When Was the Psalter Compiled?

In recent research, some scholars have proposed extremely late dates for the final formation of the Hebrew Psalter, even as late as the 1st century CE. By amassing evidence from the manuscript tradition, ancient translations, secondary reception and reuse, as well as book history, I argue that the basic framework of the traditional, five-book Psalter is rather the product of the late Persian period (5th–4th centuries BCE). 

See also Scribes and the Psalter: Situating Ancient Psalm Collections in Their Scribal Contexts (Mohr Siebeck, forthcoming in 2026).

By Drew Longacre
Research Fellow
University of Münster
September 2025

Introduction

            Dating an ancient anthology entails numerous complexities, not least of which is the relationship between the parts and the whole. According to the Babylonian Talmud (b. B. Bat. 14b–15a), David compiled the Psalter from Psalms written by himself and ten other elders. According to a fragment attributed to Hippolytus of Rome (Fr. Ps. 1), the learned tradition attributes the compilation of the Psalter to Ezra. In the 19th and early 20th centuries it was common to date many Psalms to Hellenistic (332–63 BCE) or even Hasmonean (152–37 BCE) times, but by the late 20th century a consensus was emerging that the Psalter was a product of the Persian period (539–332 BCE). Recent specialist studies on the fluidity in contents in the Dead Sea Psalms scrolls have pushed back in the opposite direction (e.g., Wilson 1985; Flint 1997; Pajunen 2014; Mroczek 2016; Willgren 2016), sometimes supposing that the contents of the Psalter did not achieve a fixed inventory and order until as late as the 1st century CE. The historically proposed dates of compilation thus span a range of over one thousand years, and there remains no secure consensus among modern critical scholars. This all raises the question: can we really know reliably when the Psalter was compiled? And, if so, when? 

Dates of the Individual Psalms

            Since the whole cannot be earlier than any of its parts, it is critical to rightly assess the dates of the individual Psalms, especially the latest ones. This is rather difficult, however, since the Psalms—which are largely tailored for general cultic performance and reuse—are notoriously lacking in references to particular, datable, historical events and individuals. Only in the editorial paratexts of the superscriptions do we find secondary traditions associating Psalms with named individuals and known events. But here it is not entirely clear whether these were originally meant as strict claims to authorship, and modern critical scholars generally do not assume they accurately record the names of the Psalms’ original composers.

            A few Psalms do contain fairly clear references to major historical events, which suggest exilic or post-exilic dates. Psalms 74 and 79 refer to the Babylonian destruction of Jerusalem and the temple. Psalm 137 mourns the Babylonian exile. Other Psalms could potentially be read in light of the return from exile (e.g., Psalms 107 and 126) or later events in Hellenistic or even Hasmonean times, but there is no scholarly consensus on these situations, and so it would be hazardous to base too much on them.

            Psalms scholars often attempt to date the Psalms based on ideas and cultural references reflected in the texts. In many cases, it seems that Psalms fit best within the temple cult of pre-exilic Judah with its Davidic kings. In other cases the Psalms seem to reflect later socio-political realities. Often, however, there is considerable ambiguity and disagreement among scholars. Psalm 110 is a classic example, with its merger of kingly and priestly authority. Some have resolutely argued that this only makes sense in the context of the Hasmonean rulers, while others have suggested this may be one of the oldest Psalms with close connections to older Canaanite patterns of kingship. To me, the archaic language and difficult state of the text suggest an earlier (rather than later) Psalm. All such attempts to date Psalms based on ideas and cultural contexts are dependent (explicitly or implicitly) on proposed typologies and histories of ideas, and they are generally subject to widespread disagreement among modern scholars. It is an open question just how reliable this type of information is for dating texts, as well as what level of precision can be expected. Can we really date the Torah piety of Psalms 1 and 119 with accuracy and precision? Can we really know when people started repenting of their sins or when people started thinking that the afflicted and poor are the righteous?

            Readers here are likely already familiar with the complexities and debates surrounding attempts to date texts on the grounds of linguistic typology. But among those making such attempts, I think it would be fair to say that there is a strong consensus (for a critical review of the consensus, see Young, Rezetko, and Ehrensvärd 2008, 2:50). Roughly speaking, the first half (or two-thirds) of the Psalter tends to exhibit older language, while a significant number of Psalms in the latter half (or third) of the Psalter yield evidence of later developments in the Hebrew language that suggest dates in the Persian period. Stark linguistic and formal differences with the corpus of Hebrew psalmody from the Hellenistic period make it difficult to imagine that many (if any) of the Psalms from the traditional Hebrew Psalter date from this period, and so it seems likely that most (if not all) of these Psalms were composed no later than the Persian period. I have not seen any evidence—linguistic or otherwise—that compels me to accept a Hellenistic or later date for any of these Psalms.

Date of the Psalter as a Collection

            If the latest individual Psalms that can be relatively securely dated seem to come roughly from the Persian period, that suggests that the collection as a whole is no earlier than this. But can we determine the date of the collection more definitely than that?

Redaction History

            One of the most conspicuous features of the traditional Psalter is that it is composed of multiple layers of sub-collections, each with distinct profiles. The entire Psalter is divided into five books, with markedly similar doxologies placed at the ends of each of the first four books. In Hebrew manuscripts, these books are typically divided with several blank lines, just as in the divisions between books of the Prophets. Within the first three books, Psalms are primarily grouped according to their named attributions, and often they are sub-grouped by Psalm type according to the superscriptions. Books two and three have attribution-based groups arranged in a concentric structure with the Davidic Psalms in the middle and Psalms of Asaph and the sons of Korah on the outside, suggesting that these two books once circulated as an independent collection. It is noteworthy that these two books also underwent an Elohistic revision, where someone fairly systematically replaced the tetragrammaton with the generic name Elohim. This is seen, for instance, in the resulting grammatical anomalies and in cases where Psalms were repeated twice in the Psalter (in one instance using the tetragrammaton and in the other Elohim). Thus, it seems quite clear that the Psalter was not just put together at one time by one person, but rather was formed in multiple stages as small collections were gradually combined to produce the full version we know today. This greatly complicates any attempts to date the completed whole, since an established date for a subcollection would not necessarily be the same as for the entire collection.

            Beyond these fairly obvious blocks, some scholars have proposed more nuanced redaction histories for the formation of the Psalter that complicate the picture even more (most prominently, the prolific Psalms scholars Hossfeld and Zenger). According to these scholars, not only were diverse Psalms collected, but also edited by redactors with ideological tendencies and agendas. So, for instance, existing collections may have been edited to reflect Torah piety or a theology where the righteous present themselves as the poor and afflicted. While the plausibility of such changes cannot be denied, no such proposed complex redactional histories have yet proved compelling enough to establish a consensus in the field, and so I remain hesitant to rely too heavily on such theories.

The Septuagint

            The search for more concrete temporal anchors for dating the Psalter as a collection must begin with its textual history. Here the Old Greek or Septuagint translation of the Psalms looms large, because it is essentially identical in terms of inventory and sequence (if not delimitation) with the traditional Hebrew Psalter. All surviving Greek manuscripts also include Psalm 151, but notable differences in translation technique suggest this was appended after the original translation, and its superscription makes explicit that it remains outside the traditional collection of Psalms.

            Scholars have debated when and where the translation was first made, but a date somewhere in the 2nd century BCE is generally considered most plausible. Most likely, the Psalms were translated earlier in that range than later, perhaps even the late 3rd century, since the translator drew on the existing translations of the Pentateuch but seems in turn to have had considerable influence on later Greek translations typically dated to the 2nd century. The early translation of the Psalms is understandable, given their well-documented prominence in the life, faith, and prayer of the Greek-speaking Jewish community. Since most surviving manuscripts of the Greek Psalms come from a Christian milieu, some have expressed skepticism about the feasibility of reconstructing the contents of this first translation, but specialists working on the Greek Psalter have largely found those sentiments unpersuasive. The Greek Psalms tradition is remarkably stable and coherent and is substantially supported by quotations from a wide range of early Jewish and Christian authors. The Septuagint thus provides rather strong evidence for the existence of the traditional Psalter as a completed collection by the late 3rd or early 2nd century BCE.

The Dead Sea Scrolls

            The Dead Sea Scrolls provide another crucial chronological anchor for establishing the date of the Psalter as a collection, since here we have preserved, datable, material artifacts. The work of scholars like Sanders (1969), Wilson (1985), and Flint (1997) identified and highlighted the diverse contents of some of the Psalms scrolls and problematized their relationships to the traditional Psalter. In more recent years, scholars like Lange (2012), Pajunen (2014), Mroczek (2016), and Willgren/Davage (2016) have stressed this content variation and fluidity and suggested that it fundamentally undermines the search for a definitive, standard collection in the period. On this view, there were many different Psalm collections, of which the traditional Psalter may or may not have been one, but no one collection could claim pride of place. The result has been a growing consensus among specialists that the traditional “proto-MT” Psalter we know from the medieval Hebrew manuscripts was not truly completed and/or established as the definitive traditional version until as late as the 1st century CE.

            Much of my own recent work has addressed precisely this question, as I seek to situate these manuscripts in the context of the material book culture of their time. My approach to the question focuses on two major arguments. First, 11QPsalmsa, the best preserved and most frequently divergent manuscript from the traditional arrangement, is an expanded and rearranged version of the complete Psalter (Longacre 2022). Where its contents can be reconstructed, it retains all of the Psalms known from the traditional proto-MT Psalter (largely in the same subgroups and same general order as the proto-MT, with its base text ending with Psalms 149–150), as well as all its editorial paratexts such as superscriptions and book-dividing doxologies. Where the 11QPsalmsa version innovates, it adds nine new poems and a prose epilogue towards the end of the collection (about 5% text growth) and moves some Psalms to different locations, mostly based on perceived formal or thematic connections. Thus, this alternative version of the Psalter demonstrates the tradition’s openness to new developments, but also indirectly attests to the priority of the traditional, proto-MT Psalter. As it turns out, there are actually multiple copies of this alternative version of the Psalter from Qumran, with overlapping witnesses surviving only for the last half of the Psalter. But several fragmentary manuscripts containing parts of the first half of the Psalter have a similar profile and probably reflect additional copies of the same “proto-Qumran” version of the Psalter. One of these, 4QPsalmsa, also happens to be the earliest of the Psalms manuscripts, probably dating to the late 3rd or early 2nd century BCE on paleographic grounds. Thus, even though the traditional proto-MT version is not directly well-attested at Qumran, there are good grounds to argue that the proto-Qumran version indirectly attests to the full proto-MT no later than the late 3rd or early 2nd century BCE, roughly the same time as the Old Greek translation.

            The second major argument is that the distinctive material features of the remaining manuscripts suggest they are derivative of one of the two known Hebrew versions of the Psalter, rather than remnants of early collections from the formation history of the Psalter or evidence of free arrangement of Psalms without prioritization of the Psalter as an established collection (Longacre 2021b). In addition to the traditional proto-MT and proto-Qumran versions noted above, there are at least three further Psalm collections that differ from both of these known versions of the entire Psalter (especially 4QPsalmsb, 4QPsalmsd, and 4QPsalmsf), as well as many manuscripts that apparently contained only a single Psalm. While recent specialist studies have tended to assert that this formal diversity undermines the existence and or importance of the Psalter as a collection, I argue instead that these manuscripts reflect a vibrant tradition of reusing the Psalter. Particularly noteworthy is that virtually all of these manuscripts yield clear physical indications of informal manuscript production, in sharp contrast with most of the known copies of complete versions of the Psalter, which tend to be very large, professional, and even beautiful productions. These non-standard manuscripts are almost invariably rather small (both in terms of included inventory of Psalms and in physical dimensions and layout), are written in non-calligraphic handwriting styles and/or relatively careless hands, often have layout and textual anomalies that suggest a degree of carelessness, and are never attested in multiple copies. Furthermore, two of the three alternative collections awkwardly switch poetic layout formats in the middle of the scrolls. I have argued that this is due to temporary interference from the scribes’ exemplars, which used the more standard stichographic layout for large copies of the Psalter. All of this suggests that these manuscripts do not in any way undermine the existence and importance of the Psalter, but in fact presuppose it in their rather ad hoc production processes. This textual situation with frequent informal reuse of the Psalter can be easily paralleled in the roughly-contemporary Greek manuscript tradition and the later Hebrew Cairo Genizah documents, which were discovered after having been stashed away for centuries in a Jewish synagogue in Cairo. In both corpora, there are similarly many different forms of—mostly informal—manuscripts containing Psalms and parts of Psalms in assorted collections and arrangements (including with non-Psalms material), though the existence and even canonicity of the Psalter are not in question. This argument from material book culture for the informal reuse of the Psalter undercuts a critical argument for those who would push the date of the Psalter late into the Hellenistic or even early-Roman (63 BCE–135 CE) periods. Instead, it suggests that by the time of the Dead Sea Scrolls the Psalter was already well-established, even if circulating in more than one version.

Chronicles

            In my opinion, the single most important source of evidence for the completion of the traditional version of the Psalter is the book of Chronicles. Though a few scholars have tried to argue that the Psalter depends on Chronicles, most scholars assert the reverse. What is especially noteworthy about Chronicles is that we have specific evidence not just for the use of individual Psalms from the Psalter but also for the use of a particular form of the Psalter as a five-book collection.

            Most significantly, the composite hymn of 1 Chronicles 16:8–36 is clearly built from a catena of verses taken from Psalms 105:1–15; 96:1–13 (selectively); 106:1, 47–48. All of these are from book four of the Psalter, and it is probably not coincidence that Psalms 105 and 106 are right next to each other in the Psalter and provide the dominant and framing content for the catena. Not only does this suggest that the Chronicler knew a collection that had book four, but here we actually have strong evidence for the full five-book collection. 1 Chronicles 16:36 preserves Psalm 106:48, which is the book-dividing doxology at the end of book four. Though some have suggested this doxology might originally have been part of Psalm 106 itself, it is evidently outside of the main structure of the Psalm with its enveloping inclusio of praise/thanksgiving (ydh) in vv. 1 and 47. This, combined with its distinctive similarities with the other doxologies, suggests that the doxology was added to the end of this part of the collection as part of its compilation history where book four was to be joined to the other books.

            We can take this argument even further. The doxology in Psalm 106:48 ends with a liturgical call for all the people to say Amen and Hallelujah, which Chronicles duly narrates as a performance. The exclamation Hallelujah never occurs in books one to three of the Psalter, but it is ubiquitous in book five, and some have argued it is a key framework for the entire fifth book. In book four, it only occurs between Psalms 104–106, the last three Psalms in the book. Thus, it seems reasonably clear that the addition of Hallelujah to the doxology was added to smooth the transition to book five, and Chronicles must have known a five-book version of the Psalter. This is further supported by the fact that 2 Chronicles 6:41–42 extensively reuses Psalm 132:8–10 from book five of the Psalter.

            When we look at the rest of Chronicles (especially 1 Chronicles 9; 15; 16:1–7, 37–42), we find the hypothesis of a five-book Psalter readily confirmed by evidence that the Chronicler drew heavily on the Psalter. The Chronicler’s discussion of David’s establishment of temple psalmody features all of the familiar Levitical singers from the first three books of the Psalter, including Asaph, the sons of Korah, Heman, Ethan the Ezrahite, and Jeduthun (interpreted as a personal name).

            Thus, in Chronicles we have strong evidence for a five-book Psalter that overlaps substantially (if not entirely) with the traditional Psalter. Without a detailed Psalm-by-Psalm inventory, we must of course leave open the possibility that a few individual Psalms could have been added in after the time of the Chronicler, but the conclusion seems unavoidable that the basic macrostructure and major sub-groupings of the Psalter were already present in the version used by the Chronicler. Moreover, the influence of the Psalter on Chronicles seems so pervasive that it cannot easily be dismissed as late and secondary within the Chronicles tradition, as some have attempted to do. While the precise date of Chronicles is itself not entirely certain, Chronicles provides rather strong evidence that the compilation of the Psalter was essentially complete by the end of the Persian period (4th century BCE).

Circumstantial Evidence from Book History

            The argument for the Persian-period formation of the Psalter can be bolstered by one final piece of circumstantial evidence from the scribal practices of the period. The five-book division of the Psalter has always been a bit of a problem. Since the Psalter was small enough to fit on one scroll, there did not seem to be any need to have divided it into five books, so scholars generally have explained the division as late, secondary, and imitative of the Pentateuch (Haran 1989). Yet books of the Psalter are not of equal length; the divisions rather align with evident editorial seams from the piecemeal collection of the whole, which seems unlikely to be coincidence. Recent advances in understanding material book culture have suggested a rather more compelling answer, namely that the book divisions are actually early remnants of the collection process from multiple subcollections. The arrival of the Greeks from the late 4th century BCE instigated major changes in Jewish Hebrew/Aramaic book formats (Longacre 2021a). Writers started using a different, sharper reed pen (instead of the earlier rush brush) and preferred to write much smaller, denser scripts. As a result, scrolls from the Hellenistic period and later (e.g., the Dead Sea Scrolls) could hold much more text on them than earlier scrolls, and so the Psalter could fit on one scroll. But when we look at Persian-period literary formats, we find that the larger scripts meant scribes could fit far less text than later scribes. In short, in the Persian period, each book of the Psalter would have measured about 3–7 meters (which fits well with expectations), but a full Psalter would be considerably longer than expected. In other words, a five-book series of Psalms fits perfectly within the scribal practices of the Persian period, but does not make much sense at all in the Hellenistic period. Thus, the clues from material book culture further support the idea of a Persian-period, five-book Psalter.

Conclusion

            In my research, I have found the extremely late dates for the final formation of the Psalter proposed in some recent studies to be unrealistic and unable to account for the full evidence from the tradition. The evidence amassed here (and in greater detail in my forthcoming monograph) establishes—to my mind conclusively—that the basic framework of the traditional, five-book Psalter is the product of the late Persian period (5th–4th centuries BCE). Lacking a detailed inventory from this period, it is always possible that some individual Psalms were added later, but we can say confidently that all of the traditional Psalms were included by the late 3rd or early 2nd century BCE at the latest.

References

Flint, Peter W. 1997. The Dead Sea Psalms Scrolls and the Book of Psalms. STDJ 17. Leiden: Brill.

Haran, Menahem. 1989. “The Four Doxologies and the Five ‘Books’ in the Book of Psalms.” Proceedings of the National Academy of Sciences 8: 1–32 (Hebrew).

Lange, Armin. 2012. “Collecting Psalms in Light of the Dead Sea Scrolls.” Pages 297–308 in A Teacher for All Generations: Essays in Honor of James C. VanderKam, edited by E. F. Mason, Samuel I. Thomas, Alison Schofield, and Eugene Ulrich. JSJSup 153/1. Leiden: Brill.

Longacre, Drew. 2021a. “Comparative Hellenistic and Roman Manuscript Studies (CHRoMS): Script Interactions and Hebrew/Aramaic Writing Culture.” Comparative Oriental Manuscript Studies Bulletin 7 (1): 7–50. http://doi.org/10.25592/uhhfdm.8898.

———. 2021b. “Paleographic Style and the Forms and Functions of the Dead Sea Psalm Scrolls: A Hand Fitting for the Occasion?” VT 72 (1): 1–26. https://doi.org/10.1163/15685330-bja10019.

———. 2022. “The 11Q5 Psalter as a Scribal Product: Standing at the Nexus of Textual Development, Editorial Processes, and Manuscript Production.” ZAW 134 (1): 85–111. https://doi.org/10.1515/zaw-2022-0004.

———. 2026. Scribes and the Psalter: Situating Ancient Psalm Collections in Their Scribal Contexts. FAT I. Tübingen: Mohr Siebeck, forthcoming.

Mroczek, Eva. 2016. The Literary Imagination in Jewish Antiquity. Oxford: Oxford University Press.

Pajunen, Mika S. 2014. “Perspectives on the Existence of a Particular Authoritative Book of Psalms in the Late Second Temple Period.” JSOT 39 (2): 139–63.

Sanders, James A. 1969. “Cave 11 Surprises and the Question of Canon.” Pages 101–116 in New Directions in Biblical Archaeology, edited by David N. Freedman and Jonas C. Greenfield. Garden City, NY: Doubleday.

Willgren, David. 2016. The Formation of the “Book” of Psalms: Reconsidering the Transmission and Canonization of Psalmody in Light of Material Culture and the Poetics of Anthologies. FAT.2 88. Tübingen: Mohr Siebeck.

Wilson, Gerald H. 1985. The Editing of the Hebrew Psalter. SBLDS 76. Chico, CA: Scholars Press.

Young, Ian, Robert Rezetko, and Martin Ehrensvärd. 2008. Linguistic Dating of Biblical Texts, 2 vols. BibleWorld. London: Equinox.