Gavin Evans, There was no Jesus, at "Aeon", provides a very good example of the sort of pseudo-history that is found in Jesus-Mythers, so it's worth taking a moment to consider how historical evidence works, the historical evidence in the case of people like Jesus and Socrates (the latter being a comparison case brought up by Evans), and the problems with the general kind of argument that Evans is attempting to make.
Historical inference is fundamentally based on a few general inferential patterns. The first, and by far the most important, of these, is causal. We have an effect; from the effect, we conclude the cause. In historical matters, the effects are usually things like texts and monuments. The kind of causality is testimonial (in a broad sense in which monuments as well as texts testify to something). Also quite important is the fact that in historical matters the causes are usually causal chains and networks of various and indeterminate kinds. Because of these facts, an immense amount of historical scholarship is devoted to study of possible defective causes in the chain.
In a causal inference involving testimony, we have an effect that does not merely indicate its cause but says something (directly or indirectly) about it. If I pick up a newspaper and read about something happening in the Persian Gulf, I have an effect, the newspaper with the article about the event in the Persian Gulf. For this to be at hand, it had to arrive there; it had to have been printed; and there had to be someone or some group of people who wrote the article. Now, if I take the article at face value, it tells me something about its own causal chain. In particular, it tells me that, if I trace back the causal network that led to its being read by me, among the causes in that network, I will find the newspaper office, the reporter, the sources mentioned in the article, and (most importantly) the Persian Gulf event itself.
Of course, we know that while testimony often informs us, it also often leads us astray. Whenever it does so, it is because the causal network is defective for some reason. There are many reasons or causes that make testimony defective. For instance, someone could have printed up the newspaper as a joke, a satire, an attempt to confuse me, or any number of other things. The newspaper may have made a mistake in printing the story (e.g., a search-and-replace function accidentally changed 'Red Sea' to 'Persian Gulf' right before it went to print). The editor may have revised the story badly or with malicious intent so that it appears to say something that was not intended. The reporter may be misinformed, or deluded, or lying. The reporter's sources may be misinformed, deluded, or lying, or they may have misinterpreted some other event, or they may have jumped to conclusions, or any number of other things. So again, there are many possible defective causes that make a testimonial causal network faulty.
Crucially, they are not equally likely. It is much more likely that the reporter is misinformed than that the reporter is deluded; it is much more likely that the reporter made a mistake than that the entire newspaper did; it is much more likely that the newspaper is genuine than that someone is trying to mislead me. Some of these differences in probability are due to the need to be consistent with other causal inferences. For instance, I might find the story in other newspapers, and so need a causal explanation that can account for more than just this particular newspaper. Much of this coherence of causal explanation is probabilistic, but causal inferences of the right kind can even rule out particular defective causes entirely, just as they can sometimes establish with certainty that there is a defect. But there are two other major kinds of inferential patterns that also have a large effect on assessing how likely or unlikely defective causes may be. The first is extrapolation by analogy; the second is interpolation into a profile. In analogical extrapolation, of course, we are reasoning on the ground that similar things happen elsewhere; in profile-interpolation, we are reasoning that something coheres with an overall pattern. In the newspaper case, genuine newspapers reporting on events are relatively common in our experience but fake newspapers intended to mislead are rare; in a reputational field like journalism there are penalties for getting caught lying, so reporters usually have at least some incentive not intentionally to report easily discoverable falsehoods, and a reporter who is caught lying once is often found to have been doing it either for a very specific reason (like money) or habitually; certain kinds of events tend to happen in the Persian Gulf (or so other testimony has suggested) and other events tend not to do so; if this event happened, it may make sense or not depending on other things I know or think I know about what is happening in the region. None of this is definitive, but cumulatively extrapolation and interpolation play a significant role in historical reasoning.
The most perfect testimony occurs when the combination of these kinds of inferences manages to rule out any relevant defectives, but testimony is still very good as testimony if even the more likely possible defective causes are not very likely at all. Much testimony is a little more spotty than this -- there are often gaps in our knowledge such that particular kinds of defective causes are still seriously possible. Human beings in general and historians in particular still use such testimony extensively, but these kinds of situation are precisely one of the things that historians look for, because it is by dealing with such issues that progress occurs in a field like historical scholarship.
Suppose we take someone like Socrates. Nobody proposed as a purely abstract hypothesis, "Maybe there was a guy named Socrates in Athens during the Peloponnesian War." We start with what the texts and monuments seem to say. We have lots and lots of texts that refer to Socrates. Those texts are often based on other texts, and so forth, and this goes back a very long way. The causal networks here are immensely complicated -- we have different manuscript traditions, all of which have their possible defective causes, and those go back to originals that had an original context. But some of these texts, as part of their own testimony, or due to the testimony of other texts, go back to people who are testified to have lived more or less at the time: Aristophanes, said to be a contemporary of Socrates; Plato, Xenophon, Antisthenes, and a few others, said to be students of Socrates; Aristotle, said to be a student of Plato; and so forth.
An interesting feature in this case is that, if this testimony is taken at face value, many of the texts that refer to Socrates are fiction. The earliest extant record of Socrates (again, if the testimony is taken at face value) is Aristophanes' The Clouds, which is quite clearly a fictional work and has always been taken by everyone to be so. Both Plato and Xenophon give us works referring to Socrates that they quite clearly did not expect their readers to take as historical facts. Plato's Apology and a few of the death dialogues and Xenophon's Apology apparently purport to describe historical events, and it is in each case historical events that they did not themselves experience -- according to the testimony we have, Plato was too ill at the time to be a participant, and Xenophon was in exile and attributes all his information about it to Hermogenes. Both Plato's and Xenophon's versions of Socrates's defense speech show signs suggesting that they are not independent accounts; one of them knew of the other's Apology, and it's just a question of which came first. The earliest testimony that we have that avoids these problems is that of Aristotle, who wasn't even born when Socrates is said to have died. And all of our evidence boils down to this: Aristophanes's fictional story, Plato and Xenophon (much of whose testimony is clearly fictional and whose most historical-seeming testimony is about events to which they were only indirectly connected), a few fragments attributed to other students like Antisthenes (whose first definite traces in history are much later), and Aristotle's passing mentions.
And yet there is no serious doubt that Socrates existed. Our attributions for most of these texts are reasonably well founded. The Clouds is a satirical comedy, which makes the likeliest possibilities either that Socrates was real or was dreamed up by Aristophanes to spoof a certain class of sophists, but the jokes that are made are more like the jokes you would tell of somebody people already know than jokes you would make of a purely fictional character, to such an extent that even if we only had The Clouds, we would have excellent reason on the basis of it to conclude Socrates actually existed despite The Clouds being a purely fictional work. But we don't just have The Clouds to explain; we need a causal network that explains the references of Aristophanes and Plato and Xenophon and various fragments and Aristotle. Plato and Xenophon even in their fictional works associate Socrates with figures we have independent reason to think existed; Aristotle clearly assumes that Socrates existed and was not made up by a group of people passing themselves off in fictional works as Socrates's students. But more than that, there is a massive shift and change in the very nature of philosophy between the sophists and the rise of the Socratic schools of philosophy that needs to be explained, and just saying, "Well, a bunch of people got together and decided to write works about a fictional character named Socrates" doesn't explain anything about it. There are so many ripples in the lake, there has to be something that disturbed it, and the texts themselves say that it was Socrates.
Much, much more could be said about this, but let's shift to Jesus. Without question, the earliest evidence we have are the Pauline letters; we then have various other New Testament works; we have the Apostolic Fathers, particularly Ignatius and Clement; we have traditions recorded by Irenaeus and indirectly by Pliny the Younger and Tacitus. Much more disputed is a reference in Josephus's Antiquities. The testimonial causal networks for all of these are quite complex, and probably the most closely and intensively examined testimonial causal networks in historical scholarship. And we have, of course, the existence of Christianity itself, which is an effect that actually ends up being of quite considerable importance in this case.
Paul represents himself as a follower of Jesus writing with a bunch of other followers of Jesus to yet other followers of Jesus; at least some of the latter are communities that he represents as not originating with himself but with yet other people. He is clear that Christian communities, particularly that at Jerusalem, pre-existed his even becoming Christian. It is clear in several ways that Paul is not the only authoritative member of the entire movement, and indeed, at several points he has to defend his having any authority at all. Paul repeatedly refers to Jesus as having been crucified, claims he was betrayed, identifies a ritual that is attributed to Jesus originating it, and claims Jesus was the son of a woman and a descendant of David. Thus the purport of the testimony of Paul is that Jesus existed. Given that Paul's letters were preserved, there must have been communities that preserved them, which makes it very unlikely that the addressing of letters to already existing Christian communities is purely a fictional device; the fact that we have the letters at all is an effect that needs to be considered in our causal explanation. Given that, it is unlikely that most of the people explicitly named in the Pauline letters are fictional. Thus already, less than thirty years after the alleged death of Jesus, there are multiple communities around the entire Roman Empire acting as if Jesus had existed, an effect that needs to explained, and is most easily explained by assuming that there was someone named Jesus who existed; every other explanation runs into problems with explaining how the Christian movement actually ends up clearly existing so early after the alleged death of the one whom the movement explicitly treats as its founder. It's worth underlining that Paul, despite being later, the Pauline letters, even the stripped down 'authentic Paul' of most liberal scholarship, are in and of themselves excellent reasons to think that Jesus probably existed. They are certainly better evidence of Jesus than Aristophanes is of Socrates, because it's even more obvious that Paul cannot himself be simply making Jesus up, since there's no reason to think the letters are intended to be fiction and it's obvious that if the Pauline letters are authentic at all that they are addressed to people who already knew what Paul was talking about and who did not get all their information from Paul; they are arguably better evidence than either Plato or Xenophon alone (although Plato and Xenophon together are a different matter -- but the relevant analogy in that case would be to Paul and all the other New Testament authors). Is this definitive? No, hence the 'probably'. Can you make a just-so story resulting in the Pauline letters that does not imply that Jesus existed? Certainly, because we cannot definitively rule out all the defective causes in the apparent testimonial causal network; but it would in fact be a just-so story, certainly less supported by the actual evidence we have than one in which Jesus existed.
But here's the thing, Paul is not our only source. We have all the other New Testament writings; they are later, yes, but they are, each alone and all together, effects that need an adequate cause. The New Testament writings were not uniformly preserved; different communities preserved different collections of writings, so we have good reason to think that the whole New Testament was not a single package. And several of them are different enough from Paul that they seem to represent different traditions. We have the Gospels, which purport to tell us about Jesus's life and associate Jesus with people we have independent reason to think existed, like Pontius Pilate. The testimonial causal networks for each are complicated and there's a lot we don't know. They are not completely independent; but the dependence itself tells us more than it might seem -- virtually all of the possible permutations of the Synoptic Problem have been considered, and in every single one we get multiple traditions for each that have to be earlier than the Gospel itself. For instance, if we assume Markan Priority, Luke and Matthew depend on Mark; but neither of them can only depend on Mark. In some cases, Matthew and Luke agree, or almost agree, independently of Mark. Thus on Markan Priority there are at least five lines represented by the three Synoptic Gospels: Mark, at least one source for Mark, at least one source of the Luke-Matthew agreement (often called Q), at least one source for things distinctive to Luke, and at least one source things distinctive to Matthew. Similar things happen regardless of what solution you give to the Synoptic Problem. (This is why Evans's complaint that Q might not exist is irrelevant; sure, but that's because it's only required if standard Markan Priority is assumed. Change the solution, you get a different set of apparent causal lines. But in every single case, you get multiple lines.) Whichever solution is accepted, the multiple lines are effects that need a cause, not just individually but also in the fact that there are multiple lines. And the simplest explanation is, again, that there were multiple traditions about Jesus, and that there were multiple traditions about Jesus because Jesus existed.
And so it goes. You can just-so any particular evidence, but the explanation for each that is most probable is that Jesus existed, and the explanation for having multiple evidences that is most probable is that Jesus existed. There is just no room for doubt; all the multiple effects we have point to Jesus having existed, and none point to Jesus being fictional. We see some of the latter aspect in the fact that Evans keeps bringing up the dating of New Testament texts as if it cast doubt on Jesus existing, but we can only even hypothesize dates for any of the New Testament writings by coordinating at least some of them as being at least reasonably historically accurate for at least most things. No date for any New Testament text survives the assumption that all the texts are false; every date requires us to assume that there was a fairly extensive Christian community from early on. If you are going to hypothesize dates, you can only do so by assuming things that are most easily explained by Jesus having existed. Again, the fact that this is not immune to just-so storytelling does not change the overall tenor of the evidence. You cannot reject a conclusion based on the most obvious reading of the evidence in favor of a hypothesis that is directly supported by none of the relevant evidence. In history, as in Newtonian physics, you don't evade induction by feigning hypotheses; if you want to claim that all the evidence on the table is misleading, you need the causal inferences giving you reason to think it so, not the bare fact that you can make up a story in which it might happen, which proves nothing at all beyond that you can make up stories. And you have to cover it all, because all it takes to have good reason to think that something or someone exists is one good causal inference.
And all of this is without even considering even more elaborate extrapolations and interpolations beyond those simply associated with the texts themselves, like the fact that in most other cases in which we have a religion that claims to have a historical founder living at a particular time, there does in fact seem to be a historical founder living around that particular time. No other messianic movement in Judaism itself has ever been so self-originating as to base itself on devotion to a purported Messiah who was a fiction the movement itself completely made up. If Christianity were different, that is a difference that would also need to be explained. There is no need to make a 'case' that Jesus exists; the likelihood of his not existing on the evidence we have is so low that we can take it as morally certain that he did.
This is, I reiterate, not an exhaustive look at the evidence; there are so many things that would have to be explained if Jesus didn't exist that they easily narrow down the possible conclusions to the point of reasonable certainty. Even on a lot of very skeptical assumptions, like the assumption that most or even all the New Testament is pseudepigraphal and fictional, the most likely conclusion is that it is fictional pseudepigraphy by a community that traces itself back to an actual man named Jesus.
So, let's turn to Evans's argument and see some of the problems that he doesn't solve in trying to make the argument.
(1) As is very common in these kinds of history-skeptical arguments, Evans fails to grapple with one of the major difficulties that has to be faced by anyone doing historical work of any kind: time does a number on your evidence. People leave extensive traces in the world, but those traces begin to deteriorate almost immediately, and it all will tend toward vanishing. Traces don't deteriorate at the same rate, which is why, for instance, ancient monuments are so important. It is also why various accidents of preservation loom so largely in ancient historical work; these accidents of preservation, like being buried so that oxygen, moisture, and bacteria can't easily break things down, become our primary connections to the past the farther back we go. One of the effects of this is that as we go back, it becomes less and less certain how representative our evidence is; and since arguments from silence depend at the very least on knowing that the evidence we have is representative, the more difficult it is to build a successful argument from silence.
Our best evidence suggests that Euripides wrote over ninety plays. We have nineteen full plays, one of which we don't know for sure is authentic, and a few scattered fragments; a few of the full plays survived in an anthology (often called the Select plays or the School plays) and most of the rest are from one volume of an alphabetically arranged multivolume collected works of Euripides (often called the Alphabetical plays), which is why nearly half of them begin with H or I. Because of this, it would be very difficult to argue that Euripides never considered a theme or responded to an event; if our extant evidence doesn't indicate that he did, that's all we can say, that our evidence doesn't indicate that he did. But it doesn't follow from the fact that our evidence doesn't touch on something that Euripides didn't. We can try to bolster the argument by extrapolation or interpolation, but it's not going to get us very far, and certainly not with certainty. This is a common issue with using arguments from silence in ancient matters.
Evans tries to lay the groundwork for the argument with an argument from silence which fails conspicuously to wrestle with this problem:
You’d think that a cult leader who drew crowds, inspired devoted followers and was executed on the order of a Roman governor would leave some indentation in contemporary records. The emperors Vespasian and Titus and the historians Seneca the Elder and the Younger wrote a good deal about 1st-century Judea without ever mentioning Jesus. That could mean simply that he was less significant an actor than the Bible would have us think. But, despite the volume of records that survive from that time, there is also no death reference (as there was, say, for the 6,000 slaves loyal to Spartacus who were crucified along the Appian Way in 71 BCE), and no mention in any surviving official report, private letter, poetry or play.Would you in fact be reasonable to think that, though? If the Gospels are right, Jesus' public ministry lasted only about three years. It occurred entirely in Galilee and Judea, the latter being a semi-autonomous province on the edge of the Empire. The sources we have that talk about Judea are often clearly not concerned with giving us detailed information about the Judeans themselves or their various religious differences, and the source we have from the period that does make the most effort to do this, Josephus, gives us a picture of Judea in a considerable amount of religious ferment at the time (and is also one of the sources that might have indirect reference to Jesus, but let us assume that the references are later glosses by Christians that made it into the text). But the biggest problem is that there is no "despite the volume of records that survive from that time"; the original records we have surviving from that time are a tiny proportion of the records that actually existed, and most of what was happening in the Empire we can only reconstruct by extrapolation and interpolation, some of which is very shaky. Compared to most ancient events, we do know quite a bit about what directly concerned the emperors and the city of Rome; but even there, we're quite patchy. A lot of what we know about most of the internal politics of Rome itself from this period depends entirely on Tacitus and Suetonius, each of whom selects his matter with his own ends in view. In the broader Empire, there are plenty of people whose names are on monuments, and thus were undeniably of some local importance, about whom we know practically nothing else. We can be reasonably sure that there were lots of important people in the Roman Empire about whom we know nothing at all. An argument from silence to a particular conclusion won't work here; we can't establish the precondition for making such an argument work: that if it had happened, it would probably be in the surviving evidence. The most we can say is that the Roman sources that have survived don't register it, not that it didn't happen, or even that it wasn't registered. A serious historian might be disappointed (being a historian requires a willingness to be disappointed by the state of the evidence), but would hardly try to build a direct case on it.
We run into a similar conspicuous failure to face the nature of our information about the ancient world in another argument in discussing Paul; but it also touches on another problem with this kind of argument, so I'll turn to that.
(2) Skeptics of historical events and personages regularly exaggerate when it comes to describing what the evidence tells us. For instance, Evans says:
If Jesus lived and died in Paul’s lifetime, you might expect he’d refer to Jesus’ ministry on earth – to his parables, sermons and prayers – and that his readers would want this crucial life story. But Paul offers nothing on the living Jesus, such as the stories or sayings that later appear in the gospels, and he provides no information from human sources, referring only to visionary communication with Jesus and to messianic Old Testament quotes.This is an argument from silence, and again we cannot establish the precondition required for making such an argument. The Pauline letters themselves indicate that we do not have all of Paul's letters, and they regularly make reference to things that Paul preached in person or took his readers already to know without giving us any detailed run-down of what they were. Nor are letters normally places in which we go into detail about things that the receivers would already have to know, and all of Paul's letters are addressed to communities that are already Christian. But more seriously, the claim is false. Paul does say things about the living Jesus. We learn that Jesus existed and was Jewish (Romans 1:3), that he appointed apostles, and that he was crucified, having been betrayed. 1 Corinthians explicitly attributes sayings to Jesus -- 1 Cor. 7:10-11 attributes to him a saying on marriage that is paralleled in other sources, and 1 Cor. 11:23-25 attributes to him words on the night when he was betrayed. If you add 1 Timothy, we learn that he was crucified under Pontius Pilate and we have a case in which the Jesus seems to be quoted without explicit attribution (1 Tim. 5:18, which parallels Luke 10:7). There are other cases that are less obvious but obviously would have to be at least considered; for instance, 1 Thessalonians 4:13-5:11 possibly preserves echoes Jesus' eschatological sermons and parables -- it's attributed to "the Lord" and has undeniable parallels to Jesus' preaching in the Gospels. There is no reason why one would suppress this obvious evidence. 1 Corinthians and 1 Thessalonians are almost universally considered authentic; 1 Timothy is often considered inauthentic, but even that is definitely not a universal consensus, nor is it certain enough that 1 Timothy can be simply ignored. It may be less, and sometimes less certain, than a historian would like, but that is true of everything in the ancient world, and it is just false to say that "Paul offers nothing on the living Jesus".