Timpany and the Fish

Stephen Jay Gould in “An Earful of Joy” recalls a moment of rapture while rehearsing Berlioz’s Tuba Mirem at Tanglewood with the Boston Symphony. The “thunder of the timpany [....] entered the wooden risers under my feet and rose from there to suffuse my body; sound became feeling.” He continues, “I do not believe in distant phyletic memory. Yet, in an odd and purely analogical sense, I had become a fish for a moment. We (and nearly all terrestrial vertebrates) hear airborne sound through our ears; fish feel the vibration of waterborne sound through their lateral line organs. Fish, in other words, `hear’ by feeling–as I had done through a set of wooden risers with a density closer to water than to air” (96). As a former member of the brass section in a concert band, I remember when my own mortal coil has shivered from such vibrations. I thank Gould, an evolutionary biologist with a gift for analogy, for connecting that shiver to a fish’s sense of “hearing.”

Gould, Stephen J. “An Earful of Joy.” Eight Little Piggies. New York: Norton, 1994. 95-108.

Misspacing as misspelling: modernization and “The man ’s mine”

In the 2-volume John P. Jewett edition, thin spaces precede the apostrophe in contractions. So one has “I[thinsp]‘ll” or “he[thinsp]‘d.” Negative contractions have the thin space before the n, so “could[thinsp]n’t” “should[thinsp]n’t etc.

But the edition has no space before an s to indicate possession. So (in chapter 1), Haley does not observe “Murray’s Grammar,” and Eliza’s son Harry takes up his “master’s stick.” (no space). In chapter 2, Mr. Harris claims to own George under the logic of slavery. He says “The man[thinsp]’s mine” (1:31). But he does not use possession. Mr. Harris uses a contraction for “man is mine.” When the thin space is present in the Jewett 2-volume edition, that form is distinguishable, by the presence of a thin space, from all possessive forms of man’s, such as when Chloe gets her “ole man’s supper” (1:38-39). When the text is modernized and that thin space is removed, the distinction between a possessive form and a contraction disappears.

In the modernized editions that I’ve looked at (Douglas, Ammons, Yellin), this distinction, which was present in the nineteenth century first book edition, has gone the way of the i/j and u/v distinction in Renaissance print. Is mis-spacing a form of a mis-spelling?

Errors in Chapter I of John P. Jewett’s Uncle Tom’s Cabin

During the process of collating four early versions of Uncle Tom’s Cabin, I have been able to identify some passages in the most commonly reprinted text of John P. Jewett’s 1852 2-volume edition that from many editorial perspectives could be considered errors. I cite four documentary sources that conceivably represent authorial preference: 1851-52 National Era newspaper (symbol NE), 1852 Jewett 2-volume edition (symbol J2V), 1852/53 Jewett one-volume paperback (JPB), and 1853 Jewett Illustrated (symbol JIL). Houghton Osgood’s New Edition (1879) is also noted. My intent is to list some of the most important examples here. This post will be devoted to chapter I.

Error 1: scrachin or screachin’
I all’ays hates these yer scrachin, screamin times. (NE 89)
I al’ays hates these yer screachin’, screamin’ times. (J2V 18)
I al’ays hates these yer screechin’, screamin’ times. (JPB 6)
I al’ays hates these yer screechin’, screamin’ times. (JIL 18)
I al’ays hates these yer screechin’, screamin’ times. (HO 6)

The first example that I consider an error in J2V, the reading closest to Stowe’s authorial manuscript is the newspaper, so the preferred reading is scrachin, screamin. I believe that scrachin is the most interesting reading because a woman whose child has been sold to a slave trader (text is from Haley’s example of previous parallels should Arthur Shelby agree to sell Eliza’s son Harry) resists with hands and nails also, not just voice.

The Jewett edition’s screachin’, screamin’ is probably a compositor’s error, caused by anticipating the subsequent word. This two words are needlessly repetitive, and the first word is misspelled. J2V twice has the correct spelling, screechin’ or screeching. See J2V 1:19 (also Haley) and 1:118 (Sam). In PB, JIL and HO, J2V screachin’ is corrected to screechin’ (JPB 6 and JIL 18). Note also that Jewett editions generally include apostrophes to indicate elided g’s in dialect. The newspaper, like the manuscript, does not.

As the authorial manuscript does not survive, I infer the greater likelihood that the earliest copy set from manuscript, the newspaper, better reflects the authorial reading. The NE reading scrachin should be noted as a possible emendation even if confined to editorial notes. Even if that proposed emendation is rejected, editors (and readers) who reject the former should consider additional occurrences of screechin’ and screeching in J2V and the correction of this form to screechin’ in JPB, JIL, and HO. The evidence is strong that screachin’ in J2V is misspelled.

Error 2: Haley folds arm

And the trader leaned back in his chair, and folded his arms, with an air of virtuous decision, apparently considering himself a second Wilberforce. (NE 89)
And the trader leaned back in his chair, and folded his arm, with an air of virtuous decision, apparently considering himself a second Wilberforce. (J2V 1:20)
And the trader leaned back in his chair, and folded his arms, with an air of virtuous decision, apparently considering himself a second Wilberforce. (PB 7)
And the trader leaned back in his chair, and folded his arms, with an air of virtuous decision, apparently considering himself a second Wilberforce. (JIL 19)
And the trader leaned back in his chair, and folded his arm, with an air of virtuous decision, apparently considering himself a second Wilberforce. (HO 7)

The J2V reading folded his arm is less satisfactory in context than the other three contemporaneous editions, which have folded his arms. The NE reading is probably closer to manuscript, J2V is likely a compositor’s error, and the J2V reading is corrected in PB and JIL. The 1879 HO New Edition restores the J2V reading, but the preponderance of evidence suggests that J2V and HO are incorrect.

Part III: In which a hyphen is not a space

This is third in a series of six, and possibly seven, posts with the provisional title “Marking Uncle Tom’s Cabin: Typography, Race, and Textual Transmission.” See Part I: In which a space is not a space if you’d like to start at the beginning. This series includes much-revised versions of presentations at the Midwest MLA Conference (Minneapolis, 2008) and the Society for Textual Scholarship (New York, 2008). The revised version is intended as a draft for an article to be submitted to a journal. Comments are appreciated.

---

But even “does-n’t [sic]” is a form of abstraction that from some perspectives does not do justice to the documentary form of the text or to meanings independent of documentary forms. The word in the Jewett edition is divided by a line break, and the abstracted form omits the line break. Even a photographic reproduction does not do complete justice to the ontology of the object, as a whole range of material and social conditions that may have shaped the documentary artifact. A textual paradigm that attends to these factors has been called the “bibliographical orientation.” During the past two decades, this approach, often associated with theorists D. F. McKenzie and Jerome McGann, is the perspective that has unseated authorial intentionalist editing from its position as the dominant paradigm for Anglo-American editing. If taken as an editorial paradigm, the bibliographical orientation makes it difficult to take up the traditional editorial task of emending because, as Peter L. Shillingsburg notes, it “does not admit to any parts of the text or of the physical medium to be considered nonsignificant and therefore emendable” (SECA 23).

Strictly speaking, any act of editing involves judgment. To select a single copy of a work for photographic facsimile is still an act of judgment. Regardless, the bibliographical orientation been associated with the practices of photographic facsimile and digital editing (and elements are embodied in McGann’s own Rossetti Archive), and digital too is a physical condition. The physicality of digital media is often inaccessible to the user’s conscious attention, and theorists of media have argued that digital acts of remediation express “the desire to get past the limits of representation and to achieve the real” (Bolter-Grusin 53). But by outlining markers for line breaks in the oral-print divide, as well as markers along the contours both of an historical typographical condition and of some present-day digital typographical conditions, we can show that the conversion from oral to written and from typeset by hand to digital transcription depend upon multiple levels of abstraction. By investigating abstraction from the perspective of computable definitions embedded within encoding systems, we begin to see the consequences of translating hyphen and a line break from from one media (print) form into another (digital).

According to McGann, texts are “autopoietic mechanisms operating as self-generating feedback systems that cannot be separated from those who manipulate and use them. Their autopoiesis functions through a pair of interrelated textual embodiments we can study as systems of linguistic and bibliographical codings” (Textual 15). If one attends to bibliographical codings, a concept named by McGann for an approach pioneered by McKenzie, every physical aspect of a text is part of its ontology. This view challenged the dominant paradigm for Anglo-American editing (associated with Sir Walter W. Greg, Fredson Bowers, G. Thomas Tanselle, and the MLA’s Committee on Scholarly Editing) because the abstract work or version cannot be invoked to authorize emendation of the text of a particular document. The dispute is usually waged on the level of definition and ontology. Whether “text” is an abstraction that exists independently of any individual document depends finally on one’s philosophy of text. McGann, for example, insists that the only condition of texts is the physical (qtd. in Shillingsburg Resisting 40).

A classic work on how meaning is differently embedded in oral and documentary forms of the same work is D. F. McKenzie’s “The Sociology of a Text: Oral Culture, Literature, and Print in Early New Zealand.” McKenzie shows that the work known as the Treaty of Waitangi ceded sovereignty of Maori territory to Britain in the English documentary version. But McKenzie’s reconstruction shows that the Maori oral version of the treaty cannot have the same meaning. The belief in “high-level literacy of the Maoris in the 1830s,” which validates the English document of the treaty, ” is a chimera, a fantasy creation of the European mind” (113). For the Maori, the oral discussion and agreement had higher authority: “the very form of public discourse and decision-making was oral and confirmed in the consensus not in the document” (117). The university classroom is also no stranger to such separation of oral performance and documentary authority. When undergraduates read poetry they dissipate the fantasy that a literate person translates from written to oral form without abstraction. When a reader pauses at the end of an enjambed line in an anthologized poem by Alexander Pope or John Milton, he or she reminds us that the skill of of poetry recitation is not a quality of the natural being but a cultural convention. Every translation from oral to written or from written to oral, from making treaties to reading poetry, includes reminders that our seeming seamless negotiation is a learned convention.

By focusing attention on the means of makers, whether the makers create space or end lines, we can see that translation of texts originally printed from type or from stereotype plates to machine-readable form differs little in kind from translations from oral to print or from print to oral–but the difference to me is in the nitty-gritty details untrodden by critical angels. I’ve already cried and dried the tears over my fallen state, but the maddening complexities of translating space and line break from stereotype impressions to transcribed digital texts offer a great deal of fodder for reflection. So, critical angels, hold tight while I drag you through my private slough of despond. And, print-shop devils, please curse at me when like Little Eva undirtied with the steamboat coal dust I emerge at the end with my hands still unmarked with printer’s ink.

A compositor in the mid-nineteenth century had in his type case–and it usually was a he at mid-century–physical objects with an extraordinary flexibility to justify lines of type. A compositor’s means to match the width of the line of type to his stick was elegant in its simplicity. Rather than reaching for a generic space, which did not exist, the typesetter reached for the space appropriate to the needs of the individual line. The basic spaces between words in a line of type had one of three lengths. But the width of a space is always relative to the size of the type font, thus to speak of the width of a typographical space is to speak in relative terms about the relationship of a space character to the standard measure for type, the em quad, a square piece of type. The typical space between words is known as a thick space. Three such spaces are the width of an em quad, so a thick space is three-to-the-em wide. A middle space, four to the em, is used for tighter spacing. And a thin space, usually five to the em, is used for punctuation and very tight spacing (Gaskell 45). The three basic widths of spaces between words–1/5 em or thin, 1/4 em or middle, and 1/3 em or thick–and end-of-line hyphenation offered the compositor flexibility and precision.

But even these three do not exhaust the spaces in the type case. The 1-em space, an em quad, is itself a space for standard functions of indenting a paragraph and of separating sentences. A hair space (1/6 em) was used to separate punctuation. Another specialized space, the en quad, 2 to the em in width, was used for spacing headlines of small caps letters. And longer spaces, 2-em or 3-em width, were used to pad out a line or for a blank line. A late-nineteenth century printer’s guide advises that a compositor could also combine two or more spaces of different width (Southward 171, 166). To return then to the hyphen that is not a hyphen in Uncle Tom’s Cabin, the width of the “space” that the hyphen seemingly represents is not generic: the space separating parts of contractions in two-volume Jewett edition is either a middle space or 1/4 em in normally spaced type, a thin space or 1/5 em in tightly spaced type.

The form of the digital line break, abstracted out of my shorthand “does-n’t [sic], depends both on an encoding system and on a computer implementation. In ASCII alone, a line break has three different forms. A line break might be signaled by a Line Feed character (LF, hexadecimal form 0×0a, decimal form 10), by a Carriage Return character (CR, 0×0D, 13), or by both characters in series, CR+LF. UNIX systems (including Mac OS X) use LF, pre-UNIX Macintosh systems (through MAC OS 9) use CR, and Windows systems use CR+LF. (http://en.wikipedia.org/wiki/Newline, 29 May 2009 10:51 am EST). These conventions divide the basic operating systems according to a conceptual understanding of the relationship between text viewed on screen and text as printed matter, whether a carriage return is implicit in a line feed, whether a line feed is implicit in a carriage return, or whether we’re just going to get along by insisting on redundancy after the origins of a line break for on screen display or for printout have faded from our digital conscience.

Such disagreements are not irresolvable–and note that I have simplified the extent of the disagreement by ignoring other systems–but the UNICODE standard resolves them (and the need to transfer files between systems) by specifying that UNICODE-conformant applications must recognize seven characters as line terminators: Line Feed, Carriage Return, Carriage Return followed by Line Feed, Next Line, Form Feed, Line Separator, and Paragraph Separator. Literary scholars can typically ignore these distinctions, but a textual editor or a student of digital humanities (who might want to alter texts with a PERL script) forgets these things at the price of days–and sometimes weeks–of frustration. In regular expressions, a general-purpose text-matching tool common to many programming languages, the dot (.), the general-purpose expression for every character, means any character EXCEPT a line break. An expression to match all characters must use classes, so the class that includes all digits, signaled \d between brackets, and all non-digits, signaled \D, are combined into a single expression ([\d\D]) that defines “every character” such that line breaks too are counted as characters (Schwartz 102, 106).

Unlike the complexity of line breaks–which separate OS partisans into lines with picks and axes–the complexity of space in historical typography has not raised the hackles of humanists. While only a handful of historians and literary scholars gad realized that a digital revolution was in progress–and while knowledge of differences in typography remained a sign of a mind diseased with printing history and bibliographical lore–a mathematical genius solved most problems of space representation for digital typography. In Donald Knuth’s TeXbook (1984), the guide for the publishing system TeX, he on one page offered a series of control sequences that satisfactorily address historical printing practices in his digital typesetting system. He divided the em quad into 18 units of math glue: thin space (1/6 quad, 3 units of math glue), medium space (2/9 quad, 4 units math glue), thick space (5/16 quad, 5 units of math glue), and negative thin space (no such physical being, -1/6 units of digital glue) (167). When genius leads, consensus follows, and the UNICODE standard now permits us to encode XML texts with the specificity that Knuth long ago allowed in digitally typeset text. See the UNICODE General Punctuation chart, where thick space equals three per em, mid space equals four per em, thin space equals six per em, and hair space equals thinner than thin. Those who would complain about minor discrepancies between between Knuth’s definition and Southward’s and UNICODE’s –UNICODE’s 5/16 quad differs by a sixteenth of an em from Southward’s 1/3 em–are more persnickety than I, but the combination of multiple spaces, including the negative thin space permitted in TeX, really ought to satisfy everyone that historical typographical space can be represented in digital type.

These fine distinctions of historical texts could be translated into digital form, but in historical translations (with the allowance that online versions preceded successful implementations of UNICODE) one often finds the digital preference for presence or absence, for on or off, for 1 or 0. A space is either present or absent. Most digital projects treat typographical space as either present or absent. So, for example, the Early American Fiction site, one of the most carefully transcribed digital texts, which includes Uncle Tom’s Cabin, does not distinguish between, thin, or medium, or thick spaces. Medium and thick spaces are usually transcribed as spaces, but thin spaces are sometimes transcribed as spaces, sometimes omitted.

Space widths may exist in the original documents, but if the encoding system for translation into digital form does not recognize the distinction present in print forms, the fine distinctions that I discussed are edited out of the transcription. And I am not aware that former generations of textual editors have treated space width as a significant feature, even if in their silent labors they may have fretted. Fredson Bowers, in “Some Principles for Scholarly Editions of Nineteenth-Century American Authors,” one of the notable statements on editorial practice, condemned the practice of modernization for 19th-century books, in the case of “spelling, punctuation, capitalization, word-division, or paragraphing” (SiB vol. 17, 223). But Bowers did not mention typographical spacing. Bowers imagined scholarly editions in the codex form, and any process of re-setting for a book edition will introduce new line breaks and thus demand new spacing if the scholarly edition would display professional quality printing.

But it cannot be said that typographical spacing is haphazard in any of the four versions of Stowe’s Uncle Tom’s Cabin that I’ve examined. Each edition has a style and a preference–a design. We might remind ourselves—before we dismiss this work as incidental craft—that a compositor to set typographical space in the hand-setting era expended labor comparable to the setting of the letterforms themselves. Had someone edited Uncle Tom’s Cabin in a scholarly edition in print form–it is a curious feature of literary scholarship that no one yet has–the cost in paper to burden readers with expansive lists of minor alterations in typographical space would have mitigated against recording variations in typographical space. As we have seen, most editors who have reprinted the 2-volume Jewett edition treat the minor differences in typographical spacing as inconsequential, if they consciously note its presence at all. The process of editing present-day reprints for scholars is in part an inheritor of modernist ideas about type. Digitization too inherits modern ideals about type, and the process of digitizing, like the process of reprinting, renders invisible the work of earlier generations.

Coming soon: Part IV: The Modern Era: When Type Became Visible

Bolter, Jay David and Richard Grusin. Remediation: understanding new media. Cambridge Mass.: MIT Press, 2000. Print.

Gaskell, Philip. A new introduction to bibliography. New York: St. Paul’s Bibliographies and Oak Knoll Press, 1995. Print.

“General Punctuation Range: 2000-206F.” Unicode Home Page, Version 5.1. 20 Jun 2009 http://www.unicode.org/charts/PDF/U2000.pdf. Web.

Knuth, Donald. The TeXbook. Reading Mass.: Addison-Wesley, 1991. Print.

McGann, Jerome. The textual condition. Princeton N.J.: Princeton University Press, 1991. Print.

Mckenzie, D. F. Bibliography and the sociology of texts. Cambridge: Cambridge Univ Press, 1999. Print.

“Newline – Wikipedia, the free encyclopedia.” 20 Jun 2009 http://en.wikipedia.org/wiki/Newline. Web.

Schwartz, Randal, Tom Phoenix, and brian d foy. Learning Perl. 4th ed. Sebastopol CA: O’Reilly & Associates, 1997. Print.

Shillingsburg, Peter. Resisting texts : authority and submission in constructions of meaning. Ann Arbor: University of Michigan Press, 1997. Print.

—. Scholarly editing in the computer age: theory and practice. 3rd ed. Ann Arbor: University of Michigan Press, 1996. Print.

Southward, John. Modern printing a handbook of the principles and practice of typography and the auxiliary arts. 3rd ed. London: Raithby Lawrence & Co., 1912. Print.

Stowe, Harriet Beecher. Uncle Tom’s cabin, or, Life among the lowly (1852). University of Virginia Library Digital Collections. Boston: John P. Jewett, 2003. 20 Jun 2009. Web.

Part II: In which a hyphen is not a hyphen

This is second in a series of six, and possibly seven, posts with the provisional title “Marking Uncle Tom’s Cabin: Typography, Race, and Textual Transmission.” See Part I: In which a space is not a space if you’d like to start at the beginning. This series includes much-revised versions of presentations at the Midwest MLA Conference (Minneapolis, 2008) and the Society for Textual Scholarship (New York, 2008). The revised version is intended as a draft for an article to be submitted to a journal. Comments are appreciated.

---

In the two-volume Jewett edition, volume 1, page 106, of Uncle Tom’s Cabin, slavecatcher Tom Loker asks Haley, the trader who purchased Uncle Tom, to provide what his partner Marks will call a retaining fee. Loker and Marks will pursue the child Harry–for Haley–and the mother Eliza for their own profit:

Note the hyphen at the end of the third line of the image. In an editorial sense, when this passage is transcribed, the hyphen is not a hyphen: it is not there. The hyphen that is not a hyphen is not there for many reasons–which this post will explore–but the real reason that the hyphen is not there, I propose, is that the hyphen represents a space that is not quite a space, in the sense that we cannot see it because of a paradigmatic blindness about typographical space. Not all scholars of literature and cultural studies are blind to typographical space in historical printing practices. A comment on Post I in this series, by William Tozier, shows that my original assumption about the blindness of other toilers in the field may have been rash. But I only recognized my own blindness to typographical space with the assistance of many works by Randall McLeod, most recently his “Gerald Hopkins and the Shapes of His Sonnets” (2004), and with the assistance of Peter Burnhill’s Type Spaces (London: Hyphen Press, 2003). But the degree to which my work departs from McLeod’s and Burnhill’s–and may be of more interest to scholars of American literature–concerns the intersection between type space, race, and stereotype during the textual transmission of Stowe’s work, a function of Modernization.

The hyphen could be “not there” in two senses. In the first sense, which is used by the Chicago Manual of Style for the preparation of manuscripts, it could be a soft hyphen, one “used merely to break a word at the end of a line.” The alternative, a hard or permanent hyphen, which “must remain no matter where the hyphenated word or term appears,” can be rejected as impossible, unless the intended word is “does-n’t [sic]” (87). A second sense in which the hyphen is “not there” is provided by the The Modern Language Association’s (MLA) Committee on Scholarly Editions (CSE). The MLA CSE addresses cases that are neither the Chicago Manual’s soft hyphens (”signs of syllabic division used to split a word in two for easier justification”) nor hard hyphens (”signs that a compound word is to be spelled with a hyphen”). Those which fit neither category are “ambiguous,” because it is “unclear whether the word is to be spelled with or without the hyphen” (CSE 36). In a scholarly edition, the editor uses judgment to decide how the word was “intended to be spelled.” After the matter is resolved, the editor must record the emendation in the apparatus. There are two basic choices: 1) “does-n’t [sic]” is an error in which “doesn’t” was intended, 2) Or “does-n’t [sic]” is an error in which “does n’t” was intended. For an authoritative judgment, a scholarly editor consults corollary evidence, which consists, first, of other instances of the same word in this edition, and which consists, second, of the same passage in other authoritative versions of the text.

In the two-volume Jewett edition, the contraction appears 12 more times, and the first use of the word (or words) is by Aunt Chloe:

That’s straightforward. There’s a space, a somewhat thin one, between the s and the n. Trust that if I were to photoquote (the term is Randall McLeod’s) “does n’t” 12 more times the other examples would also have a space, regardless of the speaker’s race, typically a thin one but a space nonetheless. Of course, in the troubling example at the end of the line on pg. 106, we have a hyphen at the end of the line. With confidence that there is usually a space between does and n’t, we will lean toward the hyphen as an error. But if the intent is to cite this passage or prepare an edition, a more conscientious attitude may be required, especially if the caution from MLA Style Manual (2008) echoes in our head: “Accuracy of quotations is extremely important. They must reproduce the original sources exactly” (122).

Stowe had some authority over at least four other versions of the text: the manuscript, the National Era serial version (1851-52), Jewett’s one-volume paperback Edition for the Million (1852/53), Jewett’s one-volume illustrated edition (1853), and Houghton-Osgood’s New Edition (1879). This passage is not present in the surviving manuscript pages. But it does appear in three near contemporary versions on which the author may have had an influence. I’ll photoquote other examples of “does-n’t [sic]” in chronological order of appearance: the serial, which appeared before the Jewett edition; those that followed shortly, the paperback and the illustrated edition; and, for good measure though it appears 27 years later, the last edition in Stowe’s lifetime on which she was intimately involved, the New Edition:

Note: Image size is not proportional. Also, the 3rd image is from the Google facsimile of the Sampson Low edition, a printing that was prepared based on the Jewett illustrated edition plates. The fourth image is also a GoogleBooks microfilm facsimile. I will be scanning the texts for updates to this post.

If we reach a conclusion based on the predominant practice of Jewett’s two-volume edition, there should have been a space. But these examples are also typical of their respective publication form. The National Era practice for contractions was to have no space. But because we have three editions from Jewett, and I’ve checked multiple examples in all three texts, we can make a further surmise. The presence or absence of space in contractions is a matter of design. For the short paperback edition, a trim 166 pages, spaces are generally present in contractions, as they are in the two-volume edition. For the fat illustrated edition with 568 pages, in many copies gilt-edged, with ample margins and copious engravings, the design of typography included omitting spaces in contractions. If Jewett as a publisher had a practice–and I think we can reasonably infer that it did–the act of designing the edition included deciding whether contractions should have a space. The 1879 Houghton-Osgood may carry lesser authority for space in typography, but it at least seems true that a space was thought present, from which we can infer, provisionally, either that the Houghton-Osgood compositor followed the 1852 two-volume copy or followed the design, in which, again, the presence or absence of space in contractions was a matter of concern.

Unfortunately, neither multiple editions nor the publisher’s practice offers any clarity on our original question, the presence of that curious hyphen in “does-n’t [sic].” The paperback edition’s generous typographical space conflicts with its strict economy in other matters: cramped margins, no illustrations, cheap paper. And though the illustrated edition was lavished with larger type, the designer chose to close up the space in contractions. So a transcriber of this text, or an editor who prepares a new edition, must make a surmise about this curious example. This is my surmise. When the compositor for the two-volume Jewett edition decided to place the hyphen at the end of the line, he probably struggled against competing influences. His copy, probably Stowe’s manuscript but possibly a marked up printing of the National Era newspaper, lacked a space before the comma. But the compositor’s instinct to follow copy contrasted with the book’s design, which insisted that contractions have a thin space between the two halves of the contraction. At the end of this line, he compromised awkwardly between the two practices: he inserted a hyphen that stands for a thin space.

That is, in the case of “does-n’t [sic]” at line end, a particular case that finds theoretical justification in the works of Jerome McGann and D. F. McKenzie, the hyphen represents a space that is not quite a space. Why one generic space is not equivalent to another will be the subject of Part III in this series of this post. This surmise can only be supported if our concept of typographical space is both historically sensitive and theoretically sound. While I aim to provide such a background, I must address our own era of computer typesetting, in which we have we have become accustomed to flexible spacing, wherein space as definition of width or space as substitute for line end is a matter of bewildering possibilities. Part III in this series of posts, “In which a hyphen is not a space,” will explore subtle variations in typographical space, from the perspective both of historical printing practices (hand-set type) and digital reproduction (ASCII, TeX, and Unicode). And we’ll turn in Part IV of this series to Modernist attitudes toward typography.

See Part III: In which a hyphen is not a space.

Works Cited

Burnhill, Peter. Type Spaces. London: Hyphen Press, 2003. Print.

Committee on Scholarly Editions. “Guidelines for Editors of Scholarly Editions.” Electronic Textual Editing. Eds. Katherine O’Brien O’Keeffe, et al. New York: Modern Language Association, 2006. Print.

McLeod, Randall. Voice, Text, Hypertext: Emerging Practices in Textual Studies Eds. Raimonda Modiano et al. Seattle: University of Washington Press, 2004, 177-297. Print.

Modern Language Association. MLA Style Manual and Guide to Scholarly Publishing. 3rd ed. New York: Modern Language Association, 2008. Print.

Stowe, Harriet Beecher. Uncle Tom’s Cabin: or, Life among the Lowly. 2 Vols. Boston: John P. Jewett, 1852. Internet Archive. Web.

Stowe, Harriet Beecher. Uncle Tom’s Cabin: or, Life among the Lowly. National Era. 5 June 1851 — 1 April 1852. Ed. Wesley Raabe. Institute for Advanced Technology in the Humanities. Web

Stowe, Harriet Beecher. Uncle Tom’s Cabin: or, Life among the Lowly. Illustrated. Boston: John P. Jewett, 1853. GoogleBooks. [published also by Sampson Low]. Web.

Stowe, Harriet Beecher. Uncle Tom’s Cabin: or, Life among the Lowly. 1 Vol. Pbk. Boston: John P. Jewett, 1852-53. Print.

Stowe, Harriet Beecher. Uncle Tom’s Cabin: or, Life among the Lowly. Boston: Houghton-Osgood, 1879. Web.

University of Chicago Press. The Chicago Manual of Style. 15th ed. Chicago: University of Chicago Press, 2003. 87. Print.

Part I: In which a space is not a space

This is first in a series of six, and possibly seven, posts with the provisional title “Marking Uncle Tom’s Cabin: Typography, Race, and Textual Transmission.” This series includes much-revised versions of presentations at the Midwest MLA Conference (Minneapolis, 2008) and the Society for Textual Scholarship (New York, 2008). The revised version is intended as a draft for an article to be submitted to a journal. Comments are appreciated.

---

Do you see a space in the contraction I ’s?

Another page from the same book: Do you see a space in the contraction I ’m?

Both of these page images are taken from the first two-volume edition of Harriet Beecher Stowe’s Uncle Tom’s Cabin (1852), by Boston publisher John P. Jewett. A compositor who set type by hand inserted a space character, but on examining the notes on the text in most reprint editions–and the reprint editions are many–it is as if those spaces were not there, or that only one of them is there–and matters–Topsy’s.

Modern editions’ textual notes seldom offer statements about typography, spelling, or punctuation in the Jewett edition of Stowe’s work–such statements may be considered unnecessary–but Kathryn Kish Sklar in Three Novels (1982), the Library of America edition that includes UTC, The Minister’s Wooing (New York: Derby and Jackson, 1859), and Oldtown Folks (Boston: Fields, Osgood, 1869), is an exception:

The standards for American English continue to fluctuate and in some ways were conspicuously different in earlier periods from what they are now. In nineteenth-century writings, for example, a word might be spelled in more than one way, even in the same work, and such variations might be carried into print. Commas were sometimes used expressively to suggest the movement of voice, and capitals were sometimes meant to give significance to a word beyond those it might have in its uncapitalized form. Since modernization would remove those effects, this volume has preserved the spelling, punctuation, capitalization, and wording of the first editions, which, of the available texts, appear most faithful to Stowe’s intentions. (”Note on the Text” 1475)

Sklar’s statement applies all three novels. Though admirable for editorial humility–respecting both authorial intention and subtleties lost through modernization–this the most comprehensive remark on typography mentions space only by inference–”spelling” and “wording” may include typographical space. Though Sklar does-n’t state explicitly that she preserves typographical spacing, she does so in her edition to a remarkable degree (I found no errors in a four-page sample of dialect). Given our two examples, Sklar is an exception: she appears to see both spaces, and sees both as possibly significant.

Many editors modernize typographical space without comment, either because they don’t realize that they are modernizing space, or because they believe that such minor matters are beneath mention. Elizabeth Ammons (Norton, 1994), despite closing up spaces consistently, states that “no editorial changes have been made” (Note ix). Likewise, Jean Fagan Yellin (Oxford, 1998), who closes spaces, states without qualification that she uses the text of two-volume Jewett edition (”Note on the Text” xxviii). Typography and spacing are not a concern of Henry Louis Gates, Jr. and Hollis Robbins in The Annotated Uncle Tom’s Cabin (Norton 2007). They do not state their choice of a base text–though they like all the other editors reprint the two-volume Jewett edition. The text of the first American book edition, modernized, just IS Uncle Tom’s Cabin. All of these editions are valuable for other reasons, Ammons’s for its ample survey of criticism and contexts, Yellin’s for the inclusion of Douglas’s “Heroic Slave” and excerpts from Stowe’s Key, Gates-Robbins’s for its vast array of responses to Stowe’s text and Gates’s incisive effort to answer James Baldwin’s “Everybody’s Protest Novel” in the introduction. Yet not one of these editors sees the spaces typographical space as significant: the spaces are not preserved in these reprints.

A more interesting case for the present concern–an editor who modernizes some spaces but not others, without comment–is Kenneth S. Lynn (Harvard 1962), who states that he “follows the text of Jewett’s first American edition” (”History” xxviii). Ann Douglas in the Penguin (1981) edition “reprints” Lynn’s Harvard text but does not address Lynn’s modernization (Note 37). The most recent edition, introduced David Bromwich (Harvard, 2009), is an unusual case, in that the introduction implies that the text reprints the Jewett edition but in fact reprints the “John Harvard Library edition” (xxviii). In defense of Bromwich, like Douglas he introduces but does not claim to “edit” the text. Some errors in Lynn’s earlier edition (1962) remain, but some errors are corrected. In other words, these editors (at least Lynn but perhaps Douglas and Bromwich in reprints of Lynn’s text to which they attach their introductions) by their editorial practice “see” the spaces in Topsy’s contractions, but not in Eva’s.

The unstated editorial decision admits of many possible explanations–that historical conventions for spacing were not noticed, that though noticed these conventions were not deemed significant, that standard speech can be modernized but not dialect, that black speech can be modernized but not white–and all three are a part of the story that this series of posts takes up. While I will return to these partially modernized editions–which can be submitted to closer analysis because derivative versions have been published online–we’ll first make a detour through another act of looking, at an invisible hyphen and the invisible differences between spaces. Whether hyphens and typographical space are visible might depend on who is looking, and when.

See Part II: In which a hyphen is not a hyphen.

Works Cited

Stowe, Harriet Beecher. Uncle Tom’s Cabin: or, Life among the Lowly. Boston: John P. Jewett, 1852. Internet Archive. Web.

Stowe, Harriet Beecher. Uncle Tom’s Cabin: or, Life among the Lowly. Ed. Ann Douglas. New York, NY: Penguin Books, 1981, 1986. Print.

Stowe, Harriet Beecher. Uncle Tom’s Cabin; or, Life among the Lowly. Ed. Kenneth S. Lynn. The John Harvard Library. Cambridge, MA: Belknap-Harvard, 1962. Print.

Stowe, Harriet Beecher. The Annotated Uncle Tom’s Cabin. Ed. Henry Louis Gates, Jr. and Hollis Robbins. New York: Norton, 2007. Print.

Stowe, Harriet Beecher. Uncle Tom’s Cabin: or, Life among the Lowly. Introd. David Bromwich. Cambridge, MA: Belknap P of Harvard UP, 2009. Print.

Stowe, Harriet Beecher. Uncle Tom’s Cabin. Ed. Jean Fagan Yellin. New York: Oxford UP, 1998. Print.

Stowe, Harriet Beecher. Three Novels: Uncle Tom’s Cabin or, Life among the Lowly, The Minister’s Wooing, Oldtown Folks. Ed. Kathryn Kish Sklar. New York: Library of America, 1982. Print.

The Kisses of Tom and Eva: Not in Stowe’s text

In the text of Uncle Tom’s Cabin that can be attributed to Stowe, Eva does not kiss Tom, though literary critics seem to imagine that she does. Sarah Robbins refers to the “angelic mother-child Eva kissing Uncle Tom.” (539). Henry Louis Gates echoes this theme in his introduction to the Annotated Uncle Tom’s Cabin, where he says that Tom and Eva “touch, kiss, hold hands, hold each other closely” (xviii).

Below is a list of every instance of the word “kiss” in John P. Jewett’s 2-volume edition (according to text from Early American Fiction) during the portion of the work in which Eva and Tom are both present:

“O, there’s Mammy!” said Eva, as she flew across the room; and, throwing herself into her arms, she kissed her repeatedly. (1.238)

Eva flew from one to another, shaking hands and kissing, in a way that Miss Ophelia afterwards declared fairly turned her stomach. (1.238)

“Well, I want to be kind to everybody, and I wouldn’t have anything hurt; but as to kissing–”
“Niggers,” said St. Clare, “that you ‘re not up to,–hey?” (1.238)

“Do hear the darlin talk!” said Mammy, as Eva thrust it into her bosom, and, kissing her, ran down stairs to her mother. (1.261)

“You sweet, little obliging soul!” said St. Clare, kissing her; “go along, that’s a good girl, and pray for me.” (1.262)

They fell on their knees; they sobbed, and prayed, and kissed the hem of her garment; and the elder ones poured forth words of endearment, mingled in prayers and blessings, after the manner of their susceptible race. (2.104)

Her father had been in, in the evening, and had said that Eva appeared more like her former self than ever she had done since her sickness; and when he kissed her for the night, he said to Miss Ophelia,—“Cousin, we may keep her with us, after all; she is certainly better;” and he had retired with a lighter heart in his bosom than he had had there for weeks. (2.111)

There are no kisses between Tom and Eva. Period. Eva kisses Mammy. Eva kisses the St. Clare servants generally. While one might suppose that she kisses both male and female, Stowe’s “shaking hands and kissing” could as easily suggest a gender divide to these gestures. In any case, Tom is not in the assembled group, as he remains associated with the arriving vehicle, not the household’s greeting party. St. Clare and Miss Ophelia’s discussion does not insist that gender matches the gesture–kiss for female servants, handshake for male–but all of Eva’s subsequent kisses are exchanged with Mammy or with St. Clare. The only other possible kiss is that Tom is among those who kiss the hem of Eva’s garment while she is on her deathbed.

Touch, hold hands, cradle, no doubt. But no kisses. Undoubtedly, however, the suggestion of sexual energy envelops the relationship between Eva and Tom in the Lake Pontchartrain Eden-like garden, given also, for example, Eva’s “I want him,” as Hortense J. Spillers points out (558, cf. UTC 1.218). But the eroticism is so coded as to dare the 19th-C. reader to infer a reading that the text resists. Hammat Billings’s illustration has been glossed as suggestive by James F. O’Gorman (84, ctd. in Morgan 27). But Jo-Ann Morgan (Uncle Tom’s Cabin as Visual Culture) responds that the “implication of physical intimacy between them would have been highly incendiary” (28).

And this fact is crucially important. Eva does not kiss Tom because the gesture would have elicited a firestorm of criticism. Billings’s illustration brings out the suggestive eroticism while the caption instructs readers not to allow the possibility that is being suggested (cf. Morgan 28). Eva’s youth, her Christian faith, Tom’s faith, and his status as a black slave close off any suggestion in the wreath of flowers or in her hand on his knee–present-day reader, keep your mind on high-minded topics.

Tom and Eva do exchange kisses, just not in Stowe’s text. For example, in on acting script for an early 20th-C. version by the Harmount company. See text at UTC & American Culture. Tom does kiss Eva. But this is not Stowe’s version: this is a modern remake. The literary criticism with which I began is also a remake: 20th and early 21st-century re-imaginings are applied to a text in a manner that I don’t think is conceivable for serious public fiction at the time Stowe wrote. By bringing a below-consciousness subtext to conscious attention with the text’s nonexistent kiss, we miss Stowe’s use of Christian morality to ward off any such suggestion.

Most present-day readers of a Jewett edition reprint would likely pass over mixed-race kissing without notice–though my student readers thought that Tom’s efforts to lure the child Eva a little creepy–but the kisses between Tom and Eva are not part of Stowe’s Uncle Tom’s Cabin. While the text is of our time, it is also of the mid-19th Century, and historically sensitive critical reading is better served if critics actually read the text with norms of 19th century conduct in mind, not create in the text imaginary episodes not supported by evidence.

Gates, Henry Louis, Jr. Introduction. Annotated Uncle Tom’s Cabin. New York: Norton, 2008, xi-xlvii.

Morgan, Jo-Ann. Uncle Tom’s Cabin as Visual Culture. Columbia: U of Missouri P, 2007.

O’Gorman, James F. Accomplished in All Departments of Art–Hammatt Billings of Boston, 1818-1874. Amherst: University of Massachusetts Press, 1998.

Robbins, Sarah. “Gendering the History of the Antislavery Narrative: Juxtaposing Uncle Tom’s Cabin and Benito Cereno, Beloved and Middle Passage.” American Quarterly 49 (1997): 539.

Stowe, Harriet Beecher. Uncle Tom’s Cabin; or, Life among the Lowly. Vol. 1 and Vol. 2. Electronic Resource. The Electronic Archive of American Fiction, 1850-1875. Charlottesville: Electronic Text Center, 2003. Online.

To do list

As my previous post makes clear, the danger to my career health is distractibility. I work all the time, but I know of only two ways to motivate myself to complete work that is in progress: shame and deadlines. I report to myself, so I shame myself here if the work is not done. I create a deadline of late August 2009. So it’s time to put these motivations to work (after I answer the phone, of course, and the email). Ah, that done–only 5 minutes later–it’s time for the list of tasks to complete:

• Course Planning: 15 to 20 hours over next two weeks, with attention to L. Dee Finks’s Creating Significant Learning Experiences, the syllabus, a schedule, and 4 or 5 planned activities as the goal.
• Revising and posting here (in parts, at least every other day) my conference presentation on race and typographical space in UTC, as preparation for revising it, probably for 12-15 pg essay in M/MLA
• Drafting on canon formation and UTC, for a planned October presentation to ADE
• Completing draft collation of four book versions of UTC that are part of my project (80 percent done).
• Transcribing chapter a day of UTC for two weeks so I have more representative sample, and then correcting
• Sight collating two copies of Jewett paperback, correcting paperback transcription, and completing survey of known copies according to WorldCat: to repeat survey of known copies for National Era newspaper (done, just need to post), Jewett 2-vol. edition (start with Kirkham and update), for illustrated edition, and for 1879 edition.
• Revising submission to American Periodicals.
• Preparing article for Notes and Queries on source for Senator Burr’s becoming Senator Bird.
• Drafting NEH Scholarly Editions Grant Proposal for November deadline
• Fixing up IBE site, personal web site, and doing twenty small tasks in preparation for teaching

Creeping distractibility, sounds nice…

Rather than reading beyond Laura Miller’s quote on “creeping distractibility” (where was that from?) in review of Winifred Gallagher’s Rapt: Attention and the Focused Life (thanks, Humanist) and then checking how much it costs on Amazon, remembering while there that I wanted to look up works by Christopher Lasch and Richard Rorty, suggested to me by George Scialabba’s Divided Mind, downloaded because my mother wrote a letter to me asking if I’d heard of George (Shalivo???) or something she heard on NPR, because he had written a book that sounds interesting–What are Intellectuals Good For?–I think I’d better get back to work.

While I would note that Scialabba’s Divided Mind is worth the download–I finished the little gem this afternoon–what I really need to do is grade two sets of papers and work on an essay. I have graded a little over a third of the essays. But I really need to get back to transcribing Uncle Tom’s Cabin, though I don’t know whether I should do that or collate the four versions that I now have, which should I discover that illustrated edition and the paperback version have even more variants than are presently known, I’ll really be doing what I want to do. And it should help the planned grant application.

Or maybe I should resume listening to the German tapes from Deutsche-Welle, so nice now that I’ve figured out how to use the MP3 player instead of the old CD route. Or maybe really work at Blackmon’s Slavery by Another Name, which I also started today. And then, as I mused on the essay, there are the two little volumes by Montaigne and Bacon, not to mention that I also picked up Fish’s Surpised by Sin and Greenblatt’s Renaissance Self-Fashioning. Department meeting prep? Grad faculty meeting prep? Dot Porter’s essay linked from SHARP?

Submit copyright requests for summer class? Submit ILL requests for Stowe essay I mean to get back to, repair the CD drive on the old laptop, uninstall xMetal from the Virtual PC on the laptop, re-install and update on the desktop PC, order the loupe for collating, finish up earlier post on paperback editions of UTC, visually collate two paperbacks on my desk, go back and revise essay on race and space from STS, begin work on the essay on the many current editions of UTC? Call the HelpDesk about my continuing troubles with email? Worry about whether the sense of unease among the student body will lead to more fires and clashes with police this weekend at Kent State?

Or maybe I should be hopeful about the grant application, and get started on that, plan for class tomorrow, set up a dentist appointment, an eye appointment, make kolaches to take to class, make the crust for the pie I’ve been promising to bake since I bought the frozen cherries on sale 3 weeks ago.

Mind unmanageable I bequeath thee to the ether, and send myself back to work. But before I go, L. Frank Baum wrote “Man does not live by bread alone but principally by catchwords,” at least according to Edward Wagenknecht in Utopia Americana (34). At least I know Wagenknecht said Baum did, but not that he did, because I just don’t know and have not the presence of mind to find out. My distractibility is not creeping, it’s at no petty pace, day to day. It’s a wildfire burning.

On Reading Milton’s Paradise Lost in the Norton Anthology

Graduate school is a time of ever-narrowing specialization. Any reader who has spent more than a moment on this blog will note my obsession with Harriet Beecher Stowe’s great work. But one of the pleasures of teaching in a department is that you must teach outside of your specialty. With that comes the opportunity to discover what you don’t know.

I don’t pretend to have read all of the major works in English literature, but one of the large holes in my background, despite having completed a Ph.D., is to have never read John Milton’s Paradise Lost in its entirety. Influenced by the syllabus on which I modeled this semester’s survey of English literature to 1800–by a scholar who took a Milton seminar in graduate school–I added it to the syllabus. We have just completed Milton’s poem in class, and I found it to be a wonderful read.

Satan is the most delicious anti-hero. I subscribe to the commonplace that the poem’s great test is to entice one to Satan’s side while continually indicating that being susceptible to his charms is a sign of one’s fallen state. Although I feel some responsibility to share with my students the ways in which Milton continually undercuts Satan’s appeal, most resisted to the end. One student expressed disappointment that Satan disappears from the final two books.

And speaking of the final two books, I’m thoroughly annoyed by the decision of the Norton Anthology of English Literature editors on the form in which it is presented. The anthology version is divided into 12 books to accord with the text which is being reprinted, the second edition (1674). The first edition (1667), which had 10 books, appears in two forms. The first printing lacks the argument. Apparently at the printer’s request, though possibly at Milton’s, prose summaries were added to the front matter of the 1667 printing (Fletcher, Milton’s Poetical Works, vol. 3, 177). When the poem was divided into 12 books for the 2nd edition, each book’s argument is now paced at the beginning of each individual book.

The arguments are a wonderful aid to readers, and it seems to me that readers who are encountering this work for the first time would derive considerable benefit if the arguments were included in the anthology. The Norton Anthology editors only include the argument for the first book. The anthology has a brief footnote to explain the complex textual situation, and it also offers this note on the presence of the argument for the first book: “We reprint the ‘Argument’ for the first book” (1831 n. 1). As a factual statement, it’s unimpeachable. But the statement has the following implication: For reasons we don’t care to go into, we do not reprint the ‘Argument’ for the remaining books.

The arguments are part of the version that the anthology claims to reprint, just as the defense of unrhymed verse only appears in the 2nd edition. Which leads me to surmise as follows: the arguments for books 2 to 12 are omitted to save space. The editors presume, I suppose, that most readers of the anthology will only read a small selection of books. I think that this decision is an error: the argument for each book should be included with the text. At least for myself in the future, I’m going to prep this better so that I can provide students with the arguments to accompany the text.

Harris Francis Fletcher, John Milton’s Complete Poetical Works: Reproduced in Photographic Facsimile vol. III. (Urbana: Univ. of Illinois Press, 1945).

Note: This post, in its former state, was poorly informed about the printing history. Thanks to some time with Fletcher, I think I’ve at least figured out the printing history of the Arguments, at least as that history was understood in 1945.