Book Work: Toward an Extended Codicological Intentionalism Free

This essay theorizes codicological intentionalism as an array of critical methods for interpreting medieval manuscripts as self-sufficient objects of study that condition the meanings available to the texts they contain. Building upon recent work published by Sebastian Sobecki, it argues for a shift in attention from texts to books as primary objects of inquiry. By extending codicological intentionalism to the production of books, scholars can recover historically situated meanings of texts in specific manuscript contexts, relying on evidence left by their immediate scribal makers, rather than relying upon internal documentary evidence about their originating textual authors.

Recently, in the pages of this journal, Sebastian Sobecki set out initial terms for theorizing codicological intentionalism as a method for interpreting medieval texts through the examination of manuscript witnesses.¹ His publication in JMEMS was accompanied by a separate case study, published concurrently in Speculum, which used codicological intentionalism to ground an understanding of how medieval literary narratives might reflect the textual conditions of their composition.² Both articles develop Sobecki's prior account of the “indexical self,” wherein “the narrative ‘I’ in such texts [i.e., texts featuring an authorial persona] is deictic and thus contingent on its biological author.”³ In both articles, Sobecki describes codicological intentionalism as a critical praxis in medieval studies that leverages the medieval manuscript book as evidence for authorial intention. Sobecki's recent research record demonstrates how valuable the approach can be for archival discovery and literary history alike.⁴

In this essay, I propose to extend the scope of codicological intentionalism in order to examine the manuscript book, rather than the manuscript text, as a primary object of study and vehicle for interpretation. Throughout, I engage directly and at length with Sobecki's reasoning, specifically because his recent publications in two major journals provide a significant foray into committed theorization of interpretive methods surrounding manuscript texts and their conjoint authors. In Sobecki's account, evidence for intention preserved by individual codices, by way of the text or texts they contain, ultimately resolves to originary authorship. There, codicological intentionalism has primarily to do with texts as malleable constructs subjected to and produced from mouvance, such that each manuscript text refracts unevenly some informing authorial intention.⁵ Manuscripts are witnesses to texts, but individual manuscript texts may not themselves accurately represent authorial productions.

The extended formulation I theorize below frames codicological intentionalism as an array of critical methods for interpreting medieval manuscripts as self-sufficient objects of study that condition the meanings available to the texts they contain. Rather than taking textual witnesses as evidence for intention in authorial production, I understand manuscript books as evidence for intention in scribal production. The intention preserved by each codex resolves directly, if complexly, to the scribal authors who produced it. My revision to codicological intentionalism, in this way, posits a more focused mode of interpreting manuscript texts without abandoning the historicist commitment to the recovery of intention central to Sobecki's formulation.

Central to any future consideration of codicological intentionalism is Sobecki's key insight concerning what is recoverable about the intentions of historical producers of texts—that is, scribes and their directors. Where Sobecki's initial formulation gives pride of place to what he terms “effective authorship” (i.e., authors as makers of texts), my alternative formulation focuses on what he terms “operational authorship” (i.e., authors as makers of books) in the recovery of meaning. In both cases, historical agents are understood to play a significant role in delimiting the interpretive possibilities for manuscripts they produce. The two versions of codicological intentionalism are complementary whenever originary authors play a direct or proximate role in the production of manuscript books containing their own texts—that is, when narrative deixis and scribal craft substantially align. Approaches diverge, however, in situations where anonymity and scribal distance from original authorship play a larger role.⁶ Ultimately, with the adoption of a codicological intentionalism attuned to the production of books, scholars can recover historically situated meanings of texts in specific manuscript contexts by relying on evidence left by their immediate scribal makers, rather than relying upon internal documentary evidence about their originating textual authors.

Defining codicological intentionalism

“Codicological intentionalism,” Sobecki writes, “infers the purpose, occasion, and objective of a text, all the while using text as a proxy for intention, whether authorial or scribal.”⁷ This precise formulation is crucial for understanding the stakes of Sobecki's position: it appears nearly verbatim in both companion essays in JMEMS and Speculum, and it usefully captures the complex interrelationship between originary authorship, scribal authorship, and critical interpretation that forms the foundation for all engagements with medieval texts. A text, in this account, is the sort of thing that cannot help but bear intention, and intention is the sort of thing that must be ascribed to some agent. What matters for codicological intentionalism in this formulation is the text, and the text matters because it can provide insight into the intentions of historical agents. Through the recovery of intention, in turn, we come to understand documentary records and cultural contexts for textual production.

Sobecki insists that codicological intentionalism does not describe a new mode of engaging with premodern texts, but rather captures a set of assumptions commonly employed within current critical praxis. Sobecki exhorts “medievalists to come clean” about the assumptions that undergird their critical interpretations of manuscript texts, and in this way “to admit our investments in the project of intentionalism.”⁸ Fundamentally, codicological intentionalism describes a kind of activity that scholars of premodern literatures in Europe undertake all the time without confessing (or, perhaps, realizing) what it is that they are doing. Through long-overdue self-examination, Sobecki states, we will come to understand better the implications of the “de facto methodological position assumed by many if not most manuscript scholars.” Codicological intentionalism emerges as a set of practices at once prevalent in scholarship and disguised within scholarly discourse: “the implied position . . . of (re)constructing authorial intention through the study of manuscripts and handwriting.”⁹ Despite enduring interest in the relationship between book production and cultural criticism, the practice remains undertheorized.

In Sobecki's view, scholarly rhetoric obfuscates the importance of its own argumentative grounding. Accordingly, it is fundamentally a good thing to account for the intentions of historical authors, since historical authors were producers of historical texts. He writes elsewhere, “This assumption of proximity between author and audience . . . always opened up the shortest possible route to the new archival documents that, in turn, confirmed the reliability of the textual self.”¹⁰ Admitting to the project of codicological intentionalism means uncovering more opportunities to understand the historical authors who played active roles in the dissemination of their compositions. Theorizing codicological intentionalism means refining current critical practice for a more robust understanding of literary history.

Intentionalism and context

Sobecki draws a methodological connection between his assessment of codicological intentionalism and influential forms of textual intentionalism as a tool used in both interpretive and text-critical practice. He dispenses judiciously with Wimsatt and Beardsley's infamous “intentional fallacy”—a “reductionist stance rooted in the artificial (and debilitating) assumption that text and context are demarcated by a fixed boundary”—and clarifies his own objections to editorial techniques that have origins in “the carefree celebrations of editorial authority.”¹¹ Though editorial concern with “final authorial intention” championed by textual critics such as Bowers and Tanselle may correspond in some ways to Sobecki's concern with historical agency, the force of codicological intentionalism “move[s] in the opposite direction of” deductive projects aimed solely at reconstituting authorial texts.¹² For Sobecki, the better part of textual criticism engages in “reverse engineering an inductive urtext based on available information” from surviving manuscript witnesses. This situation describes a powerful form of criticism that “erect[s] a sophisticated theoretical scaffolding around the text that amounts to inductive intentionalism.” The reconstruction of a text implies the reconstruction of a singular historical author—what Sobecki terms “effective authorship.”¹³

Sobecki compares his position at length to Bart Ehrman's work on scribal modifications to Christian scriptures. He cites Ehrman for his method designed “to cope with the instability of the handwritten word.”¹⁴ Most important for Sobecki is Ehrman's account of “functional intentionalism,” an alternative to any attempt to recover the inaccessible internal information of “what an individual scribe may have had in mind”—a facet of intention that shades easily into “motive.”¹⁵ Rather than asking what some anonymous historical agent might have been intending to accomplish in taking some action on a text, Ehrman's functional intentionalism takes scribal alterations to biblical texts as evidence for understanding the breadth of meanings available to texts themselves.

In Ehrman's construct, the exact private cause of a change must remain unknown. The effects of that change, however, remain explicable without consideration of any private motives the scribe may have had in making the change. “There are passages where we can conjecture that a change was made intentionally,” Ehrman writes, “but we cannot know for certain what the intention was.”¹⁶ His methodology asks how scribes doing the altering may have understood the meanings of the texts that they altered, rather than what they meant more broadly by altering the text at all. The force of intention is rather mechanical than conceptual. Ehrman's formulation is comparable to Peter L. Shillingsburg's observation of difference, in theorizing the purpose of textual criticism, between “an intention to record on paper, or in some other medium, a specific sequence of words and punctuation according to an acceptable or feasible grammar or relevant linguistic convention” and “an intention to convey an idea.”¹⁷ To identify an authorially intended lexeme is not to identify an authorially intended meaning. Simply put, what is relevant to functional intentionalism is not what a scribe meant to do (his “intention in writing”), but how what he did altered textual meaning (“his intention to write”).

Functional intentionalism in this way pivots on a distinction between types of intention strongly associated with the contextualist school in intellectual history and made most influentially by Quentin Skinner. Skinner's account differentiates between an author's intention “by writing” (“an intention to do x”) and an author's intention “in writing” (“embodying a particular intention in x-ing”). For Skinner, both meanings of “intention” must be considered in comparison with motive and with intention “by writing,” which each are conditions “antecedent to and contingently connected with” the appearance of a work. Intention “in writing,” however, describes “a feature of the work itself.” Accordingly, authors’ intentions “in writing” must be, “in some sense ‘inside’ their texts, rather than ‘outside’ and contingently connected with their appearance.”¹⁸ In Skinner's account, it is impossible to know with certainty whether an author wanted an audience to feel a certain way (sad, angry, bemused, startled) “by writing” a given text, just as it is impossible to know the author's precise personal motives for writing in the first place. It is nevertheless possible—and indeed necessary for responsible historical study—to try to discover what an author might have meant to accomplish in undertaking the writing of the text in the first place (“in writing”).

There is nearly uniform consensus among published scholars that privately held motives in writing (as in any action) cannot be accessed directly by external agents. What a specific historical agent meant to accomplish by taking a specific textual action cannot be the primary pursuit of scholarly interpretation. This is true even though each critic uses competing terminology to describe the same concept—for Ehrman, “intention in writing”; for Skinner “intention by writing.” Both critics align with Wimsatt and Beardsley's most rudimentary formulation of their “intentional fallacy,” which relies upon “external evidence” for understanding meaning.¹⁹ It is this kind of intention, too, that John Farrell has recently termed “practical intention” and declared immaterial for literary criticism.²⁰ Humans do not, by and large, read minds through texts; and the purpose of scholarship is not to capture the privately held purposes or desires of authors living or dead. Sobecki, Ehrman, Skinner, Farrell, Wimsatt and Beardsley, and many others agree: insofar as “intention” overlaps conceptually with “motive,” its recovery is not relevant to critical inquiry.

It is a telling prepositional infelicity that Ehrman's “intention in writing” and Skinner's “intention in writing” refer to contrary concepts. For Ehrman, what can be recovered from employing functional intentionalism is the effect of a textual alteration (usually, in the context of textual criticism)—a lexical alteration to an otherwise putatively stable text). Changes in words effect changes in meaning, and both forms of change are the domain of responsible scholarship. Examination of lexical change reveals, rather than presupposes, relevant historical context. Discarding any hope of recovering information about why changes were made by historical agents (their “intention in writing”), Ehrman can nevertheless assess the breadth of possible meaning in an unwieldy textual tradition.

Skinner, on the other hand, is resolute in his pursuit of what is, in fact, recoverable about historical agents according to “what [they] say in a given context.” He borrows J. L. Austin's influential distinction between perlocutionary and illocutionary force, the former having to do with lexical meaning and the later having to do with the full range of communicative purpose a speaker may engage in speaking. “An understanding of the illocutionary act performed by a speaker,” Skinner reasons, “will be equivalent to understanding their primary intentions in issuing their utterance.” Illocutionary force provides Skinner an avenue for recovering context, not in service of text but alongside any understanding of text: “it [i.e., the recovering of “intention in writing”] is equivalent to being able to say that he or she [the author] must have meant the work as an attack on, or a defence of, as a criticism of, or as a contribution to, some particular attitude or line of argument, and so on.” Moreover, for Skinner, “whatever an author was doing in writing what he or she wrote must be relevant to interpretation.” Skinner is careful to acknowledge that it is possible to interpret a text without consideration of authorial intention “in writing”: “I see no impropriety in speaking of a work having a meaning which its author could not have intended.”²¹ Careful consideration of authorial intention can nevertheless yield fruitful understandings of textual significance. While a text might exceed or depart from its author's intentions, it can also be understood to correspond to them in some way, through the medium of language.

Codicological intentionalism must address the kinds of intention that can be recovered, along with the degree to which such intentions can be recovered, once the recovery of motive has been set aside. It is at this conceptual boundary that Ehrman's and Skinner's accounts begin to diverge, and the source of their divergence is useful for understanding the boundaries of codicological intentionalism in Sobecki's formulation. Although Sobecki cites Ehrman, his strong claim that codicological intentionalism “incorporates but goes beyond functional intentionalism” relies upon a theoretical distinction more apt to Skinner's scheme.²² It is true that Sobecki and Ehrman share a common interest in how scribal transmission affects the meaning of texts. But because “medievalist manuscript scholars, by contrast, have at their disposal many more data points, sources, and records” than do biblical manuscript scholars, “codicological intentionalism concentrates on operational authorship (primarily scribal intention), working back toward establishing effective authorship (authorial intention).”²³ Where Ehrman's functional intentionalism tries to understand the interpretive effects of scribal decisions through examination of textual change in manuscript copy, Sobecki's codicological intentionalism tries to understand the interpretive effects of authorial decisions through the examination of textual production in manuscript copy.

Accordingly, Sobecki is not only concerned with the interpretive possibilities suggested by lexical changes between versions of a single posited documentary text. Rather, Sobecki is concretely concerned with what texts, as examined within their manuscript contexts, can reveal about their originary authors. “The frequently purposive conflation of the biological self and narrative persona blurs the text/context boundary,” Sobecki writes in both JMEMS and Speculum, “not only on the textual but also on the contextual side, drawing life records and historical documents into the process of approximating authorial intention.”²⁴ Because medieval textual cultures expected a kind of proximity between historical person and narrative construct, the texts authors wrote, even if guised in the premise of fiction, can be shown to have bearing on the lives those authors lived. Texts written by medieval authors can reveal authorial intentions without necessarily raising the problem of authorial intentions as transparent windows onto motive—that “design or intention” so influentially lambasted by Wimsatt and Beardsley as to create an enduring disciplinary taboo. Codicological intentionalism diverges from functional intentionalism on the question of recovering authorial intention through the examination of material evidence potentially distant in production from originary circumstance.

On the same question, however, codicological intentionalism aligns with Skinner's “intention in writing” insofar as it pursues an understanding of what a “writer means by what he or she says.”²⁵ Implicit within both accounts is a concern with writers or authors on their own terms, as a source of meaning for a text and not simply as an efficient cause of a text's existence. Intention matters to codicological intentionalism not because it settles the meaning of any single text but because it lends depth to how a single text might be understood to mean for different audiences. “Texts do, after all, have authors,” Skinner writes, “and authors have intentions in writing them. Perhaps the right aspiration is to try to close the gap between claiming that a text is doing something and claiming that its author is doing it.”²⁶ So too, for Sobecki, “the conceptual stakes are so high because the closer we bring the biological and historical Hoccleve to his narrative self, the more meaning we recover and ultimately restore to the ‘I’ in this connected group of poems.”²⁷ In both cases, authorial intention cannot help but be relevant to the pursuit of recovering textual significance.

Authorial intention and the communicative text

I have argued that Sobecki's description of intention aligns substantially with the contextualist approach commonly associated with Skinner. The same alignment can be seen elsewhere within Sobecki's essay in JMEMS, as for example in his reading of Rita Felski's Limits of Critique as “a plea for more, and more sophisticated, material historicism.” Sobecki's theoretical commitment to intention as recoverable through close examination of manuscript contexts likewise informs his treatment of editorial scholarship, which must aim “to reconstitute the agentive link tying an effective author—a biological person—to a final and approved text.”²⁸ In the framework Sobecki establishes, codicological intentionalism has an opportunity to succeed where textual intentionalism might otherwise fail, by placing a significant evidentiary burden on the material conditions surrounding textual production, rather than resting all interpretation of intention on textual content and variations thereof. Sobecki's account further leverages material evidence to overcome an epistemological obstacle (and, so, a methodological obstacle) already present within the contextualist paradigm. The advantage of locating interpretive practice within material archives becomes evident when considering the obstacles faced within the broader scope of intellectual history and semiotic theory.

Consider Skinner's touchstone example for understanding the force of intention, which rests upon a toy narrative, “adapted from Wittgenstein.” I quote Skinner at length, so as to capture the full force of the example's implications:

Skinner's point is that human communication depends constantly upon the search for, and functional recovery of, intention. The analogy to scholarly interpretation is clear: just as the man in the field waves his arms, so also the historical author writes down words. Each communicative act can yield multiple competing interpretations.³⁰ Crucially, neither the waving nor the words are sufficient for understanding the full significance of the act. Effective communication relies upon a broader understanding of how the act might signify, followed by some verification of how it does actually signify for the communicator.

Skinner identifies these two central components in the italicized sentence in the passage above. I will refer to them as possibility and actuality. First, some interpretation must be possible for understanding the action in a general sense; and second, that same interpretation must actually apply in the specific case at hand. Between possibility and actuality lies intention: what the communicator intends to communicate. Men in fields might wave their arms for a number of competing reasons. So, too, authors of texts might make use of language in equally plausible but incommensurate ways. For Skinner, intention functions as a kind of catalyst for verifying that one possible interpretation is in fact the interpretation most appropriate to a given context. Arm-waving, in this particular case, refers to danger, as intended by the man engaged in the activity, as recognized by the observer in the nearby field.

Skinner's example serves to illustrate a fundamental theoretical plank in his argument, namely, how context and intention are both inextricable from interpretation. It does not, however, provide a concrete explanation for how intention functions as an interpretive mechanism designed to narrow down argumentative possibilities within a stable communicative context. Consider again the example given above. Skinner originally believes that man is waving his arms to chase away a fly. Then, subsequently, Skinner “comes to understand”—note the use of middle voice—that the waving constitutes, in fact, an attempt to warn him of impending danger. The narrative implies that the second interpretation is correct. The narrative does not, however, provide a mechanism by which Skinner thought to alter his interpretation of the arm-waving in the first place—how he came to understand intention differently than upon first observation. What alters Skinner's understanding of the act? What causes him to believe otherwise than he first believed? How exactly—by what means—did Skinner “come to recognize” that his first interpretation was incorrect and his second was correct?

The answer cannot be that Skinner perceived the bull itself. The presence of the bull can later verify the communicator's intention, in this example, but it does not play a direct role in Skinner's ascertainment of that intention. For the example to demonstrate Skinner's point, the intention itself must be contained within or expressed through the communicative act of arm-waving. It is one thing to validate the conclusion that the bull is about to charge through direct observation—to turn around, see its head lowered, its breath heavy, its hoof pawing at the dirt, and so on. (There, too, another intention is understood.) It is another thing entirely to conclude that there is bull about to charge solely on the basis of seeing a distant man waving his arms. Surely it is not only the theoretical knowledge of bulls—that they sometimes exist in fields and sometimes charge—that prompts his shift in perception, or else Skinner would never have considered, initially, the incorrect interpretation (that the man was chasing away flies). Nor can it be direct knowledge of this specific bull about to charge in this specific field, or else the waving would not have been necessary.

Some other information, besides the simple act of waving, and besides observation of the bull, must have prompted his reassessment of the situation. Otherwise, the narrative might imply that interpreters should simply believe whichever interpretation comes most recently to mind. (There is a gory Hemingway-like counter-narrative implied in that reasoning, wherein Skinner first believes that he is receiving a warning about some danger, then “comes to recognize” that his neighbor is merely chasing away a fly.) Between the fly and bull lies the problem of assessing intention “in waving.” While it is straightforward to grant that accurate interpretation may rest upon accurate recovery of intention, it is much more difficult to assess how an inaccurate hypothesis about intention may be altered and subsequently verified.

Skinner is aware of this problem. He asks: “[H]ow is this process of ‘uptake’ [i.e., coming to understand] to be achieved in practice in the case of the vastly complex linguistic acts in which literary critics and intellectual historians are characteristically interested?”³¹ As a solution, Skinner presents his theory of “the social imaginary,” which captures “the complete range of the inherited symbols and representations that constitute the subjectivity of an age.”³² According to Skinner, critics should

In other words, Skinner argues that the recovery of intention depends first upon a recognition of all interpretive possibilities available to a communicative act based upon what was available to the communicating agent at the time of communication. Then, from an array of possible interpretations, scholars should come to recognize which is actually most appropriate for the communicative context. Recovery of intention thus constitutes a recovery of meaning as well.

The difficulty I am pointing toward has to do primarily with verification, understood as what undergirds the motion between possibility and actuality. In historical inquiry, an examination of intention may provide possible interpretations by way of identifying relevant context, but context cannot, alone, verify which particular interpretation necessarily applies in any particular case. The man is in a field, not in a pool or on an airstrip: the context for his arm-waving is known. And yet the question remains as to how to decide whether his intention is to chase flies, or warn of charging bulls, or something else.

The problem I have identified differs from common objections in theoretical responses to Skinner's school. Consider the three obstacles to contextualist approaches identified by Elizabeth A. Clark, in summary of relevant scholarship: that it assumes the primacy of speech over writing; that it assumes some contexts are more significant than others; and that it assumes about intention what it should seek to demonstrate. According to Clark, most forms of historicism, including Skinner's, “fail to explain how historians select the contexts they find pertinent, a task that becomes more problematic for the distant past.”³⁴ These interpretive schemes fall afoul of the root problem that “a context is never absolutely determinable,” as Derrida influentially put it.³⁵ Moreover, as Alastair Minnis observes, bearing his own historicist commitments, any attempt to fully determine context, as in Skinner's “social imaginary,” heaps an “immense—perhaps even impossible—burden of work . . . on the shoulders of the poor exegete.”³⁶ Such critiques reasonably point out the impossibility of determining context absolutely, and so for absolutely delimiting interpretation by way of contextualizing intention.

The rendition of Skinner's argument that I have sketched above, however, does not demand absolute context or absolute interpretation. Skinner goes to great lengths to distinguish his position “from the much stronger claim often advanced to the effect that the recovery of these intentions, and the decoding of the ‘original meaning’ intended by the writer, must form the whole of the interpreter's task.”³⁷ If we grant that context need not absolutely delimit interpretation, the separate problem still remains of how to choose between competing viable interpretations presented by a single bounded context. This hermeneutic problem is not predicated upon the difference between speech and writing, nor does it directly contradict the possibility that intention governs certain forms of meaning. Rather, the root of the problem lies in the fact that any single context may yield multiple competing and equally valid possible intentions. The contextualist schema in its most general form generates possible communicative intentions, but it does not provide a mechanism for selecting from among competing available interpretations within a single context.

Sobecki's formulation of codicological intentionalism provides one materially grounded solution to the problem of selecting interpretive context and verifying interpretive hypothesis. Consider this formulation by Sobecki:

And compare this to the following formulation by Skinner:

The difference in scope of context differentiates Sobecki's practice from Skinner's. For both theorists, relevant contexts must be discerned from an array of possibilities for production. For Skinner, “sufficient accuracy” can take any number of forms, as generated from an examination of what a writer was doing “in writing.” In the case of Sobecki's codicological intentionalism, however, the nature of the context relevant to inquiry must be narrowed from the start. Textual witnesses, specifically, serve as primary context for understanding authorial intention. Moreover, those textual witnesses “that were overseen by the author or produced in their lifetime” are most useful for generating accurate textual interpretations (“original or first meaning”). Sobecki's codicological intentionalism uncovers literary-historical significance through the examination of authorially produced texts within authorially produced manuscripts. To uncover the intention behind the production of the codex and to uncover the intention behind the production of the text become functionally identical modes of inquiry.

Although codicological intentionalism necessarily narrows the range of contexts applicable to inquiry into a given author or text, the utility of Sobecki's approach is in its capacity to verify, through the examination of archival materials, the accuracy of some intentions fitting to the context over others. He observes this in theory: “By dint of their material reality, medieval texts, in particular those written in Middle English, offer codicological support for probability-based, inductive intentionalism.”⁴⁰ The results of the approach are easier to see in practice, however; and in this regard Sobecki's recent scholarly record stands as compelling evidence for the value of the approach.

For example, in demonstrating that Hoccleve's narrative account of the production of the Series bears fidelity to the actual circumstances that surrounded his writing of that text, Sobecki prioritizes the relationship between what he terms the “biological author” (i.e., Thomas Hoccleve, the historical person who could only exist in one place at one time) and the “literary self” (i.e., Thomas Hoccleve, the narrative person who appears in every copy of the Series at the same time). He then proceeds to demonstrate how Hoccleve's narration of textual production in the Series is supported by the surviving manuscript record. As Hoccleve describes the situation, his copy of “The Chaste Empress” from the Gesta Romanorum originally lacks its moralization. He borrows his friend's copy, in order to translate the moralization into his own verse. Prior to doing so, however, he must first transcribe the source text into his own copy. In this way, the Series describes a series of scribal acts—in other words, “the occasion . . . of textual witness overseen by the author.” Although this sequence of scribal acts takes place within the narrative, Sobecki demonstrates how it also describes the historical contexts surrounding the narrative's composition. In his reading, a single manuscript, London, British Library, Harley MS 219 “preserves the anatomy of the copying and production process that preceded the Series,” and indeed can be shown to have served as the source manuscript for Hoccleve's own translation practice.⁴¹ Archival materials reveal a close connection between the historical Hoccleve and the literary Thomas, who each transcribed relevant textual sections in surviving manuscript copy in order to translate them into his own poetry. Sobecki concludes:

What Hoccleve as author intended “in writing” corresponds substantially with what Hoccleve as scribe accomplished by undertaking the act of copying (“in copying”) a portion of the Gesta Romanorum and then composing the Series. As the Series provides context for understanding the manuscript record, so the manuscript record provides context for its narrative events. Together, these contexts might generate multiple possible interpretations, as is evident in the robust scholarly literature that surrounds the Series and its manuscripts. By way of codicological intentionalism, however, Hoccleve's intention to record a version of historical reality as narrative content can be verified through an examination of the manuscript record and the text together. Using the manuscript record as context, the scholar can “come to understand” more fully what Hoccleve intended. The strength of the approach lies in its capacity to verify an interpretive understanding of a communicative act.

Sobecki's use of codicological evidence to recover authorial intention cannot absolutely capture every interpretive possibility presented by the manuscript contexts or the literary text; nor does it make any claim upon Hoccleve's motives or personal intentions in arranging the narrative of the Series. He observes, for example, that “in a significant plot twist . . . the Series appears to describe the production of Harley 219 rather than its own production,” and for this reason the Series presents “an effective dramatization that only partly blurs the contours of historical reality as represented in the text.”⁴³ Neither the Series nor Harley 219 can answer, absolutely, the question of why Hoccleve might have substituted his Friend's book for his own. Further questions remain open to investigation: whether the Friend represents a specific historical person, and if so who, and what Hoccleve's relation with that person may have been, etc. In Skinner's terms, there is still no intrinsic accounting for what Hoccleve intended “by writing” in this way; and in the lingering parlance of Wimsatt and Beardsley, there is no specific “biographical knowledge” to be gained of inner motive or purpose. Future readers will surely dispute Sobecki's subsequent claim, for example, that “the splitting of the author and persona, of reality and literary convention, is post-medieval.”⁴⁴ Accordingly, none would conclude that Sobecki's critical venture in codicological intentionalism has delimited the boundaries for understanding the Series in all further interpretive contexts. It is nevertheless the case that reading the intention of the author “in writing” and harnessing the manuscript record to verify the implications of that intention can lead to an improved understanding of textual significance and authorial activity in conjunction.

Scribal intention and the manuscript work

We are now in a position to ask how important the conditions of originary authorship must be for an extended understanding of codicological intentionalism. The vast majority of medieval manuscripts were not produced by, or under the direction of, the originary authors of the texts they contain. Scribes working at various textual and chronological removes from the conditions of original composition undertook the task of transcribing myriad texts into competing manuscript situations, according to their own purposes and intentions. On the one hand, it must be acknowledged that these medieval scribes had no direct access to the communicative intentions of the authors who composed the texts they copied. Scribal alterations, whether intentional or accidental, need have no direct bearing on authorial meaning. On the other hand, scholarship has demonstrated how scribes engaged constantly with the meanings they perceived in and expected from the texts they copied.⁴⁵ If a theory of codicological intentionalism can unearth interpretive possibilities for manuscript texts with respect to involvement from their originary authors, to what degree can an extended theory of codicological intentionalism likewise unearth interpretive possibilities for manuscript texts according to the communicative intentions signaled by individual scribes? How might codicological intentionalism aid interpretation in those manuscript instances when an author has not played a direct role in the production of any authorized text?

Sobecki's broader theoretical formulations leave open the possibility of research undertaken with nonauthorial agents in mind. He distinguishes throughout “The Author's Three Bodies” between effective authorship, which names historical writers as “autonomous agents of new content,” even (and perhaps especially) in those instances where the effective author might have conferred responsibility upon a past authority, and operational authorship, which encompasses all of the various ways in which the production of texts and the production of books might be diffused across agents who did not originally compose a work, including “scribes, editors, commentators, compilers, redactors, continuators, and others.” Sobecki claims that “operational or secondary authorship, is the only genuinely medieval form of authorship,” suggesting that, in order to understand fully the function of authorship for medieval texts, codicological intentionalism must account for shifting manuscript contexts. Accordingly, “manuscript scholars are particularly sensitive to reconstructing not just the judgments of the effective author of a text but also of its many operational authors.”⁴⁶ Scribal authorship and forms of scribal participation in textual meaning would seem to play a necessarily substantial role in the use of codicological intentionalism.⁴⁷

The force of Sobecki's argument on the whole, however, suggests codicological intentionalism provides insight most effectively in situations where the originary author plays a primary scribal role:

On the one hand, clear space is provided for inquiry into scribal activity as part of codicological intentionalism's broader remit: “whether we pursue effective or operational authorship.” On the other hand, the emphasis on textual witnesses directs attention consistently toward the authorial text as recovered from its manuscript conditions, rather than toward the significance of the manuscript as a material production in its own right.

The practice of refering to copies of works as “witnesses” derives from textual criticism and is particularly important within any Lachmannian methodology. There, a “witness” describes one representation of a “work,” where “work” is used not in the broad sense of a thing that is made by an artist or artisan (“The result or product of action, labour, etc.”), but in the specific sense of an archetype most proximate to the author's composition.⁴⁹ Alfred Foulet and Mary Blakely Speer, for example, define “the direct witnesses” as “all manuscripts and fragments containing that work,” where “work” refers roughly to “the text for his edition.”⁵⁰ This sense of the term is common among editors of premodern texts. Furthermore, as Paul Maas writes, it is possible for a “manuscript copy” to fail to be a functional “witness,” insofar as a prior exemplar representing the work “can be reconstructed without its help.”⁵¹ It follows that not all manuscript copies need be witnesses to the work. Implicit within the history of “witness” is the primary importance of the authorially composed text.

Even where editorial practice aims to reconstruct the work itself rather than its first recoverable archetype, the idea of a witness (in the sense of “manuscript copy”) is subordinated to some prior understanding of what the originary author finally intended.⁵² In Shillingsburg's evocative phrasing:

The “work” captures information about all available resultant copies but is itself captured by none of them. To lay emphasis on the idea of a witness, understood as a manuscript copy of a work, is necessarily to attend to the “objectives of the work's originator.”⁵⁴ When “work” refers implicitly to some overarching or archetypal form of “text”—including “the manuscript text”—then the intention brought to bear by its witnesses must in some way always relate to the originary, or effective, author who produced the work that the archetype best represents. Thus, in the case of an authorially produced text, this emphasis corresponds to an understanding of effective authorship, so “bringing us closer to the author's intended work.”⁵⁵ To witness a work is to provide testimony as to the author's intentions in composing the resulting text.

The terminology of effective and operational authorship drives home the same point. To describe operational authorship as “secondary” demands some prior form of authorship which is “primary”—namely, the historical or biographical agent who first composed the text under consideration. In this framework for understanding codicological intentionalism, the most productive instances for literary inquiry are those when the originary author has also acted in an operational capacity involving “textual witnesses that were overseen by their author or produced in their lifetime.”⁵⁶ When the originary author acts in a scribal capacity, the witness accomplishes a proximity to the originator of a textual work otherwise unavailable in further scribal witnesses precisely because the conditions of production of the manuscript copy provide context understanding for the textual author's intentions.

As long as primacy is given to individual texts as objects of inquiry, codicological intentionalism must ultimately resolve to the study of originary authorship. Manuscript copies produced at significant removes from the biographical and historical author must be, in this framework, objectively less useful as objects of interpretation: “The further they [manuscript witnesses] are removed from their author and originating context, the more fragmented and less satisfactory their meaning becomes.”⁵⁷ To interpret a textual witness by way of its author's intentions requires some assessment of its proximity to that author in production. Insofar as work describes a textual object, then its terms of intention must be authorial.

In order to inquire into other forms of historical intention held by agents who were not originary authors of the texts they acted upon, the object of inquiry for a more robust codicological intentionalism must shift. Specifically, in order to inquire into forms of intention held by producers of manuscripts, scholarship must shift attention from the text to the book. Insofar as work describes a manuscript object, as produced according to its own conditions and contexts of intention, the primary authors of the work are the “scribes, editors, commentators, compilers, redactors, continuators, and others” who undertook the labor of its production.⁵⁸ I refer to such historical actors as “codicological agents.” This category is intended to capture a broad cross-section of relevant labor and intention in the manufacture and modification of the many integrated components—material, textual, decorative, and otherwise—found in surviving manuscript books.

As with historical authors, so too codicological agents had intentions, some of which can be recovered through an examination of what they were doing “in making” manuscript books. The intentions of codicological agents need not correspond in any meaningful way with the intentions of the historical authors who originally composed the texts included in the manuscripts they created. Rather, codicological agents possessed their own varied communicative intentions, held separately from any authorial intentions, and these agents expressed those communicative intentions through the creation of textual artifacts. As D. F. McKenzie has argued under the banner of the sociology of texts, “each reading is peculiar to its occasion, each can be at least partially recovered from the physical forms of the text, and the differences in readings constitute an informative history.”⁵⁹ Moreover, as Shillingsburg demonstrates, “The significance of the sociological confluence of events is vested in, and indexed by, the whole material text—the tangible product of the forces at work at the time—not by the lexical text alone.”⁶⁰ If it is the case that the manuscript copy of a single text witnesses some lexical work, then it is also the case that the manuscript object witnesses some codicological work. The codicological work may have bearing on the lexical work, but the two productions are not identical.

A manuscript witnesses, therefore, the conditions, sociological and otherwise, of its own production as a work. Just as a manuscript text might serve as witness to a textual work (the sort of thing made by authors), so too a manuscript might serve as witness to a codicological work (the sort of thing made by codicological agents). A manuscript is not merely a vehicle for text; rather, the composition of a manuscript informs textual meaning by preserving and representing the intentions of its makers. Here, the technical term witness may come to the limit of its usefulness when applied to the book, as opposed to the text. In the production of a manuscript where codicological agency forms the primary locus of inquiry, the only resulting copy of the work is the manuscript itself.

Most literary manuscripts, especially those containing European vernacular literatures, are the only witnesses to the work they represent. These manuscripts are informed broadly by the intentions of one or more codicological agents in their physical makeup, textual contents, hierarchies of script, and decorative programs, among other qualities that make up the “whole book.”⁶¹ Of course, there are forms of the medieval manuscript book in which layout and production were sufficiently standardized to afford study along comparative lines, such as the Paris Bible or the manuscripts represented by sigla BmBoCot in the tradition of Piers Plowman.⁶² This situation is especially common in standard texts of religious practice and ecclesiastical training necessary for university life.⁶³ Such cases are the exception, however, not the rule. Even for widely transmitted compilations and collections, such as the Glossa ordinaria or the shifting group of school-texts known as the Liber Catonianus, individual productions vary widely.⁶⁴ Most commonly, singular manuscripts witness an accretion of textual material unique to their own circumstances of production and reception.⁶⁵ Even where expectations for manuscript presentation are relatively uniform, the range of books produced can be wide.⁶⁶ Accordingly, if a singular manuscript is to be understood as a witness, then what it witnesses is some codicological work for which there is only one copy.

In order to fully comprehend the extent of codicological intentionalism as an array of interpretive methods that attend to the historical intentions recoverable from medieval codices, it is necessary to understand individual codices as meaningful works apart from their significance for individual texts. Codicological intentionalism need not only infer “the purpose, occasion, and objective of a text, all the while using text as a proxy for intention.”⁶⁷ Rather, separating the textual work from the codicological work opens further possibilities for understanding the distinct objectives of manuscript texts for the codicological agents who created them. A robustly theorized codicological intentionalism can account for multiple, overlapping, complex forms of intention and agency in the construction of medieval manuscript books.

Extending codicological intentionalism

Throughout this essay, I have aimed to demonstrate the utility of codicological intentionalism as a method for interpreting manuscript texts within the context of manuscript production, rather than textual composition. Alongside Sobecki, I have argued for the value of perceiving limited forms of intention as recoverable from the communicative labors of historical agents. Through engagements with the contextualist school and editorial theory, I have further theorized the competing ways in which conditions of manuscript production might contextualize, and so verify, competing interpretations of singular communicative acts. We need not discard an interest in textual significance in order to extend codicological intentionalism to an understanding of the individual manuscript as a work worthy of analysis. Neither need we route all interrogations of any given manuscript through some prior interest in one or several texts it contains. Codicological intentionalism need not be circumscribed by textual interpretation, nor need the methods it encompasses resolve ineluctably to textual authorship as a framework for understanding historical intention.

Shifting primary attention from authorial texts to manufactured books as objects of interpretation opens up multiplying vectors for further critical inquiry. Such inquiry may, in practice, revolve around textual contexts within individual manuscripts or groups of manuscripts. It may also revolve around significant aspects of production and use only loosely connected to literary interpretation. Recent scholarship has demonstrated how necessary some understanding of material production must be in arriving at a more robust understanding of textual production at variously local and global scales.⁶⁸ Paleographers have long understood the value of categorizing and contextualizing script in pursuit of historical understanding of text.⁶⁹ Programs of decoration and illustration reveal alignments and misalignments between the intentions of scribal copyists and their collaborators.⁷⁰ Multitext manuscripts have received extended attention and theorization along competing axes of historical and readerly intention and interpretation.⁷¹ My own first monograph argues for the employment of methods from analytical bibliography, in what I term “analytical codicology,” in order to distinguish forms of competing intention and unconscious action held by codicological agents that contributed to the manufacture of a single composite production.⁷² As with the understanding of codicological intentionalism limited to the study of authorial production, so too with the extended version I have posited above: in theorizing how medievalists might uncover forms of intentionalism, we also come to understand the myriad ways that medievalists have already uncovered possible paths forward in current critical praxis. To understand the affordances of the manuscript book as a historical work worthy of attention is also to understand the labors many scholars have undertaken in demonstrating codicological significance. Further examination will yield further methodological assumptions necessary for recovering the intentions of codicological agents.

There is, then, significantly more to say on the topic. The present essay is toward an extended codicological intentionalism, testing rather than determining the boundaries of the methods it might comprehend. To borrow Sobecki's forceful language once more: “It is time, therefore, to . . . accept that the functional codicological intentionalism espoused by manuscript scholarship is empirical, cautious, and mindful of its own relative probability and shifting frontier.”⁷³ We cannot fully occupy the interpretive contexts that conditioned the codicological agents we study, nor can any amount of careful inquiry fully recapture the intentions they brought to their craft. Extending our theoretical commitments to codicological intentionalism in order to account for the making of books—and so the making of meaning through books—nevertheless has the potential to transform broadly the parameters of premodern literary study. Evidence for how and why codicological agents pursed their crafts grows year over year. It remains to understand what meanings their recoverable intentions might communicate, through close examination of the objects they produced.

Notes