither shared nor contradictory. Of these, we have no separate basis for knowing anything about the angelic encounters, or the place where Jesus was born, or the shepherds. Conceivably, the various characters mentioned only by Luke (Zacharias, Elizabeth, Simeon and Anna) or even the magi in Matthew could have been mentioned in some kind of contemporaneous records, though I am not aware of any. (Josephus does mention John the Baptist, but I do not know if that includes his parents.) In any case, history's silence in those cases would not be surprising so it is hardly a basis for comment.

In the case of Herod ordering the slaughter of all of the male children under two years of age in and around Bethlehem, that would have been an event worthy of comment. As Richard Carrier notes, due to the life expectancy at the time, this would have represented approximately 10% of the population in the area. Nobody who wrote about Herod mentions this, even though it would be highly noteworthy, so the lack of mention makes the actual historicity of the slaughter questionable.

The other historically significant event mentioned was the census. Here the problems are more substantial. Luke claims the census took place while Herod was alive and while Quirinius was governor of Syria. Herod died in 4 B.C. Quirinius did not become governor of Syria until 6 A.D. (If Luke is referring to Herod Archelaus instead of Herod the Great then this is another contradiction between Luke and Matthew, since Matthew is clearly referring to Herod the Great.) Caesar Augustus did order a census of Judea, which was conducted by Quirinius, but this was not a census of the entire Roman empire or the entire inhabited earth. (An interpretation of Luke 2:1 consistent with this fact is not particularly unreasonable.) Some skeptics have questioned the need for Joseph to return to his ancestral home for the census, but apparently this was not so uncommon. More troublesome is why he would bring Mary along for the journey when she was expecting to give birth at any time. Why not leave her in the care of friends and family in Nazareth?

The answer to that question is, I think, the answer to many of the questions raised by the contradictions and other features of both accounts. Both Matthew and Luke needed to provide an explanation for Jesus being born in Bethlehem (as required by prophecy) while being raised in Nazareth. Matthew chose to represent Bethlehem as their original home and then manufactured a reason to leave and eventually settle somewhere else. Luke chose to represent Nazareth as their original home, so he needed to manufacture a reason for Mary to travel to Bethlehem to give birth, at which point they could return home to Nazareth.

Matthew's solution to the problem allowed him incorporate two (supposed) fulfillments of prophecy, one regarding the murdered children and one regarding God calling his son out of Egypt, though in neither case do the "prophecies" appear to be prophecies at all. He also is able to incorporate the "dangerous child" motif into the story, a theme that is very common in literature and mythology. But this solution required the introduction of a significant ahistorical event in addition to Matthew's tendency to manufacture prophetic fulfillments where no prophecy was intended.

Luke's solution, on the other hand, is far more plausible. There are some questions about the timing of the events he describes, and the inclusion of Mary on an unnecessary journey while in the final days of her pregnancy is rather suspicious, but so long as one is willing to accept that Jesus was born in 6 A.D. the difficulties are far smaller with Luke than with Matthew.

The Christmas story as told in Matthew contradicts the story as told in Luke in almost as many points as they agree. On the whole, Luke's version is more plausible. I consider it most probable that both accounts are almost entirely fabricated, but the account in Luke is sufficiently consistent internally and externally that I would not expect to convince anyone of that on the basis of the story itself, but only as part of a much larger evaluation of the trustworthiness of Luke.

Richard Carrier has written a far more detailed examination of the dating issues The Date of the Nativity in Luke, which also touches briefly on the other points of contradiction between Matthew and Luke. Mostly, though, he examines various suggestions that have been put forward for reconciling the dates implied by Matthew and Luke.