Paul and Scripture Seminar
SBL 2006 Annual Meeting Seminar Papers
Stanley E. Porter, McMaster Divinity College, CA
Paul and His Bible: His Education and Access to the Scriptures of Israel
There have been many introductions to Paul’s use of Scripture over the last twenty years, or so. Most would agree that we have increased in our knowledge of how Paul uses scriptural texts in a variety of ways. Some of these studies have concentrated upon the various forms of the text that Paul used, such as Christopher Stanley’s detailed study. As a result, we have a clearer idea of the variety of textual traditions that may have been available to Paul, and the ones that he may have drawn upon in his writings. Others have focused upon the various ways in which these texts have been incorporated into Paul’s writings. Some of these treatments have been broad in scope, such as Richard Hays’s attempt to distinguish various types of intertextuality. Others have been more specific, such as Ross Wagner’s treatment of the use of Isaiah in Romans. Still others have been concerned, not with the way that the texts have been incorporated, so much as with what function the texts have played in Paul’s writing. Again, Christopher Stanley has aided us in deciphering this, as he has analyzed the various rhetorical purposes and effects that Paul’ use of Scripture has had. Most of these treatments—with the noteworthy exception of Hays and his loyal followers—have concentrated upon explicit quotations, especially those that have been marked or indicated with a quotation formula. This has been the case to the point that many studies have been limited to treatment of those passages that are introduced by quotation formulas.
There have been a number of other assumptions that seem to have driven and regulated the study of Paul’s use of Scripture. One is that Paul had a relatively low level of education, especially in Greco-Roman circles, even if he was better educated in the rabbinic (or equivalent) traditions of Jerusalem. Another is that there was widespread illiteracy in the ancient world, including the world of Judaism, perpetuated by the fact that writing was difficult, that those competent in writing were difficult to find, that writing materials and people were costly, and that, as a result, books were in limited supply and hard to find. Following on from that is the assumption that, when texts were read in the ancient world, they were read aloud. These are jut some of the fairly common assumptions.
In this paper, I have been asked to address some of these technical issues that accompany the study of Paul’s use of Scripture. If we acknowledge that Paul used the texts of the Old Testament (not necessarily the Hebrew Bible), then he must have had access to them in some form. The questions that I wish to raise here are related to how he had access to the texts that he drew upon. Therefore, I wish to address several issues. These issues include: (1) the educational system of the ancient world, and along with it the educational level that Paul could reasonably have achieved; (2) the nature of the book culture in the ancient world, including the availability of books and what books were available; (3) the question of reading, as a phenomenon and in particular with reference to Paul; and (4) the process by which Old Testament texts became Pauline citations, and what we can determine from that process. I will only be able to make suggestive remarks under each category, but I believe that they will open up potentially new insights into if not how we interpret Paul, at least how we view Paul as an interpreter of Scripture.
2. Paul and the Greco-Roman Educational System
Paul was born and reared, at least up to a certain age, in Tarsus, in Cilicia. One of the most extensive treatments of Tarsus in ancient literary sources is from Strabo (64/63 B.C.–A.D. 21 at least, according to the OCD), the first-century author who would have been writing about Tarsus at the time Paul was born and lived there. Strabo makes a number of important comments about the city. Regarding their philosophical and educational aspirations, he states, “the people at Tarsus have devoted themselves so eagerly, not only to philosophy, but also to the whole round of education in general, that they have surpassed Athens, Alexandria [the two leading centers other than Tarsus], or any other place that can be named where there have been schools and lectures of philosophers” (Strabo, Geography 14.5.13; LCL). Some of these distinguished people included Stoic and other philosophers and rhetoricians, among them one rhetorician who had been the teacher of Julius Caesar (14.5.14). However, according to Strabo, there are some things that distinguished the educational system in Tarsus, and that are of special interest in our consideration of Paul and his education. Strabo says, “But [Tarsus] is so different from other cities that there the men who are fond of learning are all natives, and foreigners are not inclined to sojourn there” (14.5.13; LCL). Whereas it appears that the custom was for young men of other cities to go to another city to study and become educated, it was the custom for Tarsians to stay at home, at least for the initial stages of their education. Strabo states further, “neither do these natives [of Tarsus] stay there, but they complete their education abroad; and when they have completed it they are pleased to live abroad, and but few go back home” (14.5.13; LCL). By contrast, other cities, such as Alexandria, had a mix of native and foreign students. This description by Strabo of Tarsus is uncannily close to the description of Paul’s development that we find in the New Testament. Paul was born and reared in Tarsus (Acts 9:11, 30; 11:25; 21:39; 22:3; cf. Gal 1:21), and so had the possibility of gaining some education there as Tarsians did, before leaving to complete his education abroad in Jerusalem, under Gamaliel (Acts 22:3), and never returning there to resettle.
In the light of these considerations, there are two pertinent questions to answer. The first revolves around what the educational system would have been like that Paul may have participated in, and what evidence do we have that he had been educated within such a system. The nature of his education has direct bearing on the access he would have had to the Bible. The number of educational centers in the ancient world was relatively small. For a Roman intending to “study abroad,” the two primary choices were Athens and Rhodes, some of the universities in now France such as Marseilles or Autun, or some of the cities in the east, such as in Asia Minor including Mytilene, Ephesus, and Berytus (and later Constantinople). Few went to Alexandria (neither did Paul, it appears!), perhaps, Daly suggests, because of the political tensions. Paul, however, being a Tarsian, did not have to worry about going abroad, as the custom was to stay at home, at least to the point where one left for the final educational stages. We do not know at what age Paul left home. There is much dispute over interpretation of Acts 22:3, and whether, when Paul refers to being raised in “this city,” he is indicating Tarsus, which he has just referred to, or Jerusalem, where he studied with Gamaliel, whom he mentions next. Most scholars, on the basis of three factors—their understanding of the educational system (but see below), their perception of Paul’s Greco-Roman knowledge and education, and his being Jewish—conclude that Paul was not educated to a very high level in the Tarsian Greco-Roman educational system. There are a number of considerations, however, that make resolution of the specific meaning of Acts 22:3 less important than getting a grasp of the organization of the educational system. The nature of the educational system points to the fact that Paul could have been educated in both Tarsus and Jerusalem.
The traditional and still standard viewpoint is that the Greco-Roman educational system had three successive tiers. This position has been argued by a number of scholars in the past, such as Marrou, Bonner and Clark, among others, and repeated more recently by such scholars as Morgan and Hock. The three tiers were: the primary school or ludus litterarius, the grammar school or schola grammatici, and the rhetorical school or schola rhetoris. These three were sequentially arranged, so that one attended the primary school first, where one learned the basics of reading and writing, and perhaps some mathematics, before proceeding to the grammar school. At the grammar school, one learned grammar and some composition, including letters, and was introduced to literature, especially the writings of Homer. At the third level, the rhetorical school, one learned rhetoric or oratory, and concentrated on various prose authors. This tripartite organization, however, has been revised in the light of recent research. A.D. Booth noticed that, in the ancient evidence, there were many instances when the first teacher of a student was equated with the functions of both the primary teacher and the grammatical teacher. As a result, he suggested that the Greco-Roman educational system, rather than being sequential, was concurrent and organized socially. There were essentially two tracks. The first track, or elementary school, was for those of the lower social orders, including slaves. The second track was for those of status, or the upper class. The latter received elementary education either at home or in the initial stages of their grammatical education. The system then consisted for them of essentially two levels, grammatical and rhetorical education. Even this scheme, however, as Robert Kaster has shown, is overly rigid, with there being abundant evidence that the educational system of the time was much more flexible and less rigid than previous scholarship has realized, and more subject to regional variation. Nevertheless, the evidence Kaster marshals does seem to support the notion that the distinction between the primary and grammatical teachers was blurred and their function greatly overlapped (there is little evidence that a student going through the grammatical system normally attended primary school), both in Rome and in other places throughout the Greco-Roman world.
If this is the case, and it seems like the evidence points in this direction, it has implications for Paul’s education. In this revised organization, students began their schooling somewhere around six to eight years old, and would have finished the grammar school at around twelve or thirteen years. As a Roman citizen, and the son of a person with a productive trade (which probably resulted in the citizenship of his father, or an earlier relative), Paul would apparently have had the status and economic support that would have financed and facilitated his attending the grammar school. This chronology is consistent with the ages of man as discussed by the rabbis, according to W.D. Davies. As Pirke Aboth 5.24 (c. A.D. 150) says, “At five years the Scriptures; at ten years the Mishnah; at thirteen the commandments; at fifteen the Talmud etc.…” It was thought that generally at the age of thirteen a boy was made a “son of commandment” and welcomed into the Jewish community. Even though the tradition may be later, this pattern of development is consistent with Paul being raised in a Jewish home in Tarsus and learning the Scriptures while being educated in the grammar school, and then, after his bar-mitzwah (or equivalent at the time), going to Jerusalem to complete his education in Jewish law and related matters. The word that is translated “reared” in Acts 22:3 does not necessarily mean being reared from birth, but may indicate being educated, without reference to when the process began, and is compatible with either Tarsus or Jerusalem, without necessarily restricting his education in the other. This educational structure is also consistent with the educational framework that Birger Gerhardsson outlines. He notes a number of similarities between the Torah schools and the Greco-Roman schools, parallels that are complementary to the portrait of Paul that we are creating here. Rather than the poets, a student of the Torah school would study the Old Testament, and rather than a student advancing from poets to rhetoric, in the Torah school they proceeded from the written Torah to the oral Torah, in other words, from Mishnah to Talmud, as it was later codified. Although the evidence for advanced Jewish education is later, from the tannaitic but especially the amoraic periods, the general pattern was that the few fortunate students who could would study with a rabbi and attend a “study house,” where they would engage not so much in reading the Torah (basic Torah knowledge would have been assumed), but in disputing over interpretation of the Torah, that is, Talmudic discussion. According to this model, then, Paul would have had plenty of time to have been instructed in the written Torah in Tarsus, even as a student of the grammar school and possibly beyond, before leaving for Jerusalem to continue his education with Gamaliel the rabbi on the oral Torah.
As important as the organization of the educational system—and perhaps the most important element for this paper—is the content of the educational system. What exactly would Paul have learned as a student in a grammar school? As noted above, if he did not learn the basics of writing, reading and arithmetic from a private tutor or at home before entering the school, he then acquired these skills right at the outset. However, there was much more to the grammatical education. Students started with simple exercises, in which there was syllable or word separation, and gained increased facility in reading. Students increasingly learned about words and then sentences, and then about basic literary features of a text, including punctuation and marking of change of speakers. At this stage the major authors that were read were poets, with an eye toward identifying the features of poetic texts and how to evaluate them. The major poet that students read was ‘The Poet’ Homer (especially his Iliad), along with Hesiod, Pindar and other lyric poets including contemporary ones like Callimachus, Euripides mostly among the tragedians, Menander, and the like. It appears that there were certain authors who were at the core of the curriculum, while others were more at the periphery. The students learned to read, recite and explain these authors, in other words, there was a huge emphasis upon the oral and mnemonic character of education. However, there was also a written component. It was during the first century that grammatical study became a more or less fixed part of the grammar school curriculum. There was also practice in composition. These included the so-called preparatory exercises, or progymnasmata. These included such forms as fable, narrative, chreia, aphorism, confirmation or refutation, commonplaces, eulogy or censure, comparison, ethopeia or prosopopeia, the thesis and discussions of law. These were all considered necessary preparation for moving to the next stage in education, rhetoric proper. Though not central to the curriculum like the progymnasmata, at the grammar school, the student learned how to write letters. Learning to write letters has long been recognized as one of the skills learned by those who were becoming professional scribes, but it is increasingly recognized that learning to write letters was a part of grammar school as well. Students probably did not learn by copying letters out of books, but by constructing them themselves, perhaps under the direct guidance of their teacher.
In regard to the curriculum, there are at least three observations to be made in the light of recent Pauline scholarship. The first concerns the progymnasmata. Recent rhetorical studies of Paul attribute to him the ability to write a number of the different stylistic characteristics found in the kinds of exercises he would have learned in the grammar school curriculum. The natural place to learn these would have been in the grammar school. There is plenty of evidence that he learned how to use these exercises, and used them in his letters. The second concerns the literature read at this stage. It is worth noting that, of the secular authors that Paul cites or alludes to, or is depicted as citing or alluding to in Acts, most of them come from the canonical authors mentioned above, including especially a variety of poets (Menander, Thais frg. 218 in 1 Cor 15:33; Aristotle, Politics 3.8.2 1284a14-15 in Gal 5:22; Aeschylus, Eumenides 1014-15 in Phil 4:4; Pindar frg. from Strabo 6.2.8 in 2 Tim 2:7; Epimenides in Titus 1:12; Aratus, Phaenomena 5 or Epimenides in Acts 17:28; Euripides, Ion 8 in Acts 21:39; Euripides, Bacchae 794-95 in Acts 26:14). Some speculate that Paul did not know these authors’ works but that he simply cited quotations from major writers found in testimonia. If this is so, he probably gained such exposure during his grammar school training (see discussion below). The third area is in terms of Paul as a letter writer. Despite the debate over whether they are letters or epistles, scholarship has continued to recognize that Paul was one of the major letter writers of the ancient world, as the size and character of his letter collection attests. Paul’s use of the letter form seems to indicate an author who knew the conventions of letter writing but who also innovated in his use of this form, by development of various sub-sections of the letter (such as expansion of the sender or receiver in the salutation), or development of various sections of the letter, such as the thanksgiving or paraenesis.
The chronology of Paul’s life and the evidence from his letters and Acts make it plausible that Paul was a product of the Greco-Roman educational system in Tarsus. The structure of the system, including the ages of study, and the curriculum point to Paul having completed the grammar school, before leaving Tarsus to complete his education in Jerusalem in the rabbinic school. During his grammar school studies, besides basic literacy and numeracy, Paul would have been exposed to the major poets and authors and learned composition through the progymnasmata and letter writing. Whereas he would have already have studied the written Torah during his studies in Tarsus, he would have learned the oral Torah in Jerusalem. The evidence indicates that there is nothing in the educational system in Greco-Roman times that prevents Paul from having finished the grammatical school level, and with it gaining the various skills that that required.
3. Oral and Book Culture
Paul’s being able to read and write is also important for understanding how he utilized Scripture. However, it is not enough to explain how it was that he gained access to and then utilized the texts that he did. In order to understand that, we must understand the interplay of oral and book cultures. There is a tendency in the current climate of biblical interpretation to emphasize the elements of oral culture, at the expense of the book culture of the time. I wish to say something about each of them here. The first statement that I will make, however, is that the situation is much more complex than simply that between oral and written. As Robbins has stated, the situation of terminology needs to be clarified. He differentiates between the following when talking about culture: oral culture, scribal culture, rhetorical culture, reading culture, literary culture, print culture and hypertext culture. I am not sure that we need all of these categories, but the situation is admittedly more complex than simply oral vs. written, as we shall see.
I probably do not need to say as much about orality, as this is a subject that has been widely studied lately. The emphasis in recent studies has been upon the oral nature of Judaism and early Christianity, as they were positioned within the predominantly oral cultures of the ancient Mediterranean Greco-Roman world. Following on from work by those such as Walter Ong, a scholar like Werner Kelber emphasizes the oral dimension of the ancient world and especially of Paul’s work. Kelber draws a contrast in Paul’s thought between the stultifying effect of the written word, which is equated with the law, and the liberating and spirit-filled oral word. According to Kelber, Paul emphasizes the oral word, which is related to the gospel and it strength and power. Kelber’s ideas have been picked up and developed further by a number of scholars, to the point that the oral is clearly emphasized to the point of dismissing the written.
There is no doubt that oral culture played a significant role in the ancient world. However, there has been evidence gathered on both the Greco-Roman and Jewish sides that attempts to put this in perspective. To adapt the terminology of Robbins, there is not a need to create a false disjunction between oral and written culture, but there is a need to recognize that there are levels of inscribed culture, such that there may be a place for a scribal culture, even if there is not as large a literary culture. In fact, this is what we find. In the ancient Greek and then Roman world, even though there was an emphasis upon orality, such that the major poets wrote for oral performance, in such media as epic and tragedy, with the growth of empire there came a need for record-keeping that led to a scribal culture that developed more fully into a literary culture. Thus Rosalind Thomas distinguishes between documents and records. There are potentially all sorts of documents, written for different purposes. However, there is a transition that takes place when one recognizes that written documents may become records of persons, possessions or other things. With this recognition there comes a certain power in literacy, to the point of developing systematized record-keeping and the like. This is one of the hallmarks of the Roman empire, especially as evidenced in the papyri remains from Egypt. The confluence of record-keeping and power can be clearly seen in the tax records that were kept every seven and then every fourteen years. The power to record became the power to tax and the power to continue to control the people. In the Jewish world of the time, there was a somewhat similar phenomenon, as Gerhardsson has made us aware. Gerhardsson offers a more sophisticated (and the history of research tells us, more prescient) view of the interplay of orality and literacy than many others. He notes the creative and dynamic role that oral tradition played in the rise of Judaism. He also notes that early on, and a number of centuries before the Christian era, written tradition also began to play an important part in Judaism. Gerhardsson’s analysis ends up claiming that the written Torah, what we would equate with our Old Testament, was complemented by the oral Torah, which constitutes the oral tradition that goes hand in hand with it.
A factor that is not as fully realized as it might be, however, is the type of book culture that was already present in the first century A.D. The standard position is often to dismiss such a notion on the basis of purported widespread illiteracy, the high cost of materials, and the lack of printing. Nevertheless, the evidence is that there was a widespread book culture at that time. In some ways, the disparagement of writing and the book culture is an unfounded consequence of a paradigm that draws a disjunction between orality and literacy, and, because the culture is posited as oral, then claims that there must not be significant written sources. How then do we account for such factor as Galen wandering through a market and seeing books for sale, and his checking and seeing that some of those for sale were attributed to him and he knew that he had not written them? In other words, there was enough of a market for books that it made forgery a desirable option for some (clearly not for Galen, who was incensed). Similarly, the literary tradition associated with Lysias is that a significant number of the works attributed to him even in ancient times were known to be forgeries, a problem that has kept scholars busy for years. Likewise, the Qumran community, though always a relatively small community living in isolation, was responsible for creating a significantly large number of books. There were also a number of well-known large libraries, such as the ones at Alexandria and Ephesus, among others. There were also a number of small private libraries, which may have been more important to the book culture than the larger libraries. In other words, books appear to have been so plentiful by the first century A.D. that Seneca goes so far as to denounce what he sees as the ostentatious accumulation of books.
There are two major features of the book culture to note. Without minimizing the importance of orality—and that the major means of ‘publication’ was in terms of oral performance, such as orations, the reading of poetry, lectures delivered in public, and theatre—we must also recognize that there was a parallel book culture that was large and significant. There was no publishing industry as we would know it today, but there were means of getting books produced, nevertheless. Papyrus, the paper of the ancient world, was widely available and was not expensive. As Thomas Skeat has argued, the cost of papyrus was reasonable in the ancient world, both on the basis of statements by the ancients and the bountiful evidence of discarded papyrus discovered throughout Egypt—very few of which were ever re-used, even if they were blank on one side. The cost of getting a book copied was not exorbitant, and ranged from two to four drachmas, which is the equivalent of from one to six days pay.
The second feature of the book culture is the nature of that culture itself. Kenyon notes that as early as the fourth century B.C. there was a “considerable quantity” of “cheap and easily accessible” books to be found in Athens, which shows that a reading culture was growing even during that time. In Hellenistic Egypt, as is indicated by the documentary and especially the literary finds, there were numerous books available, to the point that “Greek literature was widely current among the ordinary Graeco-Roman population,” and this was a likely pattern throughout the Hellenistic world. That these books were accessible to a wide range of people is indicated by the fact that the most common literary author found in the papyri is Homer, the major author read in the grammar school. Access to these books was gained through a variety of means. There was what amounts to what Loveday Alexander has called a commercial book trade. Alexander notes that the main motivation of authors to publish in ancient times—perhaps not too unlike modern times—was fame, not necessarily money. Hence there were not the same kinds of restrictions on access as we have created today. Authors themselves would probably have been involved in “publication” of their books. Raymond Starr notes that the author would have written a rough draft, and then had it reviewed by others, such as slaves or friends. Once a final form of the text was formulated, it was circulated to a wider group of friends, before being more widely disseminated. As a result, perhaps the easiest way to secure a book was to borrow a copy from a friend and either have a slave or scribe copy it, or make one’s own copy. As Alexander notes, Aristotle, Cicero, Galen and Marcus Cicero the younger all seem to have been a part of this process. There were other means by which books were circulated as well. Alexander notes four. The first is the copying and exchanging of books by professional scribes. Another is the transcribing of orations into written speeches. A third was to copy down lectures given by teachers and circulate them among students. A fourth and final was to dedicate a book to a patron who may have been involved in the publishing process. Harry Gamble has noted that there were associations of people connected with book production, so that those who were interested in a particular type of literature would produce and share these books. This could include those who were interested in the Greek text of the Bible (i.e., the Old Greek or Septuagint). Josephus records that a rebel in Galilee confronted him with “a copy of the law of Moses in his hands” as he tried to work the crowd (Josephus, Life 134).
As a result of the book culture, there were a number of types of collections of books that we have from the ancient world of the time. When, at the end of the fourth century B.C., Ptolemy I Soter founded the Museum in Alexandria, an effort toward standardizing the most substantial Greek texts of antiquity was initiated. The mass scale of book production at the Alexandrian library played a crucial role in standardizing both the selection of literary works and the form they took (book size, material, etc.). As Easterling and Knox note, “the corpus of the ‘best authors’ was given official recognition in the classifications made by the scholars of Alexandria and perhaps of Pergamum, and came to exercise a very powerful effect on Greek culture.” Specifying the “canons” of Greek literature (the ten orators, the nine lyric poets, the three tragedians, etc.) is typically accredited to Aristophanes of Byzantium. The first formal standardization of a collection of authors seems to have been the formation of the canon of the attic orators around the time of the Alexandrians and the Roman Empire. Several writers of antiquity mention the canon, including Caecilius, who wrote the now lost On the Character of the Ten Orators, Quintilian (Inst. or. 10.1.76), Lucian (Scythian Or. 10) and Pseudo-Plutarch’s Lives of the Ten Orators. Canonization was a common Greek educational practice that allowed the selection of a list of authors from a particular genre to be standardized as the objects of instruction in the schools. Literary collections of poets and canons of various philosophers followed soon after. These collections of authors provided the basis for curriculum within the ancient schools, although which authors were emphasized or selected remained much more fluid. The writings of various canonized authors thus became the basis for instruction, imitation and criticism in the Hellenistic schools. The works of Homer were perhaps most affected by this process under the influence of Aristarchus, who successfully standardized Homeric compositions to conform to a common “vulgate,” reflected in nearly all Homeric papyri found after the middle of the second century B.C. The works of Homer, in particular, were imputed with a normative status in Greco-Roman culture comparable—if not on par with—that of the Jewish Scriptures in early Christianity.
Hellenistic authors or speakers who desired to quote from any or all of these canons of literature would have faced some of the same material difficulties that early Christians and Jews would have dealt with in their attempts to cite authoritative texts. The cumbersome nature of scrolls, the lack of ease in navigation, cost of ownership and the sheer size of many ancient books all made employing citations from written material extremely unpractical. This certainly explains the emphasis upon imitation and memorization in Hellenistic education. As Stanley notes with regard to the citation of Homeric literature, “When one takes into account the difficulties associated with unrolling a large scroll to check and find references (hence the frequent reliance on memory in the ancient world), it is…the faithfulness of the authors of this period to their sources that appears remarkable.” However, Stanley may give the Greek memory more credit than it is due. While the majority of Homeric papyri consist of the two major epic poems—the Odyssey and the Iliad—there is nevertheless an ample papyrological supply of Homerica, various lexica, scholia, anthologies, glossaries, summaries and paraphrases of Homer’s work. P.Mich.inv. 4832 (late second c. to first c. B.C.) provides an example of a Ptolemaic papyrus of an anthology from Iliad S-T. Other significant Homeric anthologies on papyri include P.Strasb. inv. 2374 (third c. B.C.), P.Vindob. gr. inv. 26740 (second c. A.D.), P.Hamb. II.136 (early third c. B.C.). As Renner observes, together “these fragmentary anthologies bear abundant witness to continued popularity through five centuries in Greco-Roman Egypt to a type of study which had probably been in use in the schoolroom since at least Plato.” They would have also provided authors with a more manageable form of material than a book length scroll for citing Homer and other classical authors without having to rely solely on the facility of memory.
Perhaps the most common collection of quoted material that circulated in the Hellenistic world was the Greco-Roman anthology. Greco-Roman anthologies were diverse and abundant, taking a variety of forms (e.g. gnomologium, chreia, poetry, biography, history, rhetoric, philosophy) and being employed substantially as schooltexts within the system of Hellenistic education. Gnomologies were especially significant in Greco-Roman education. As Morgan insists, “Gnomic schooltexts form part of a long and complex tradition in Greek literature, and one in which Greek literature was part of an even older and more extensive tradition in the literatures of the Near East.” These anthologies emerge in the Hellenistic period and typically consisted of a collection of sayings on divergent subjects by the same or various authors relevant to individuals at all layers of social strata. Manuscripts of collections of gnomic sayings and various anthologies are in greater number than any other form of schooltexts, and were apparently used at every level of education. Quotations from Menander or in Menandrean style were most popular, but sayings from other prose poets are represented as well. While not all of the topical categories for gnomic sayings mentioned in literary sources are found among the schooltext papyri, several emerge with a great deal of consistency, evidencing their important role in early education. Some of these reflect themes that Paul takes up in his letters, such as wealth, ‘word’ and speech, family, women, friends and associates, old age. The authority of these texts undoubtedly derives “from their being part of classical Greek culture, as well as from the authority which is awarded conventionally to poets of all sorts.” From the above analysis of Greco-Roman anthologies it becomes clear that literary and papyrological sources confirm the use of literary compilations as a significant aid in citing ancient authors in the production of original compositions. This makes sense of the accuracy and consistency with which Greek authors quoted Homer and other ancient literature.
Philosophers, rhetoricians and historians were anthologized as well. These collections served as essential compositional tools used by ancient authors for accessing classical literature. As Barns has convincingly shown, the famous bee simile was likely employed as a suggestion for students to create compilations of significant poets as a tool for later composition. The simile seems to have been used first by Isocrates (Ad Dem. 51) where students are encouraged to compile collections of poetic writings for study. It takes on a similar connotation in Plutarch (Moralia; De recta ratione audiendi 8.41E), but refers to philosophers instead of poets. In Lucian (Pis. 6), we find evidence that rhetoricians used philosophical anthologies as a tool to aid in their compositions. Similarly, Seneca (Ep. 83) suggests employing a written compilation when composing a literary work. The simile is even employed in early Christianity to refer to the selection of the best pagan authors (Clement, Strom., 1.11; 33.6; 6.89.2; PG 31.564). Barns concludes that the form of literary compilation in these examples has relevance to four areas: the selection of literary excerpts, their compilation in writing, their application to education, and their use as an aid to original composition. These selections were made either for the purposes of pleasure or for intellectual and educational purposes.
Collections of authoritative material took a variety of forms in Hellenistic Jewish sources as well. As far as collections of literature go, there was in New Testament times the notion of the Old Testament as a body of Jewish literature, but, as the discoveries at Qumran indicate, the understanding of the canonical status of many books was still rather fluid. Canons of rabbinical material, whose boundaries became fixed much more quickly, also emerged, including the Mishnah and various compilations of tannaitic tradition including the Tosepta, the Amoraim and the halakic midrashim. Like other writings of the first century, we have testimony to compilations in the Mishnah such as Perque ’Abot and ‘Eduyyot. Similar anthologies are represented in Hellenistic Jewish literature like Pseudo-Phocylides as well. Perhaps the most significant Jewish parallel to the Greco-Roman anthologies of classical authors are the collections of excerpts from the Hebrew Bible found in Qumran Cave 4: 4QPatriarchal Blessings, 4QTan (4Q176), 4QOrd (4Q159), and especially 4QTest and 4QFil. The two most significant compilations, 4QTest and 4QFil, are collections of excerpts revolving around the eschatological hopes of the community and seem (at least in the case of 4QFil) to have served liturgical purposes as well. 4QTest contains five passages: Deut 5:28-29; 18:18-19; Num 24:15-17; Deut 33:8-11; and a selection from a book accredited to Joshua. Unlike 4QTest, 4QFil includes midrash upon its three central selections: 2 Sam 7:10-12; Ps 1:1; Ps 2:1-2 (cf. also 4QTan [4Q176]).
The evidence suggests that there was a book culture that cut across all of the various cultural and ethnic groups of the first century. The result was that there was an abundance of written material available. Some of this was in the form of whole works, while other documents attest to a process of selection and anthologizing. The anthologized texts made access to extended works more manageable as well as providing ease of transportation.
4. Literacy and Reading
The last topic that I wish to address—literacy and reading—does not need to be treated as extensively—literacy and reading. Though important, they are not as germane to the topic of how Paul handled Scripture, as it is clear from what has been said that Paul was literate. Nevertheless, I wish to say something about literacy in general and then about reading to round out the picture I am creating.
The issue of literacy in the ancient world, and especially among Jews, has become a subject of renewed interest and debate. William Harris’s well-known monograph was the first to undertake a historical rather than a social-anthropological study of literacy in the Greco-Roman world. He was not the first to address the issue, but was the first to study systematically the historical, documentary and literary evidence, and then attempt to quantify the literacy of the ancient world. His work distanced him from those who had emphasized the oral culture of the ancient Greeks, and established a target to shoot at. Some rejected his findings by claiming that he had overestimated the influence and capability of writing, but the majority of the response was to find that he had underestimated the impact of literary culture at least in terms of the Greco-Roman context. Even if his statistics are accurate and not underestimated—that overall 20-30% of men in the Roman empire were literate, and that 10-15% of women were, for an overall ratio of no more than 15%—he seems to have neglected the fact that even those who were illiterate came into contact with literate culture. For example, a person illiterate in Greek might need to have a contract written out, and so would be a part of literate culture by virtue of needing to have this document prepared, and deal with the consequences of it, such as a document sent in return. Bowman goes so far as to state that a “large proportion” of the 80% who may have been formally illiterate were to some degree participants in literate culture. Further, there are far more documents still to be deciphered, as well as a number that were destroyed—all evidence of literacy that needs to be taken into account. Hopkins notes that by Harris’s figures, there were over two million adult men in the Roman empire who could read, and that this large number would have exerted a significant influence upon the society as a whole. Paul, as a Tarsian of some means and status, and as evidenced from his letters (of which he wrote a part, if not all), would have been one of these two million literate men who exerted cultural (and in this case religious) influence on his surroundings.
For the purposes of this paper, more germane to Paul’s use of Scripture would be the style of reading in the culture of the time. We have seen that the culture was a complex interplay of oral and literary elements, of which written texts were clearly a part. If we assume that there were books in the ancient world, and that Paul had access to some of them (I will explore this further below), then the process by which he read them for the sake of using them is worth considering. I am not addressing the issue here of how his letters were read out, although the findings about reading in general do have application in that area.
The standard view is that there was very little silent reading in the ancient world. In biblical studies, an important article by Paul Achtemeier agrees that papyrus was widely available in the ancient world and so books were widely available through libraries and booksellers, and the widespread number of manuscripts attests to the book culture of the ancients, even in the lower classes of the Greco-Roman world. Where Achtemeier has been particularly influential is in his view that all reading was vocalized, and hence done out loud. He goes further, however, and argues that, as this was the case, all writing came about through vocalization, and hence dictation. That is, one either dictated (read aloud) to oneself or to another, in either case one writing down what was heard. Achtemeier’s view has been influential upon subsequent interpretation of Paul’s letters, as scholars have attempted to locate and identify the oral features that are to be found in his letters as a result of this dictation process.
This position was called into question almost immediately by Frank Gilliard, but the fact that his statement was relatively buried in the critical notes of JBL probably minimized its importance. Gilliard pointed out numerous earlier examples of silent reading. These included: Theseus as depicted in Euripides, Hippolytus 856-74 (fifth century B.C.); Demosthenes in Aristophanes, Knights 116-27 (fifth century); the riddle in Antiphanes, Sappho (fourth century B.C.); Alexander the Great (according to Plutarch, On the Fortune of Alexander 340A; fourth century B.C.), Julius Caesar (according to Plutarch, Brutus 5; first century B.C.), Cicero, Tusculan Disputations 5.116 and Orator 57 (first century A.D.), Suetonius, Augustus 39 (see Gavrilov below; second century A.D.), the Peripatetic work Problems 18.1, 7 (material from Aristotle, though final form in fifth century A.D.), and even Augustine himself, Confessions 8.12 (thus mitigating the significance of Ambrose being able to read silently; fourth century A.D.). The conclusion we are forced toward is that there is an awful lot of reputable evidence for silent reading over the course of the ancient world. In a discussion of the psychology of reading, A.K. Gavrilov analyzes the reading process, as well as commenting upon several of the key examples above, and providing more of his own. He adds to those above a number of passages where “silent reading is more or less certainly implied.” Gavrilov also includes lists of passages where the advantages of reading aloud or silently are listed, such as instances of lecturing where ancient writers would use notes, in which case silent reading would have been a necessity.
5. Paul and the Use of the Old Testament
At this point, I wish to try to bring together some of the strands that I have been introducing above, to see if it helps to understand the context in which Paul cites Scripture.
The first stage is to try to gather the various strands together to create a scenario in which to place Paul. Though orality was significant in the Greco-Roman world, including the world of Diaspora Judaism, it maintained a complex interplay with literacy and a growing and developing book culture. Paul was born in a city that was one of the leading cities in providing education, especially for its native youths. The nature of the educational system, with its parallel streams, provided opportunities for those like Paul. If we can assume that the portrait that emerges regarding Paul from the evidence in Acts and the letters is accurate, then Paul could quite possibly have taken advantage of the liberal arts education that began with the grammar school and finished with the rhetorical school. His being a Roman citizen from a family of Romans, with trades of value in the ancient world, so much so that he used the trade throughout his travels, indicate that he could possibly have availed himself of this educational stream. I don’t think that this is the case, but even if he were not able to be educated in the grammar school, the primary school would have possibly been available to him, where he would have gained basic literacy and numeracy, and even some basic compositional skills. However, the sequence of events within the educational system, and his move from Tarsus to Jerusalem to study with Gamaliel, indicate a pattern that fits very well with the educational system in Tarsus. If this is the case, then Paul would have been literate and capable of not only reading and writing, but composing. Besides his exposure to a range of writers, especially poets, he would have been able to perform the various requirements of the progymnasmata, been introduced to letter writing and style, and even have had some basic rhetorical training. He would have been able to read and write, probably aloud and silently. His continuing of his education in Jerusalem would have in some ways been the equivalent of the rhetorical training of the Greco-Roman school (we simply don’t know if he was a student of rhetoric at this level, but the timing and other factors indicate that he probably was not). In any case, he would have both known the written Torah and been educated in the oral Torah and how to dispute its points of law, enough as a student of Gamaliel to become a Pharisee. As a literate person, in the company of literate persons (such as those who served as scribes for his letters), he would have had access to the necessary materials for the securing, copying and writing of books. Some of these books are his letters. As a literate person and a major letter writer (I believe it is impossible for Paul not to have known that he was a major letter writer on the basis of simply seeing how his letters compared to other, more typical letters of the ancient world), he probably wrote or had written multiple copies, with copies being kept, with later copies being made from them, and with the perhaps forming the basis of his letter collection.
This raises the question of what might be referred to in 2 Tim 4:23 by the ‘parchments’. Of course, many will want to dismiss this as non-Pauline and later, but that does not answer the question of what the reference is. Whether Paul wrote these words or not, it is entirely possible that these documents might be any number of ‘books’ of the ancient world. As noted above, these documents could have consisted of individual books or they could have consisted of anthologies of the writings of others, such as Old Testament authors. These compilations in tandem with their abundant parallels among Greco-Roman anthologies provide a significant background for the methods of collecting significant texts for easy reference in antiquity. Given Paul’s educational background, we certainly would have expected him to be familiar with Greek anthology as a fundamental part of his early literacy and literary composition training. As a Pharisee, Paul would have also likely been exposed to Jewish methods of compiling important Scriptures into testimony volumes for easy reference, the propagation of theological agendas and liturgical purposes.
Paul undoubtedly employs the Greek Jewish Scriptures in most of his citations of the Old Testament. As an initial point on terminology, ‘LXX’ has taken on a variety of different meanings and is often used in ambiguous and unqualified ways in New Testament studies. This has led several recent scholars to speak of the Greek Bible in use during the time of the New Testament authors as the Old Greek text. This terminological point highlights the fact that Paul’s Greek Bible (the ‘LXX’) was not a volume Paul (or anyone else, for that matter) could have pulled down off of the shelf and looked up a passage. There is, in fact, no evidence that there was even a collection of Greek manuscripts spanning the entire Old Testament present in one place. More than likely, the Greek text was dealt with in terms of individual books and their respective scroll(s)—it did not exist as a single volume until the invention of the codex in the second century AD. These scrolls were the result of two hundred years of scholarship by diverse translators in diverse times and circumstances. They, therefore, reflect different translation philosophies and should not be understood monolithically as a coherent textually stable tradition of Scripture readily available to most first century Jews and Christians. Thus, as Stanley notes, it is highly unlikely that any one individual or establishment would have had access to the entire Jewish Old Testament in Greek. Poor relations between the Jews and Paul may have further limited Paul and his colleagues’ ability to access the Greek Scriptures. In any case, carrying around a large bag of scrolls for reference in letter composition would have been rather cumbersome in Paul’s travels.
A likely hypothesis, therefore, is that Paul or one of his early Christian colleagues compiled an anthology of significant texts for specific purposes, such as liturgical, doctrinal or compositional tools. This would not have been a foreign compositional aid to Paul, given his Greek and Jewish backgrounds. It is not hard to imagine Paul as a person who was involved in the book culture in a number of ways. He himself might have been one to copybooks. We know that Paul could write, even if he chose to use a scribe in most instances when it came to writing his letters. He may well also have been a part of a group that was concerned with Scripture and hence had access to at least Greek if not also Hebrew scriptural texts for copying. We know that he also had scribes and those accompanying him who may well have had similar interests and abilities. This same group may have been involved in the copying and then dissemination of his own letters.
When it comes to what we actually find in his letters, there is a range of evidence to be considered. However, little of it gives direct evidence of Paul’s compositional technique. There is no way to prove it, but I suspect that it involved a combination of citation from memory and from written texts, some of which may have been in anthologies and some in manuscripts of entire books. Quite possibly the citations from individual Greek authors came from either memory or such anthologies. The citations of the biblical text probably reflect all three sources. One hypothesis is that in those places where Paul cites by author or book he is citing a written source, or where he is citing a series of passages from the same book he is citing a written source, but when he is citing an author or book without attribution or bringing passages from different sources together into one he is citing these from memory. For the sake of discussion, let me cite a couple of examples that may indicate what I am suggesting. In Rom 9:12 and 13, Paul cites Gen 25:23 and Mal 1:2-3. The fact that he brings these two passages together in these brief quotations gives the presumption that they are from memory and recalled for his occasion—the first reflects the MT and the second the MT = LXX. By contrast, I think that it is worth considering whether Paul had a scroll of Psalms, or at least Psalm (and possibly more) excerpts, in front of him when he wrote Rom 3:10-18. This passage contains citations from Ps 14:1-3 (Rom 3:10-12), Ps 5:9 (Rom 3:13), Ps 10:7 (Rom 3:14), and Ps 36:1 (Rom 3:18), reflecting various features of the Greek version. Apart from the first citation, the order is biblically sequential. Paul also cites Isa 59:7-8 in Rom 3:15-17. This may have come to his memory, or it may have been a citation within a book of citations that he had in front of him. A similar example may be found in Rom 10:15, 16, where first Isa 52:7 and then Isa 53:1 is cited (cf. Rom 11:3, 4, citing 1 Kgs 19:10, 18). The text here is somewhat mixed, but again there are elements that reflect the Greek version.
This paper has been prepared as an attempt to bring the actual process of citation into the discussion. As a result, I have explored briefly the nature of the Greco-Roman educational system and how Paul may have been involved in it. The nature of his involvement has potential implications for how he cited Scripture and other authors, because it establishes the nature and types of reading that he would have done. As a result, I believe that it is plausible at least to explore the possibility that Paul’s education combined element of both the Greco-Roman grammar school and Torah training. His exposure to a range of texts, including both continuous texts and various types of anthologies and collections, helps perhaps to account for some of the features of his use of both Scripture and other ancient authors. Actual studies of such usage are of necessity preliminary, and may well always remain so, because the type of evidence that we have, while suggestive, is indirect and circumstantial. Nevertheless, I believe that it is worth exploring that Paul’s involvement in the Greco-Roman and Torah-based educational systems can help account both for the material that he cites, and the way in which he cites it.
 Some of the issues surrounding this is surveyed in S.E. Porter, “The Use of the Old Testament in the New Testament: A Brief Comment on Method and Terminology,” in C.A. Evans and J.A. Sanders (eds.), Early Christian Interpretation of the Scriptures of Israel: Investigations and Proposals (SSEJC 5; JSNTSup 148; Sheffield: Sheffield Academic Press, 1997), 79-96; idem, “Further Comments on the Use of the Old Testament in the New Testament,” in T.L. Brodie, D.R. MacDonald and S.E. Porter (ed.), The Intertextuality of the Epistles: Explorations of Theory and Practice (NTM; Sheffield: Sheffield Phoenix Press, forthcoming 2006).
 C.D. Stanley, Paul and the Language of Scripture: Citation Technique in the Pauline Epistles and Contemporary Literature (SNTSMS 74; Cambridge: Cambridge University Press, 1992).
 R.B. Hays, Echoes of Scripture in the Letters of Paul (New Haven: Yale University Press, 1989).
 J.R. Wagner, Heralds of the Good New: Isaiah and Paul in Concert in the Letter to the Romans (Leiden: Brill, 2002).
 C.D. Stanley, Arguing with Scripture: The Rhetoric of Quotations in the Letters of Paul (London: T&T Clark International, 2004).
 As Daly points out, though Strabo is probably describing primarily Greek-speaking people, he also notes elsewhere (Geography 4.5.181) that the Romans frequently did likewise. See L.W. Daly, “Roman Study Abroad,” AJP 71 (1950): 40-58, esp. 40, 55-56.
 On these cities, see C.A. Forbes, “Ancient Universities and Student Life,” The Classical Journal 28 (1933): 413-26, esp. 414 on Paul and Tarsus.
 Daly, “Roman Study Abroad,” 55.
 H.I. Marrou, A History of Education in Antiquity (trans. G. Lamb; London: Sheed and Ward, 1956), 186-205, 242-54; S.F. Bonner, Education in Ancient Rome: From the Elder Cato to the Younger Pliny (London: Methuen, 1977), 34-75, 165-276; D.L. Clark, Rhetoric in Greco-Roman Education (Morningside Heights, NY: Columbia University Press, 1957), 9-66.
 T. Morgan, Literate Education in the Hellenistic and Roman Worlds (Cambridge: Cambridge University Press, 1998), 21-25; R. Hock, “The Educational Curriculum in Chariton’s Callirhoe,” in J.-A.A. Brant, C.W. Hedrick and C. Shea (eds.), Ancient Fiction: The Matrix of Early Christian and Jewish Narrative (Symposium 32; Atlanta: Scholars Press, 2005), 15-36, esp. 17-25, even though he recognizes the evidence against such a scenario.
 A.D. Booth, “The Schooling of Slaves in First-Century Rome,” TAPA 109 (1979): 11-19, as in R.A. Kaster, “Notes on ‘Primary’ and ‘Secondary’ Schools in late Antiquity,” TAPA 113 (1983): 323-46, here 324, who notes that in the sequential account there was a recognition that many in the elementary school did not go on to the grammar school, and many grammar school students received their elementary tuition privately (although Kaster does not mention any of those who held to this modified view). See also R. Cribiore, Gymnastics of the Mind: Greek Education in Hellenistic and Roman Egypt (Princeton: Princeton University Press, 2001), 37, 45.
 Kaster, “Notes,” esp. 328-29, 336-38, 346.
 Kaster, “Notes,” 336.
 W.D. Davies, Paul and Rabbinic Judaism: Some Rabbinic Elements in Pauline Theology (Philadelphia: Fortress, 4th edn, 1980), 24-25.
 This has direct implications for interpreting Luke 4:16, which may refer to Nazareth as the place where Jesus ‘was educated,’ rather than simply ‘reared.’
 B. Gerhardsson, Memory and Manuscript: Oral Tradition and Written Transmission in Rabbinic Judaism and Early Christianity , with Tradition and Transmission in Early Christianity (Grand Rapids: Eerdmans, 1998 [1961, 1964]), 56-66, 90-89-92; cf. C. Hezser, Jewish Literacy in Roman Palestine (TSAJ 81; Tübingen: Mohr Siebeck, 2001), 39-109; A. Millard, Reading and Writing in the Time of Jesus (Biblical Seminar 69; Sheffield: Sheffield Academic Press, 2000), 188-92.
 Hezser, Jewish Literacy, 94-109, esp. 94-96.
 See G.A. Kennedy, Progymnasmata: Greek Textbook of Prose Composition and Rhetoric (Leiden: Brill, 2003).
 R.F. Hock and E.N. O’Neil, The Chreia in Ancient Rhetoric. I. The Progymnasmata (Atlanta: Scholars Press, 1986); R.F. Hock and E.N. O’Neil, The Chreia and Ancient Rhetoric: Classroom Exercises (Leiden: Brill, 2002), but who nonetheless, despite recognizing the recent shift regarding education (pp. 1-4), still use the three sequential stages in their analysis.
 See Marrou, History of Education, 169-75; Cribiore, Gymnastics of the Mind, 185-218 (cf. R. Cribiore, Writing, Teachers, and Students in Graeco-Roman Egypt [American Studies in Papyrology 36; Atlanta: Scholars Press, 1996], esp. appendixes); Morgan, Literate Education, 24; accepted by Hock, “Educational Curriculum,” 21-23. The epistolary theorists may have relevance here, as they present types or models of letters. However, the only one early enough to be of relevance does not present whole letters, but characteristic excerpts (see Pseudo Demetrius). We do not necessarily know whether these letter types were copied directly. See A.J. Malherbe, Ancient Epistolary Theorists (SBLSBS 19; Atlanta: Scholars Press, 1988)
 See, for example, S.E. Porter, “Paul of Tarsus and his Letters,” in S.E. Porter (ed.), Handbook of Classical Rhetoric in the Hellenistic Period 330 B.C.–A.D. 400 (Leiden: Brill, 1997), 532-85, esp. 578-83.
 Other references in Paul have an aphoristic quality and may have been learned in a school setting, such as 1 Cor 9:10; 2 Cor 4:6; and Eph 5:14. For other possible knowledge of ancient sources, see E.B. Howell, “St Paul and the Greek World,” Greece & Rome 2nd series 11 (1964): 7-29, esp. 22-23, used above.
 See U. Wilamowitz-Moellendorff, Die griechische Literatur des Altertums (Stuttgart: Teubner, 1912), 232-33.
 V.K. Robbins, “Oral, Rhetorical and Literary Cultures: A Response,” Semeia 65 (1994): 75-90, esp. 77. Cf. Hezser, Jewish Literacy, 16.
 W. Kelber, The Oral and the Written Gospel: The Hermeneutics of Speaking and Writing in the Synoptic Tradition, Mark, Paul and Q (Philadelphia: Fortress Press, 1983), 140-83.
 J. Dewey, “Textuality in an Oral Culture: A Survey of the Pauline Traditions,” Semeia 65 (1994): 37-64; cf. A.J. Dewey, “A Re-hearing of Romans 10:1-15,” Semeia 65 (1994): 109-27.
 See S.E. Porter, “The Reasons for the Lukan Census,” in A. Christophersen, C. Claussen, J. Frey and B.W. Longenecker (eds.), Paul, Luke, and the Graeco-Roman World: Essays in Honour of Alexander J.M. Wedderburn (JSNTSup 217; Sheffield: Sheffield Academic Press, 2002), 165-88.
 R. Thomas, Literacy and Orality in Ancient Greece (Cambridge: Cambridge University Press, 1992), 132-44; cf. idem, Oral Tradition and Written Record in Classical Athens (Cambridge: Cambridge University Press, 1989); Dewey “Textuality.”
 B.M. Metzger, “Literary Forgeries and Canonical Pseudepigrapha,” JBL 91 (1972): 3-24, esp. 4.
 K.J. Dover, Lysias and the Corpus Lysiacum (Berkeley: University of California Press, 1968), 23-27.
 L. Alexander, “Ancient Book Production and the Circulation of the Gospels,” in R. Bauckham (ed.), The Gospel for All Christians: Rethinking the Gospel Audiences (Grand Rapids: Eerdmans, 1998), 71-111, esp. 78. Cf. Hezser, Jewish Literacy, 498, who thinks that Qumran was “exceptional,” but who also notes several other archives, such as that of Babatha and Salome Komaise.
 F.G. Kenyon, Books and Readers in Ancient Greece and Rome (Oxford: Clarendon Pres, 1932), 65.
 Kenyon, Books and Readers, 80.
 P.E. Easterling and B.M.W. Knox, “Books and Readers in the Greek World,” in Eaterling and Knox (eds.), The Cambridge History of Classical Literature. I. Greek Literature (Cambridge: Cambridge University Pres, 1989), 1-41, here 19.
 T.C. Skeat, “Was Papyrus Regarded as ‘Cheap’ or ‘Expensive’ in the Ancient World?” Aegyptus 75 (1995): 75-93.
 Easterling and Knox, “Books and Readers,” 22. Cf. Kenyon, Books and Readers, 24; Millard, Reading and Writing, 164-66.
 Kenyon, Books and Readers, 24.
 Kenyon, Books and Readers, 34, 36.
 Cribiore, Gymnastics of the Mind, 140.
 Alexander, “Ancient Book Production,” 87-89.
 R.J. Starr, “The Circulation of Literary Texts in the Roman World,” Classical Quarterly 37 (1989): 313-23, here 213-16.
 Alexander, “Ancient Book Production,” 88-99. Cf. also L. Alexander, “The Living Voice: Scepticism towards the Written Word in Early Christian and Graeco-Roman Texts,” in D.J.A. Clines, S.E. Fowl and S.E. Porter (eds.), The Bible in Three Dimensions (JSOTSup 87; Sheffield JSOT Press, 1990), 221-47.
 H.Y. Gamble, Books and Readers in the Early Church: A History of Early Christian Texts (New Haven: Yale University Press, 1995), 85.
 Millard, Reading and Writing, 161. Millard notes that Herod had a library of at least 44 Greek author, and that Philo must have had his own collection of Scriptures, which he used (in Greek!).
 Easterling and Knox, “Books and Readers,” 34.
 Easterling and Knox, ‘Books and Readers,” 35.
 See S. Usher, Greek Oratory: Tradition and Originality (Oxford: Oxford University Press, 1999).
 Hermogenes (On Ideas) provides the most extensive survey of classical canons and genres from an ancient source. For analysis of On Ideas 2.10-12 as well as the contribution of the Hermogenean analysis of literature more broadly see I. Rutherford, “Inverting the Canon: Hermogenes on Literature,” Harvard Studies in Classical Philology 94 (1992): 355-78.
 See S. West, Ptolemaic Papyri of Homer (Papyrologica Coloniensia 3; Cologne: Westdeutscher Verlag, 1967).
 See C.D. Stanley, “Paul and Homer: Greco-Roman Citation in the First Century CE,” NovT 32 (1990): 48-78, here 51-52; Morgan, Literate Education, 144-51.
 According to Quintilian, memorization is the most important element of education. Quintilian, Inst. or. 1.1.5; 1.3.1; cf. also Morgan, Literate Education, p. 352; on the function of imitation in advanced levels of education, see Cribiore, Gymnastics, 220-44.
 Stanley, “Paul and Homer,” 54.
 For a listing of Homeric papyri, see F. Uebel, Archiv für Papyrusforschung 24/25 (1976): 191.
 For analysis of the form and content of the papyri, see T. Renner, “Three New Homerica on Papyrus,” Harvard Studies in Classical Philology 83 (1979): 311-37, here 331-37.
 Renner, “Homerica,” 333.
 See Morgan, Literate Education, 120-51: non-school text anthologies include ps.-Phocylides, Menandri Sententiae, Sentences of Sextus, Distichs of Cato and Stobaeus. The Pythagorean Golden Verses is one of the few anthologies which was focused around a single doctrinal system (Morgan, Literate Education, 121).
 Morgan, Literate Education, 120.
 Plato, Leg. 810e-812a contains a reference to an ethical anthology within the classical period which would indicate the selection of ethical texts for educational purposes (Morgan, Literate Education, 121).
 Cf. Morgan, Literate Education, 123.
 Morgan, Literate Education, 125.
 Morgan, Literate Education, 144.
 For a surveys and lists of Greek anthologies, see J. Barns, “A New Gnomologium: With Some Remarks on Gnomic Anthologies,” Classical Quarterly 44 (1950): 126-37, esp. 133-137; idem, “A New Gnomologium: With Some Remarks on Gnomic Anthologies, II,” Classical Quarterly, New Series, 1 (1951): 1-19.
 Barns, “New Gnomologium I,” 132-34.
 Barns, “New Gnomologium I,” 134.
 Compilations for pleasurable reading include P.Petrie. II.49; P.Tebt. 3; P.Freiburg I.4; P.Oxy. 622; P.Oxy. 671. Barns, ‘New Gnomologium I’, pp. 134-35. See the above discussion on gnomologies for various compilations and schooltexts used in education. The literary sources quoted above also evidence the use of anthologies in composition.
 On the formation of the rabbinic canon, see D. Kraemer, “The Formation of Rabbinic Canon: Authority and Boundaries,” JBL 110 (1991): 613-30.
 See R. Hodgson, “The Testimony Hypothesis,” JBL 98 (1979): 361-78, here 362-65.
 See G.J. Brooke, Exegesis at Qumran: 4QFlorilegium in its Jewish Context (JSOTSup 29; Sheffield: JSOT Press, 1985), 169-78.
 So Heszer, Jewish Literacy, 21.
 Such as Eric Havelock in his The Literate Revolution in Greece and its Cultural Consequences (Princeton: Princeton University Press, 1982) and The Muse Learns to Write: Reflection on Orality and Literacy (New Haven: Yale University Press, 1986).
 E.g. see M. Beard (ed.), Literacy in the Roman World (Journal of Roman Archaeology Supplement Series 3; Ann Arbor: University of Michigan Press, 1991), with essays by Beard, Bowman and Hopkins; Millard, Reading and Writing. The major exception is Heszer, who argues for a significantly lower literacy rate among Jews in Palestine, perhaps as low as 2-3% (Hezser, Jewish Literacy, 496-504). However, Paul was not a Palestinian Jew, but a member of the Roman world, and is considered in that light here.
 W.V. Harris, Ancient Literacy (Cambridge, MA: Harvard University Pres, 1989), 266-67 for summary.
 The example is from A.K. Bowman, “Literacy in the Roman Empire: Mass and Mode,” in Beard (ed.), Literacy in the Ancient World, 119-31, here 122, see also 121. Harris does take this into account (Ancient Literacy, 124).
 K. Hopkins, “Conquest by Book,” in Beard (ed.), Literacy in the Ancient World, 133-58, here 135.
 Another factor to consider but that is not directly germane here is that the term ‘illiterate’ was used in Greek to refer to those who did not know how to read or write Greek. They may have been literate in other language, but those knowing such languages were considered ‘barbarians’. See H.C. Youtie, “Agrammatos: An Aspect of Greek Society in Egypt,” Harvard Studies in Classical Philology 75 (1971): 161-76.
 See Harris, Ancient Literacy, 125; cf. Thomas, Literacy and Orality.
 P. J. Achtemeier, “Omne Verbum Sonat: The New Testament and the Oral Environment of late Western Antiquity,” JBL 109 (1990) 3-27, esp. 15-19. The example of the reading silently of Bishop Ambrose in the fourth century is often cited. This point is addressed by M. Slusser, “Reading Silently in Antiquity,” JBL 111 (1992): 499, who cites another example from the fourth century.
 E.g. see F.D. Gilliard, “More Silent Reading in Antiquity: Non Omne Verbum Sonabat,” JBL 112 (1993): 689-94, who notes previous discussion among biblical and classical scholars, where similar confusion reigned.
 Gilliard, “More Silent Reading,” 690-92. Gilliard also cites Bernard Knox (“Silent Reading in Antiquity,” Greek, Roman and Byzantine Studies 9 : 421-35, esp. 421-22) who speculates on the absurdity of thinking that every ancient author read everything aloud, such as Didymus’s 3000 volumes! (p. 692).
 A.K. Gavrilov, “Techniques of Reading in Classical Antiquity,” Classical Quarterly NS 47 (1997): 56-73, esp. 70-71. See Herodotus 1.123-25; Euripides, Iphigeneia in Aulis 34ff.; Iphigeneia in Taurus 763; Aristophanes, Nubians 23; Birds 960ff.; Rn. 51-52; Xenophon, Symposium 4.27; Plato Comicus = Athenagoras 1.8.5b; Menander, Epit. 211ff.; Herodas 4.21-25; Plautus, Bacch. 729-995; Cicero, Fam. 9.20; Horace, Sat. 2.5.51ff.; Ep. 1.19.34; Ovid, Met. 9.569; Petronius, Sat. 129; Quintilian, Inst. or. 1.1.33-34, 10.3.25; Josephus, Life 219-23; Plutarch, Cato Minor 19, 34; Brut. 36.1-3; Ant. 10; Pliny, Epistle 5.3.2, Lucian, JTr. 1; Achilles Tatius 1.6; 5.24; Philostratus, Life of Apollonius of Tyana 8.1, 31; 4.17; Aristaenetus 1.10.36ff.; Cyril of Jerusalem, Procatechesis 14; Ambrose, Ep. 47; Possidius, Vita S. August. 31.
 Cf. Millard, Reading and Writing, 76-77; S.E. Porter, “Paul and the Process of Canonization,” in C.A. Evans and E. Tov (eds.), Exploring the Origins of the Bible (Grand Rapids: Baker, in press ).
 As McLay observes, “Sometime ‘LXX’ refers to the reading in the Greek Jewish Scriptures that has been judged by the editor of a critical text to be most likely the original reading, that is, what is believed to be the closest approximation that we can make to what was probably written originally by the translator. In other cases ‘LXX’ may refer to any reading that is found in any Greek manuscript of the Jewish Scriptures, which is not necessarily the original or even a very early reading. It could be any reading or word that appears in any Greek manuscript of a book in the LXX. In the same way, it is often stated that the NT writer quotes the ‘LXX’ version of a biblical text, as opposed to the Hebrew version or the MT, without any qualification.” R.T. McLay, The Use of the Septuagint in New Testament Research (Grand Rapids: Eerdmans, 2003), 5; see also L. Greenspoon, “The Use and Abuse of the Term ‘LXX’ and Related Terminology in Recent Scholarship,” BIOSCS 20 (1987): 21-29; C.D. Stanley, “‘Pearls before Swine’: Did Paul’s Audience Understand his Biblical Quotations?” NovT 41 (1999): 124-44, here 126-128.
 Cf. Stanley, “Pearls,” 127.
 For a recent survey of the history and transmission of the Greek Jewish Scriptures, see N. Ferna,ndez Marcos, The Septuagint in Context (Leiden: Brill, 2001). A less technical treatment is found in K.H. Jobes and M. Silva, Invitation to the Septuagint (Grand Rapids: Baker Academic, 2000); cf. also McKay, Use of the Septuagint, 100-136.
 Stanley, “Pearls,” 127.
 Stanley, “Pearls,” 127.
 There have been some who have taken Paul’s signing in Phlm 19 as if it were an acknowledgment of his illiteracy. This reflects a failure to understand the illiteracy formula, and instead indicates that Paul is legally affirming what has been said above. See T.J. Kraus, “Eine vertragsrechtliche Verpflichtung in Phlm19. Duktus und juristischer Hintergrund,” in J. Frühwald-König et al. (eds.), Steht nicht geschrieben? Studien zur Bible und ihrer Wirkungsgeschichte (FS G. Schmuttermayr; Regensburg: Verlag Friedrich Pustet, 2001), 187-200. The same can be said for Gal 6:11-18. Contra Dewey, “Textuality,” 52-54.
 A similar view is suggested, but rejected, by McLay, Use of the Septuagint, 26-30.
 I use the information provided on the quotations by G.L. Archer and G. Chirichigno, Old Testament Quotations in the New Testament (Chicago: Moody, 1983), loc cit.