Voyage of the Paradigm Ship

PROLOGUE

The following is the text of a proposal I prepared in 2004 for a Research Methods course while a graduate student in the School of Information Resources and Library Science, University of Arizona (Tucson). I offer it here as a follow-up to my post of February 22, 2009: “When you’re used to paper rolls it takes some time to convert to turning pages of a book.”

The proposal is entitled “Christians and the Codex: ‘Early Adopters’ of New Media Technology.” It appears from available manuscript evidence of the first few centuries of the Common Era that Christian communities—as a minority social group—showed a distinct preference for the codex book form than what was true for majority Roman society. It is a question of some interest why this was so, and has led to further questions into what role, if any, these Christian communities may have played in contributing to the eventual diffusion and adoption of the codex into majority society, and its subsequent triumph as the ubiquitous book form we know today. Many scholars have pursued answers to these questions. My purpose was to possibly offer another methodological lens for looking at the problem. Of course, I was spared the real hard work because the assignment was to develop the proposal, not conduct the actual research.

I find these historical and cultural questions intriguing in their own right, and maybe one day I will get around to some of that research work. At the moment, the most compelling aspect—as I introduced in my earlier post—is to try to imagine, by analogy, the dynamic at play as we consider adoption of the e-book as a new book form technology.

Harry Y. Gamble, Professor in Religious Studies at University of Virginia, a scholar who has inquired into these questions, and whose writing informed my proposal, later captured the essence of that dynamic (though working in reverse) when he wrote:

The gradual transition in Western Antiquity from the time-honored bookroll to the codex followed upon and imitated the popularity of the leaf-book in early Christianity, and specifically in connection with the formation of the Christian Bible. This was a monumental change in the history of the book. In significance it is sometimes compared to the invention of movable type in the fifteenth century…But the transition to the codex is perhaps more aptly compared to the advent of electronic texts in the late twentieth century, which even now is revolutionizing the way texts are made, accessed, and stored. In both cases we have to do with a major change in the format of texts, and consequently in the means of producing and using them. It may well be that for readers of the not-to-distant future the word “book” will automatically conjure the meaning “e-book” rather than the leaf-book that it suggests to us, or the bookroll that it designated in pre-Christian Antiquity. (Bible and book. In M. P. Brown (Ed.), In the Beginning: Bibles Before the Year 1000 (p. 16, emphasis mine). Washington, D.C.: Freer Gallery of Art & Arthur M. Sackler Gallery (distributed by Smithsonian Books), 2006.)

It turns out that a clear and definitive answer to the question of why early Christians preferred the codex book form continues to evade scholarship. Even a recent work by Larry Hurtado (The Earliest Christian Artifacts: Manuscripts and Christian Origins. Grand Rapids: William B. Eerdmans, 2006) ultimately disappoints. After insightfully critiquing the proposals of others, Hurtado himself sends us away largely empty-handed. Toward the end of a section entitled: Why Did Christians Prefer the Codex? Hurtado writes: “It is not my primary purpose here to argue for a particular answer to the questions involved…[M]y main emphasis in this discussion is that the early Christian use of the codex is an important matter worthy of attention by all scholars concerned with Christian origins” (pp. 80-81). O-O-O-K…

Maybe the question is unanswerable given the current state of our knowledge. The physical evidence in favor of the codex is indisputable. But motivational questions (Why?) are notoriously difficult for historians to answer. My research proposal would likely lead to a similar dead-end. But still I wondered whether a social-scientific approach might yield a different way of looking at the available evidence.

I decided to share my proposal essentially as I wrote it in 2004. I would be interested in your comments. Compared to five years ago, I think I am now very much more interested in wondering if a diffusion of innovation study on the adoption of the codex book form could provide us with any insights into the process of how the e-book might gain ascendancy as the book form of choice in our society. Notice my reference to reading an e-book at the time—George Orwell’s 1984—on a Palm handheld device! I think I meant that to be vaguely ironic.

Christians and the Codex: ‘Early Adopters’ of New Media Technology
Research Proposal
Gary F. Daught
April 26, 2004

Introduction
For at least 1,500 years the dominant and virtually unrivaled form of the book in Western civilization (and beyond) has been the codex. Roberts and Skeat (1983) concisely define a codex as “a collection of sheets of any material, folded double and fastened together at the back or spine, and usually protected by covers” (p. 1). The sheer longevity of the codex book form evidences it as an effective and efficient media technology. Subsequent inventions in printing, paper, and binding further revolutionized the book production process. But these have been improvements of degree. The essential form of the codex has remained unchanged.

I intentionally stress the codex as a book form as making a distinction from book content. It is true that the codex form has an impact upon how content in a book is organized and presented, but content itself is largely independent of form (which is the container and delivery mechanism of content). There were and are alternatives for delivering book content. Considering the long and ubiquitous nature of codex use worldwide this is no longer easily appreciated. Most people are not even familiar with the word “codex,” but that does not stop them from using the word “book” as a virtual synonym.

It is interesting to consider the codex as the dominant and entrenched book form at a time when a revolution in electronic and digital technology is presenting us with potential new book alternatives. I am currently reading George Orwell’s 1984 as an e-book on my Palm handheld computer. Is it possible that digital book content, delivered via various electronic devices will eventually unseat the codex as the dominant form of the book? How would such a profound shift come about?

In the present study proposal, I do not intend to address the developments in digital media technology that are leading to new book forms. I only raise the question as an analogy to prime my thinking about how technological change of this nature is adopted by a community or society. (This may come full-circle, however, when I consider how the results of my study might be generalized.) I am much more interested in the second question. But again, I appreciate that such a question involves a complex set of answers that touch on numerous aspects of human society and culture. I will necessarily limit my scope to an investigation into a particular case of early adoption of the codex book form that may have contributed to its eventual dominance.

Background to the Problem (integrated with a Literature Review)
The codex appears to have been a Roman invention, whose formal predecessor (of much greater antiquity) was the tablet notebook. The tablet notebook was composed of thin wooden boards joined together at one edge by a cord, and whitened to receive carbon ink or coated with wax to receive inscription with a stylus. The word codex (plural codices) derives from the Latin for a piece of wood, likely referring to the tablet notebook. A later development replaced wooden boards with sheets of leather, parchment (membranae), or papyrus (from Egypt). This development made the notebook lighter, easier to handle, and capable of greater capacity. But the tablet notebook was used primarily for impermanent writing activities, such as recordkeeping, transcription, and school exercises. As Gamble (1992) writes: “[In] its purely functional status as a notebook [the codex] was not immediately regarded as a proper book, that is, a medium of literature” (p. 1067).

A late first and second century (CE) satirist by the name of Martial, along with a book publisher named Secundus, appears to have made an early attempt to generate interest in the codex as a legitimate literary book form. As a true entrepreneur, Martial even wrote poetic “commercials” to promote the advantages of this particular form:

You want to take my poems wherever you go,
As companions say, on a trip to some distant land?
Buy this. It’s packed tight into parchment pages, so,
Leave your rolls at home, for this takes just one hand! (quoted in Casson, p. 104)

The last line of the poem makes the key point. Martial’s parchment codex was an innovation proposed against the traditional and widely accepted literary book form, the roll (Latin uolumen, our ‘volume’). Notwithstanding the practical advantages of the codex, Martial and Secundus’ “experiment…of practical Latin genius” was deemed a failure in the face of the very strong influence of Greek cultural conventions upon the Romans of the second century (Roberts and Skeat, p. 29).

Although parchment (made of processed animal skins) eventually replaced papyrus for (roll and codex) book manufacture, scholarship has laid to rest the long-held assumption that papyrus was somehow an inferior material. The issue seems rather to have turned on supply at various times and localities. Egypt was the Empire-wide producer of papyrus. But local disruptions (e.g., during times of war) could jeopardize supply. Meanwhile, parchment (once the procedure for its manufacture was mastered) could be produced anywhere. It is this fact, not the comparative quality of material, which may have eventually won the day for parchment (Roberts and Skeat, pp. 5-10). However, in the period under investigation (1st to 5th centuries) codices of both parchment and papyrus were common.

Martial appears to have failed to generate sufficient interest in the codex as a literary book form in Rome. Indeed, analysis of available manuscript evidence (excluding Christian literature) suggests subsequent adoption of the codex occurred only very gradually. Parity between roll and codex was not attained until roughly two centuries later (300 CE). During the 4th century the figure rises to about 74%. And by the 5th century the figure is roughly 90%.

The picture appears dramatically different when considering specifically Christian literature, where notebook-style codices (primarily of papyrus) are being utilized as the preferred form already by the beginning of the second century. Of the available manuscript evidence dated to the second century roughly 90% are codices! (See further in Roberts and Skeat, pp. 38-44; Gamble (1995), pp. 49-66; Turner, pp. 89-185; Aland and Aland, pp. 75-76, 83-102.) Gamble (1995) suggests an initial transitional use for codex notebooks in the Christian community—a sort of hybrid between notebooks and literary books. The aim was to secure written copies of important (apostolic) teachings for practical congregational use. But as the authority of these written teachings grew, the form associated with them was adopted as well.

Broad Problem Statement
The Christian community did not invent the codex book form. Yet by the second century CE it had already adopted the codex as the preferred medium for recording and disseminating its literature. This happened precisely at a time when the general Greco-Roman society did not regard the codex as an appropriate medium of literature, deferring instead to the convention of the roll. The codex was eventually adopted in the general society, but not in a majority way until the 4th century. The so-called practical advantages of the codex over the roll do not by themselves account for its ready acceptance among the Christians. Indeed, these advantages, which would have been observable by the general society, failed to generate acceptance there. Is it possible to account for the Christian community’s rapid adoption of the codex book form for literary purposes when the general society did not? Can it be demonstrated that the Christian community influenced the eventual adoption of the codex in the general society?

Theoretical Framework (integrated with a Literature Review)
Viewed conventionally, historical study seeks to reconstruct the “what really happened” of past events. Historical study has always and continues to capitalize on the retrieval and interpretation of relevant primary documents and artifacts in consultation with the parallel contemporary literature. Though no potential source of reconstructive evidence is dismissed out of hand, historians today are much more discerning and appreciate that there is no such thing as disinterested or purely “objective” historical reporting.

At the same time, historians are discovering new and fruitful insights about the past by applying methodologies typically directed to more modern anthropological, sociological, political and economic issues and problems. For example, social historian Wayne Meeks (1983) has compellingly applied social theory to the study of the early Christian movement. In rationalizing his method, Meeks writes:

To be sure, ordinary Christians did not write our texts and rarely appear in them explicitly. Yet the texts were written in some sense for them, and were used in some ways by them. If we do not ever see their world, we cannot claim to understand early Christianity…Since we do not meet ordinary early Christians as individuals, we must seek to recognize them through the collectivities to which they belonged…. It is in the hope of accomplishing this that a number of historians of early Christianity have recently undertaken to describe the first Christian groups in ways that a sociologist or anthropologist might. (p. 2)

In a similar vein, it occurred to me that perhaps the social theory of Everett M. Rogers on the ‘diffusion of innovations’ might be applicable to the current discussion. (Rogers’ book by the same title (originally published in 1962) is now in its 5th edition (2003). I had access to the 4th edition (1995).) It was Rogers who coined the term ‘early adopter,’ a term that has found its way into common parlance, especially when speaking of new technology.

Rogers defines diffusion as “the process by which an innovation is communicated through certain channels over time among the members of a social system” (p. 5). This definition exposes 4 main elements of the diffusion process: 1) the innovation; 2) communication channels; 3) time; and 4) a social system. These elements can be summarized as follows:

1) The Innovation: “an idea, practice, or object that is perceived as new by an individual or other unit of adoption” (p. 11, emphasis mine). The innovation may not be the thing itself but the use to which the thing is put. The parchment or papyrus codex was a new technology, but it existed alongside the literary book roll in its use as a notebook. The innovation came when someone (e.g., Martial) proposed that the codex could also be used as a literary book form. Rogers identified 5 characteristics of an innovation (pp. 15-16):

• Relative advantage: “[T]he degree to which an innovation is perceived as better than the idea it supercedes…The greater the perceived relative advantage of an innovation, the more rapid its rate of adoption will be.” Martial pointed to the advantages of his codex book editions over the roll. But other factors worked against him.
• Compatibility: “An idea that is incompatible with the values and norms of a social system will not be adopted as rapidly as an innovation that is compatible.” The codex notebook was not perceived to be a proper book form for literature. The Romans wanted to demonstrate that they were capable of being a cultured society. They may have adopted their bias for the roll under Greek influence.
• Complexity: “New ideas that are simpler to understand are adopted more rapidly than innovations that require the adopter to develop new skills and understandings.” Martial was able to secure a publisher who was willing (at least initially) to adapt scribal conventions to the new form. But the codex presented certain new challenges for the scriptorium. Too, “the [supposed] disadvantages of the roll…may well not have been so apparent to the second-century public” (Roberts and Skeat, p. 74; see also pp. 51-52, 73-74; and Turner, pp. 73-74).
• Trialability: The opportunity to “test drive” a new innovation without needing to fully commit to it reduces uncertainty and enhances the prospect of adoption. In addition to his own poetry, Martial appears to have promoted codex versions of classic works. Though Roberts and Skeat reject his notion, Kenyon (1932) suggests these “were not ordinary copies of the authors named, but were miniatures of some sort, presumably either extracts or epitomes [anthologies]” (p. 93). If so, could these have been published as (low cost) promotions?
• Observability: “The easier it is for individuals to see the results of an innovation, the more likely they are to adopt it.” Martial’s ability to secure a publisher/bookseller (he even gives the address to Secundus’ shop in his poem! [Casson, p. 105]) was an important step in moving private use of notebook codices into the public.

2) Communication channels: Communication is essential to the diffusion of innovations. Rogers identifies mass media channels and interpersonal channels of communication. Mass media is most effective for creating “awareness-knowledge” of an innovation. However, most people tend to make adoption decisions based on interpersonal interaction. Rogers uses the term homophily (and heterophily, its opposite) to describe “the degree to which two or more individuals who interact are similar in certain attitudes, such as beliefs, education, social status, and the like…More effective communication occurs when two or more individuals are homophilous.” Roberts and Skeat make an interesting comment regarding the Martial example: “It has been observed that the authors who appear in [Martial’s codex book] format are all classics and it is likely enough that the fashionable author or discriminating bibliophile would not readily accept a format which suggested the lecture-room or the counting-house; the inference is that these volumes were designed to appeal rather to the literate bourgeoisie” (p. 25).

3) Time: Simply put, the “innovation-decision process”—the point of an innovation’s inception to its adoption or rejection—requires time. Adoption (y-axis) is plotted over time (x-axis), and the resulting distribution is an S-shaped curve of various slopes depending upon the rates of adoption. Based upon analysis of available manuscript evidence, the diffusion and ultimate (majority) adoption of the codex as an appropriate literary book form into the general Greco-Roman society took roughly 200 years, but it does result in an S-curve. We don’t know to what degree Martial’s innovation influenced this eventual adoption. In the short-term, at least, his idea seems to have been quickly rejected. But in another sense, his contribution would have been added to all such efforts that led to eventual adoption. Rogers notes that the diffusion curve “takes off”—reaches a “critical mass”—at about a 10-25% adoption rate (p. 12). For the codex, this “critical mass” rate was not reached until sometime in the 3rd century. However, as noted, by the 4-5th century the adoption rate had shot up to 90%.

Rogers further defines the “innovation-decision process” by an individual or decision-making unit as involving 5 steps: a) coming to some knowledge of the innovation’s existence and understanding of it’s function; b) formation of a favorable or unfavorable attitude toward the innovation as a result of some persuasion; c) making a decision about adopting or rejecting the innovation; d) implementing or putting the innovation to use; and e) seeking reinforcement, or confirmation about the decision already made, which may lead to a reversal of decision “if exposed to conflicting messages about the innovation” (p. 20).

4) A Social System: Rogers notes that “diffusion occurs within a social system,” which he defines as “a set of interrelated units that are engaged in joint problem-solving to accomplish a common goal. The members or units of a social system…cooperate at least to the extent of seeking to solve a common problem in order to reach a mutual goal.” This is an involved discussion that deals with the complex interactions of social structure, social norms, change agents, decision-making processes, and the consequences of adoption or rejection of an innovation.

Rogers specifies 5 categories of adopters within a social system that are classified on the basis of innovativeness, which is “the degree to which an individual or other unit of adoption is relatively earlier in adopting new ideas than other members of a social system” (p. 261). As it happens, the distribution of these successive categories follows a Gaussian (bell-shaped) curve: the first 2.5% are called the innovators (and given the generalized character descriptor venturesome); the next 13.5% are early adopters (respect); the next 34% are called the early majority (deliberate); the next 34% (to the right of the mean) are called the late majority (skeptical); and the last 16% are called laggards (traditional) (pp. 262-280). These adopter categories would essentially line-up along the y-axis as a distribution accounting for 100% of a social system over the course of the innovation-decision process. When dealing with an ancient social system, like the Martial example, we have to infer from available evidence (which is limited) to categorize the social system along these lines. Again, Martial did not invent the notebook codex, but he did put it to a new and innovative use. So, we can say that he was an innovator. The bookseller Secundus was apparently early-on persuaded by Martial to publish his codices, so he would either join Martial as an innovator, or at the very least, an early adopter. The people who took a risk to purchase Martial’s books when they became available for sale would also be categorized as early adopters. In the short-term, the remainder of the populace in Rome that was even aware of Martial’s innovation but who rejected it (the majority) would, at best, be categorized as part of the skeptical late majority, but more probably, as traditionalist laggards. The distribution does not follow as expected if the diffusion process fails to garner enough adopters to reach “critical mass.” Of course, in this case we don’t have enough information to be more specific.

Research Hypothesis
The innovation/adoption of the notebook codex for literary book use in the Christian community (of the late 1st and 2nd century CE) occurred independently of Martial’s innovation. The literature cites Martial for shedding light on the historical development of the codex. I cite him as a contemporary example for putting the application of Rogers’ diffusion on innovations theory into a parallel/comparative context. While not proposing a new answer to the problem (I am fairly persuaded by the hypothesis put forward by Gamble (1995), pp. 58-66), Rogers’ theoretical framework does provide a new perspective for interpreting the available evidence by illuminating the social dynamics of innovation diffusion and adoption (or rejection, as in the case of Martial). I suggest that Martial was attempting to promote his innovation into a generalized social environment—the literate citizenry of the cosmopolitan city of Rome (which, at the same time, was biased by Greek cultural precedent toward use of the roll). Though he appears to have capitalized on mass media communication channels, his ‘market’ was heterophilous. The Christian community, by contrast, sought to differentiate itself from a generalized social environment. Though culturally diverse (Jewish, Greek, Roman, etc., drawn from general society), the Christian community sought to gather itself around a set of unified religious values that were intentionally aimed at transcending (or relativizing) those cultural differences. I suspect that communication channels were more interpersonal and homophilous, and the construct of their social system was more congenial to an innovation that was compatible with and reinforcing of their internal values and norms. This focus on internal values and a differentiated identity over against “the outside world” accelerated adoption of this initially practical and eventually characteristic technological medium—the codex as a literary book form. Once established internally, use of the codex became observable by the general Greco-Roman society (through Christian missionary efforts, etc.). While not causing its general adoption, the improved status of the Christian community and the recognition of the Christian religion in the Empire by the early 4th century certainly contributed to the stabilization of the codex book form from that time into the present.

Research Methodology
A diffusion study is interested in determining innovation adoption (or rejection) rates as a percentage of the total population over time. The theoretical framework is largely descriptive and observational. This could be handled as a casual-comparative study or field research. Since I want to understand the dynamic of how motivations, attitudes, and social and interpersonal relationship networks affect innovation adoption, the research to be conducted will be qualitative, though my hypothesis is deduced from the theoretical framework. If I were conducting this research on a contemporary population, I would select a representative sample that I could track over time—through the entire innovation-decision process. But depending upon the innovation being studied, it may be difficult to estimate in advance how much time would be required to see the process through. (My study effectively tracks adoption rates over the course of better than four hundred years!) There is no need to set-up an experiment or establish a control group. The innovation is the independent variable. A number of dependent variables, in addition to time (rate of adoption), can be identified (e.g., “beliefs, education, social status, and the like”) that would be hypothesized to influence adoption rates. It may be possible to operationalize these. (In my study, I will need to infer and generalize regarding variables like these, and will have to be content with conceptual definitions.) I would not want to intrude upon the innovation-decision process, so selection of the representational sample and application of appropriate instruments—questionnaires and interviews—may need to await post-processing. (I have not studied the application of the theoretical framework enough to glean how diffusion is researched as it happens.) Availability of other in-process data would depend upon the nature of the innovation being studied. But these could include media promotions, purchase records, meeting minutes, press releases and press conference transcriptions, letters and other correspondences, news reports and editorials, sermons and speeches, etc.

I am assuming at the start of the diffusion study that the roll was viewed as the “proper” literary book form. The innovation under study is the codex employed for literary (and not just practical) use. The populations I am studying—the communally-literate Christian community and the generalized literate (Greek and Latin) population of the Roman Empire in the late 1st through 5th centuries CE—are not directly accessible to me. What I do have, however, is access to manuscript evidence and contemporary writings that can be surveyed for attitudes relating to book form and use. I also have a context of remarkable historical breadth. The manuscripts comprise my sample. An obvious and severe limitation I face is that the current manuscript evidence (numbers) is relatively sparse, condition of the manuscripts may compromise their usefulness, and there is no way to know to what degree these are representative of the actual situation at the time. Still, the data currently available does suggest a sustained trend over time, which should enhance confidence as long as conclusions are kept tentative.

The manuscript evidence has been subjected to paleographic analysis that seeks to place a date (not an exact science!), origin, and identification of form, text and subject matter. Based on text and subject matter, the manuscripts would be sorted to their respective population. (Though Jewish and Christian communities share some of the same texts [in the Greek language], there are observational techniques available for sorting these.) Since my diffusion study is interested in adoption of form over time, the available manuscripts would then be dated and arranged chronologically (interval data) 1st through 5th centuries CE. The manuscripts in each population and century rank would then be counted (nominal data) for totals in each century. Finally, percentages of rolls and codices to the total number in each century within each population would be determined. The results are plotted and the (expected) S-curve of adoption rate for the codex within each population can be demonstrated and their slopes compared. The steeper the slope, the quicker the adoption rate.

Of course, results (which other researchers have previously reported) do not answer the questions of how and why. Interpretation of data is provided a new lens with a sociological approach like the diffusion of innovations theoretical framework. Available contemporary literature over the course of the time period under investigation would be surveyed with special attention to the hints provided by all those involved in the innovation-decision process, not only for the Christian community, but for the general Greco-Roman society as well.

Generalization of Research
Is an historical study generalizable? As I suggested in the introduction, studying the process by which one long-recognized book form gave way to a new technology may provide insight into how the next significant technological form may gain the necessary adoptive “critical mass” to provoke an eventual shift. Convention dictated form in the general society of the Greco-Roman world, and so resistance to adopting a new technology was strong. Change took much longer. The rapidity of adoption in the Christian community was related, in an intensely significant way, to the wedding of function and form. But that intense combination does not appear to be at play in our general society at present. Innovators of new digital book forms may be doing better than Martial, but I suspect there is still a long way to go.

Reference List
Aland, K. & Aland B. (1987). The text of the New Testament (E. F. Rhodes, Trans.). Grand Rapids, MI: Eerdmans. (Original work published 1981)

Casson, L. (2001). Libraries in the ancient world. New Haven: Yale University Press.

Gamble, H. Y. (1992). Codex. In Anchor Bible Dictionary (Vol. 1, pp. 1067-1069). New York: Doubleday.

Gamble, H. Y. (1995). Books and readers in the early church: A history of early Christian texts. New Haven: Yale University Press.

Kenyon, F. G. (1932). Books and readers in ancient Greece and Rome. Oxford, England: Clarendon Press.

Meeks, W. A. (1983). The first urban Christians: The social world of the Apostle Paul. New Haven: Yale University Press.

Roberts, C. H. & Skeat, T. C. (1983). The birth of the codex. London, England: The British Academy/Oxford University Press.

Rogers, E. M. (1995). Diffusion of innovations (4th edition). New York: Free Press.

Turner, E. G. (1977). The typology of the early codex. University of Pennsylvania Press.