Here's an important review of a fairly recent book on the OT canon:
By the end of the first century, he [Lim] concludes, there is a rabbinical canon of the Pharisees, which is not closed until sometime between 150–250 c.e..
Building on a theory first proposed by John Collins about two decades ago, Lim argues that the canon represents a political triumph of the main sect within Judaism that survived the tumultuous post-70 c.e. years within Palestine. The Pharisaic party represented the majority of Jewish survivors from the Roman holocaust and as a result their collection of authoritative texts became the canon. Other collections of authoritative literature simply perished since the sects or groups associated with them did not survive. The resulting canon was that of the victors.
This raises several issues:
i) We need to distinguish between the date at which that collection became the standard canon for the Pharisees, and the date at which that collection became the standard canon for Jews in general. Even if we grant that the Pharisaic canon only became the official canon of rabbinical Judaism in the 2-3C AD, that canon antedates 70 AD. The Pharisaic canon preexists its dominance. Its origins go back to an earlier time. So the date of the Pharisaic canon is much older than the date at which it became dominant–even on Lim's construction.
ii) According to Lim, the Pharisaic canon became the official canon by default. It was the last man standing after the dust settled (as it were). The rival canons of rival Jewish sects perished when the sects that sponsored them perished.
Whether you think that's a problem depends, in part, on whether you think the canon is just a sociological phenomenon or historical accident. In other words, if methodological atheism is your frame of reference, then which canon won or lost is the luck of the draw. The victorious canon has no intrinsic authority in contrast to rival collections. It isn't special, isn't more deserving, then rival canons that perished.
If, on the other hand, you believe in divine providence, then might be God was using the historical process to winnow the wheat from the chaff.
iii) There's some ambiguity in referring to other collections that perished. If they weren't preserved, then how do we know that they differed from the Pharisaic canon? How do you determine the content of a collection that didn't survive?
iv) There's nothing necessarily suspect or unsettling about the existence of rival canons. For instance, in church history you have heretical groups that produce their own canon (e.g. Gnostics, Mormons, Swedenborgians, Christian Science). They may not reject the received canon outright. Rather, their sectarian literature supplies a filter that reinterprets the received canon.
That's no reason to doubt the legitimacy of the received canon, or the illegitimacy of competing canons. Rather, that's to be expected. There's a perennial tug of war between truth and error, orthodoxy and heresy.
Moreover, the only reason that this Pharisaic canon remains open is because there remained a question about the authority of certain books. But there have always remained questions about canonical books and this need not imply an open canon.
That point is often overlooked in discussions of canonicity. The fact that ever book in the received canon aren't equally well attested doesn't mean the canon is open. The closure of the canon creates a boundary between books inside the canon and books outside the canon. But that doesn't mean all books inside the canon enjoy the same level of evidence or theological significance. The canon can have "border" states. Yet documents outside the canon may have even weaker claims than the weakest claimants inside the canon.
This evidence confirms the essential thesis, but it needs to be emphasized that from the various collections there was no unilinear progress from the many collections to the one canon. “Rather, there were the many collections and then there was the majority canon. Once sectarianism disappeared, so did the variety of collections” (p. 186).
In other words, you don't have a general evolution towards official collections. Rather, certain collections are already in place early on. It's just a question of which collection or whose collection. As OT books were being composed, you'd have a growing canon. Collections of collections, as a later collection incorporated the former collection, but updated that collection to include newer books. But once all the "OT" books were written, that process would naturally come to a halt.
The reviewer then makes a number of other worthwhile observations:
If Scripture itself is used to help determine the authority of biblical books, why not at least consider some other evidence within the text itself, e.g., that Chronicles begins with Adam, who initiates Genesis, and ends with a quotation at the beginning of Nehemiah, thus comprehending the entire canon in summary form. Moreover, many scholars now recognize the extent of canon-conscious editing of the biblical text, in which superscriptions have been added to books stressing divine authority, and also editorial additions which organize collections of books.
I am left with some other misgivings about the book. First, Lim claims that there is no evidence for a temple library or archive, which would have contained a collection of canonical books. But there is no question that sacred space in the Hebrew Bible itself was a location for sacred texts. Lim's description of the scroll during Josiah's time as a book of reform and not a canonical book is questionable (pp. 32–33). Would a book of reform cause the king to rip his clothes in grief? The fact that this book was used to institute widespread reform in Judah shows its authority. Moreover, the fact that “canonical books” were not popular or were abandoned or lost may say something more about the people at the time than the books. On the other hand, in times of spiritual renewal, I find it difficult to accept that a religion which revered the holy words of God would not have had a special place for the creation, preservation, and transmission of divinely inspired documents in its holiest sanctuary. The books which later made up the Hebrew Bible itself cry out for such an explanation. Where else would there be the necessary infrastructure for their production and their preservation? In this regard, a recent important work by Tim Stone notes the coincidence of lists of canonical books after the destruction of the temple. There was no need for listing them before since enumeration and order were assumed.
Second, what might be said about the evidence of biblical manuscripts from Qumran? The majority of them are proto-MT manuscripts. How does one explain this? Where does this tradition come from which reflects the text type of the majority canon—the canon of the winners? Emmanuel Tov has argued in the past that such a text type probably derives from scribal circles associated with the temple, and this of course implies canon. This makes a lot of sense. Lim questions why a rabbinic tradition which mentions the authoritative function of standard Torah scrolls in the temple for establishing readings for the Torah might infer canonization. He concludes that “in establishing a standardized text, they were not fixing the extent of the scriptural collection” (p. 34). But this is to confuse the effect with the cause. Why would temple scribes be concerned with text-critical matters for these books? Probably because there already was a scriptural collection. Moreover, what about pre-first century c.e. Greek manuscripts which have been corrected to the MT? Does this not reflect the importance of a particular text type, which itself implies canon?
Finally, it is worth observing that in early Christian conflicts with Judaism there is never any debate about the extent and the content of the canon, only its meaning. In my judgment this is telling.