It's probably no secret that I greatly appreciate the work of J.A.T. Robinson, specifically his Redating the New Testament. I've been thinking lately about its strengths and weaknesses, and thought I'd write some of them here, beginning with weaknesses.
1) The Significance of 70: an early chapter in Redating goes by that title. Robinson argues, I think convincingly, that Jesus's predictions of the temple's destruction in 70 do not constitute a terminus post quem (time after which) for the composition of the gospels. I think that he perhaps pushes the matter of 70 too far, though. He argues that the absence in the NT corpus of any incontrovertible reference to the events of 70 gives us reason to think that the NT as a whole dates prior to 70. I feel that this might be a somewhat illicit argument from silence. Admittedly, he does bolster this by observing that such prophetic material as the Olivet discourse (cf. Matthew 24, Mark 13, Luke 21) contain prophecies that are somewhat at variance with the known course of the First Jewish War (c. 66-73), which presents some difficulty for those who would argue that the text was written to conform with the events of those years. Given though that I see the text as more concerned with conforming itself to the Jewish prophetic tradition than to current or recent events, I'm not convinced that that would demonstrate that such variance indicates a pre-70 date either. My own feeling is that the prophecies of the destruction of the temple are probably non-probative, neither establishing a terminus post quem or terminus ante quem. That having been said, Robinson's argument does lead me to consider the following possibility: if, as the consensus chronology holds, the majority of the NT dates to between 70 and 100, and given the virtually complete absence to incontrovertibly mention of the events of 66-73, we should perhaps conclude that the events of that period simply didn't have too much impact upon the churches that produced our NT literature. If we conclude thus, then we probably also have to dispense with the argument that the eschatology that we find in things like the Olivet discourse could not have originated but in the context of the First Jewish War.
2) A Fixation with Nero: Robinson rightly objects that the reign of Domitian has become a sort of dumping ground for NT texts, especially those that mention persecution. Best I can tell, this practice began with Lightfoot. Of course, we now know that the 19th-century's understanding of Domitian's persecution is probably mistaken. Robinson is right in arguing that scholars have made too much of the events of his reign. But Robinson I think replicates this error to a large extent, by assuming that virtually any reference to persecution, unless it clearly indicates otherwise, must refer to the Neronian persecution. He replicates the error because the problem with ascribing so many texts to the reign of Domitian had not simply to do with misunderstandings about his persecutory activities, but also to do with the fact that in many cases the references to persecutions are sufficiently vague that they could refer to any number of moments in the first century of the Christian movement. In fact, Robinson's own argument with regard to the events of the First Jewish War potentially could apply here: just as there are no incontrovertible references in the NT to the events of the First Jewish War, apart from probably Revelation there are no incontrovertible references to the events of the Neronian persecution. I am inclined to think that Robinson's Nero is too big.
And now some strengths.
1) The Pastoral Epistles: Robinson was probably ahead of his time in realizing just how weak the arguments against their authenticity truly are. Let it be clear: I'm not arguing that they are authentic. In fact, I'm not here arguing anything about their status. Rather, I am noting that we are becoming increasingly aware that many of the classical arguments, such as from style, or from ecclesiastical development, or from inability to fit into the Acts narrative, are astonishingly weak. One can still argue that these texts are pseudepigraphical, but one needs to find better grounds for such an argument than the classical arguments (cf. Campbell's recent work in Framing Paul, which responds to this challenge with far greater sensitivity than most arguments against the authenticity of the Pastorals).
2) The Approach to the Data: quite simply, Robinson considers a greater range of data than most. He has no qualms looking at Eusebius. And why not? If we can consult Josephus in our reconstructions of periods two, three centuries before his time, then why can't we consult Eusebius in our reconstructions of periods two, three centuries before his (especially when much of our consultation actually consists of looking at his quotes from no-longer-extant second-century texts)? So, Robinson, properly cognizant that there is in fact no good reason to ignore Eusebius and other Patristic writers, spends his time consulting their works rather than showing why this or that text that would appear to be relevant is not, or worse, simply ignoring them. Such data informs his historical judgment, and as such he is working with more information. And working with more information, he has simultaneously more material to inspire and greater controls to govern his historical imagination.