Editorial Policies
The guiding principles for this edition of Tannahill’s work are threefold: firstly, to provide an edition as free from editorial interference as possible, secondly, to provide a text that remains accessible to the reader, and thirdly, to provide a text as complete as possible. It may seem at first that these objectives are at odds. How can an editor create a reader-accessible edition of a source text 200 years old without significant emendation while combining multiple source texts to ensure comprehensiveness?
To accomplish the first objective, I transcribed the text very literally. I maintained small caps when used in titles and Tannahill’s use of italics throughout. I did not change any of Tannahill's substantives. I presented the text layout as close to the original as possible, and noted emendations to spacing and line indentation where I made changes. I silently emended the spacing before and after punctuation to single spacing instead of Tannahill's publishers original double spacing.
I did not change Tannahill's spelling except where I was able to reasonably determine a printer's error rather than Tannahill's unique usage of a word. One potential significant loss of information due to transcription occurs through my transcription of rules. In The Soldier’s Return … engraved rules of varying sorts separate poems that share pages. Rules generally do not occur at the beginning of poems that start on their own page. Rules occur in every instance of song, including the first song. However, the songs may still follow the same protocol as poems. The first song may have a rule to divide the from the section title, “SONGS.” No other song begins on its own page, so there are no examples of songs without rules. I chose to ignore these rules when drafting the clear text of the poems. Because each poem or song stands isolated from the context of other poems in this edition, including rules that divide poems is irrelevant. I did, however, include rules that divide the title and subtitle of poems from the text of the poem. I took these short, centered rules to represent one symbol. Because of the difficulty of reproducing a double rule that Tannahill's publisher employs through HTML, I chose to represent this double rule with only one line. The JPGs running parallel to the text show the original dividing rule for interested readers wishing to reconstruct the original text. Both these emendations occur silently throughout the clear text.
Another possible loss of information occurs specifically within "The Parnassiad." This poem is the only included piece in which Tannahill employs footnotes. I chose to combine footnotes that are split across pages for the user's benefit, which changes the position of those footnotes relative to the original text. I ignored the footnotes when counting lines in the poem because they do not represent the main poem text. I also chose to not include page numbers in the clear text even if they appear in the parallel PDF because these are not essential features of the copy text, but rather typesetter's artifacts. Using these numbers out of the context of the 1807 edition is confusing and gives the reader no additional useful information.
I chose to present this transcription as clear text alongside scans of Tannahill’s work. Kline and Perdue note in A Guide to Documentary Editing that “clear text has traditionally described the preferred method for presenting the critically edited texts of published works. The texts themselves contain neither critical symbols nor footnote numbers to indicate that an emendation has been made or that some detail has been omitted” (173). As Kline and Perdue suggest, my list of emendations appears on a page separate from the text. These emendations do not need to be presented alongside the edition itself because the clear text is the finished product and is most readable without in-line notes of emendation.
In regards to emendation, I made very few alterations to the text, generally correcting only obvious grammatical errors and textual corruptions. I transcribed the text literally and generally did not emend Tannahill’s spelling. Maintaining authentic Scots-infused spelling is important to help the reader grasp the true feeling of Tannahill’s work. Attempting to determine a standard spelling is a nearly impossible battle because of the dialectical nature of Scots, both at the time and now. Greg’s position on spelling applies here. Bowers defends Greg’s position on not normalizing any spelling because, for the Renaissance texts Greg edited, the language was “idiosyncratic and in process of flux” (Bowers 124). Scots was a language very much in flux; there was no standardized spelling, and even today few definitive Scots dictionaries exist because of the many competing dialects.
In cases where misspellings occur not because of a difference between Scots and English but a true publishing error, I emended the spelling and noted the correction in the emendations list. For example, in “The Lass O’ Arranteenie” I made the emendation hazel hazle because I could find no recorded instances of the use of “hazle” in Scots. Greg agrees that the correction of such errors is necessary: “Every editor aiming at a critical edition will, of course, correct scribal or typographical errors” (30).
Kline and Perdue also note the opinion of Elizabeth Witherall, editor of the Thoreau Journals, in regards to clear text. Witherall proposes that presenting an edition in clear text is best suited to “ ‘documents that are themselves close to final versions of the documents that they represent . . . [Such as] fair copies of literary manuscripts” (Kline and Perdue 174). Presenting Tannahill’s work in clear text makes sense because few “important inscriptional details” (174) are lost by such a transcription. My copy-text has no markings besides the rules, noted above. There are no “cancelled passages” (174) to be omitted in an edition such as this.
When the medium of an edition is digital, clear text is an obvious choice. Clear text is far easier to implement than any other textual treatment style. For example, typographic facsimiles with lines and other symbols imposed on top of text are nearly impossible to recreate through solely digital means. Diplomatic and expanded transcriptions, on the other hand, can be reproduced digitally but often require a variety of non-standard characters that cause issues when displayed in certain web browsers. Clear text presents none of these problems, and provides an additional benefit as well: searchability. Computers are far faster than humans at retrieving information. Creating a PDF of a typographic facsimile removes the search function, one of the greatest inherent benefits of the web. In terms of SEO (Search Engine Optimization) it also makes the page less likely to appear in search results because the text of the poem will not be catalogued by website crawlers.
To fulfill my second goal, creating a reader-accessible text, I chose to gloss nearly all Scots words (except for non-standard contractions) and locations. In almost all cases, I used The Dictionary of the Scots Language (DSL), a searchable digital reference based on the 12-volume Dictionary of the Older Scottish Tongue (DOST) and the 10-volume Scottish National Dictionary (SND). I also consulted Semple's edition for information regarding locations. I did not place quotes around direct quotations from Semple or the DSL because I found them distracting to the reader; I did not place in-text citations for DSL references for the same reason. My basic frame of reference for determining a gloss was my own experience. If I was unsure of the meaning of a non-standard English word, I glossed the word for the reader’s benefit. In this regard, I remained cautious because determining the knowledge of a typical reader is a subjective matter. In this case, Kline and Perdue suggest that editors “err on the side of over-citation, providing all the clues necessary for readers who want to conduct further research” (250). I added these definition glosses in the most inconspicuous and non-intrusive way possible: as a mouseover event, hidden in a HTML span tag around the word or phrase in question.
Kline and Perdue generally follow the line of thought established by editor G. Thomas Tanselle in “Some Principles for Editorial Apparatus.” Here, Tanselle argues that “the kind of apparatus presented is an indication less of the nature of the text than of the type of audience for which the edition is intended. Just what apparatus is appropriate for a particular audience is a matter about which opinion naturally changes over the course of time” (“Some Principles” 42). For this audience, presenting my notes and glosses near the text is important to a clear understanding of the text. However, because of the facsimiles already present on the page and the anticipated nature of my audience, noting emendations on the page itself is not important.
The MLA Committee on Scholarly Editions’ Guidelines for Scholarly Editions is a series of procedures editors generally follow to obtain CSE approval of their edition. One parameter presents the following question: “Is it important, and is it feasible, to reproduce the material sources in facsimile as part of the edition?” (“Guidelines” 1.2.1.8). In the case of Robert Tannahill’s work, reproducing facsimiles of the texts in question provides a substantial benefit by showing readers the original, unaltered text. Comparing this photofacsimile to my clear text edition allows the reader, with the help of my emendations list, allows the reader to deconstruct the editing process and comprehend the editorial decisions I made to the copy-text.
The photofacsimiles used in “An’ Nature All Is Cheery: The Integration of Nature Imagery in Tannahill’s Poems and Songs” are .jpg picture files converted from .pdf files of Shelfmark Ry.III.f.19 in the National Library of Scotland Special Collections. I chose to use .jpg files instead of the original PDFs because .jpgs can be viewed in any web browser, including those on mobile devices. The MLA Committee for Scholarly Editions also approves of the use of .jpg images because they conform to their ISO standard (“Guidelines” 2.25.2). Furthermore, PDFs generally require a plugin such as Adobe Reader to function correctly. These plugins consume system RAM and take a longer time to load than .jpgs. I did not enhance the .pdf or .jpg files in any way; they are unmodified professional scans by National Library of Scotland staff. In terms of accuracy, both the .jpg files and the PDFs provide very clear, viewable recreations of Tannahill’s work. Clicking on any document image will bring the image fullscreen for closer scrutiny, providing readers with poor eyesight or perusing the site on mobile devices the same experience as an average user on the desktop site.
A reader may question why creating a clear text edition of Tannahill’s work is necessary at all if the original text is highly readable. Presenting both a clear text and the original text provides the reader with options. In “The Varieties of Scholarly Editing,” G. Thomas Tanselle refers to these two options as “literal (“diplomatic”) transcriptions … [and] critical editions” (“Varieties” 17). The first, photographic facsimiles, literally reproduces the document, and is thus very reliable, but not as readable. The second, a clear text, is so named because “their texts are the products of the critical judgment of editors” (“Varieties” 17), and is thus less reliable but generally more readable.The clear text is still central to the project because some emendations to the source text must be made, such as correcting misspellings, in order to “focus on … [the] work as intended at some past moment” (“Varieties” 17). This would be impossible to do in the medium of photography. The clear text is also web searchable, unlike .jpg images. Furthermore, the images do not provide line numbers, a helpful resource when dealing with poems and songs.
When adding line numbers, I chose to begin numbering with the first line of the poem’s body text and exclude the title, subtitle, and any other prior information from the numbering sequence. Stanza breaks are also not included in the numbering sequence. Choruses are included in the line count, but only if they are full instances. Abbreviated chorus lines that employ the &c. symbol are not included. These decisions can be justified because they reflect poetic and musical conventions, which do not count blank lines and stanza breaks as lines. Because of the relatively clean nature of the copy-text, no information outside the text is lost by not having line numbers that reflect the entire space consumed by the poem. Ignoring repeated chorus lines also helps to hold the integrity of the line count and preserves stanza counts (e.g. if stanzas or verses are four lines, this groups them as four lines with a chorus repeat line instead of five lines).
For information on my design process, please see the Design Principles subsection of the Project Rationale.
To accomplish the first objective, I transcribed the text very literally. I maintained small caps when used in titles and Tannahill’s use of italics throughout. I did not change any of Tannahill's substantives. I presented the text layout as close to the original as possible, and noted emendations to spacing and line indentation where I made changes. I silently emended the spacing before and after punctuation to single spacing instead of Tannahill's publishers original double spacing.
I did not change Tannahill's spelling except where I was able to reasonably determine a printer's error rather than Tannahill's unique usage of a word. One potential significant loss of information due to transcription occurs through my transcription of rules. In The Soldier’s Return … engraved rules of varying sorts separate poems that share pages. Rules generally do not occur at the beginning of poems that start on their own page. Rules occur in every instance of song, including the first song. However, the songs may still follow the same protocol as poems. The first song may have a rule to divide the from the section title, “SONGS.” No other song begins on its own page, so there are no examples of songs without rules. I chose to ignore these rules when drafting the clear text of the poems. Because each poem or song stands isolated from the context of other poems in this edition, including rules that divide poems is irrelevant. I did, however, include rules that divide the title and subtitle of poems from the text of the poem. I took these short, centered rules to represent one symbol. Because of the difficulty of reproducing a double rule that Tannahill's publisher employs through HTML, I chose to represent this double rule with only one line. The JPGs running parallel to the text show the original dividing rule for interested readers wishing to reconstruct the original text. Both these emendations occur silently throughout the clear text.
Another possible loss of information occurs specifically within "The Parnassiad." This poem is the only included piece in which Tannahill employs footnotes. I chose to combine footnotes that are split across pages for the user's benefit, which changes the position of those footnotes relative to the original text. I ignored the footnotes when counting lines in the poem because they do not represent the main poem text. I also chose to not include page numbers in the clear text even if they appear in the parallel PDF because these are not essential features of the copy text, but rather typesetter's artifacts. Using these numbers out of the context of the 1807 edition is confusing and gives the reader no additional useful information.
I chose to present this transcription as clear text alongside scans of Tannahill’s work. Kline and Perdue note in A Guide to Documentary Editing that “clear text has traditionally described the preferred method for presenting the critically edited texts of published works. The texts themselves contain neither critical symbols nor footnote numbers to indicate that an emendation has been made or that some detail has been omitted” (173). As Kline and Perdue suggest, my list of emendations appears on a page separate from the text. These emendations do not need to be presented alongside the edition itself because the clear text is the finished product and is most readable without in-line notes of emendation.
In regards to emendation, I made very few alterations to the text, generally correcting only obvious grammatical errors and textual corruptions. I transcribed the text literally and generally did not emend Tannahill’s spelling. Maintaining authentic Scots-infused spelling is important to help the reader grasp the true feeling of Tannahill’s work. Attempting to determine a standard spelling is a nearly impossible battle because of the dialectical nature of Scots, both at the time and now. Greg’s position on spelling applies here. Bowers defends Greg’s position on not normalizing any spelling because, for the Renaissance texts Greg edited, the language was “idiosyncratic and in process of flux” (Bowers 124). Scots was a language very much in flux; there was no standardized spelling, and even today few definitive Scots dictionaries exist because of the many competing dialects.
In cases where misspellings occur not because of a difference between Scots and English but a true publishing error, I emended the spelling and noted the correction in the emendations list. For example, in “The Lass O’ Arranteenie” I made the emendation hazel hazle because I could find no recorded instances of the use of “hazle” in Scots. Greg agrees that the correction of such errors is necessary: “Every editor aiming at a critical edition will, of course, correct scribal or typographical errors” (30).
Kline and Perdue also note the opinion of Elizabeth Witherall, editor of the Thoreau Journals, in regards to clear text. Witherall proposes that presenting an edition in clear text is best suited to “ ‘documents that are themselves close to final versions of the documents that they represent . . . [Such as] fair copies of literary manuscripts” (Kline and Perdue 174). Presenting Tannahill’s work in clear text makes sense because few “important inscriptional details” (174) are lost by such a transcription. My copy-text has no markings besides the rules, noted above. There are no “cancelled passages” (174) to be omitted in an edition such as this.
When the medium of an edition is digital, clear text is an obvious choice. Clear text is far easier to implement than any other textual treatment style. For example, typographic facsimiles with lines and other symbols imposed on top of text are nearly impossible to recreate through solely digital means. Diplomatic and expanded transcriptions, on the other hand, can be reproduced digitally but often require a variety of non-standard characters that cause issues when displayed in certain web browsers. Clear text presents none of these problems, and provides an additional benefit as well: searchability. Computers are far faster than humans at retrieving information. Creating a PDF of a typographic facsimile removes the search function, one of the greatest inherent benefits of the web. In terms of SEO (Search Engine Optimization) it also makes the page less likely to appear in search results because the text of the poem will not be catalogued by website crawlers.
To fulfill my second goal, creating a reader-accessible text, I chose to gloss nearly all Scots words (except for non-standard contractions) and locations. In almost all cases, I used The Dictionary of the Scots Language (DSL), a searchable digital reference based on the 12-volume Dictionary of the Older Scottish Tongue (DOST) and the 10-volume Scottish National Dictionary (SND). I also consulted Semple's edition for information regarding locations. I did not place quotes around direct quotations from Semple or the DSL because I found them distracting to the reader; I did not place in-text citations for DSL references for the same reason. My basic frame of reference for determining a gloss was my own experience. If I was unsure of the meaning of a non-standard English word, I glossed the word for the reader’s benefit. In this regard, I remained cautious because determining the knowledge of a typical reader is a subjective matter. In this case, Kline and Perdue suggest that editors “err on the side of over-citation, providing all the clues necessary for readers who want to conduct further research” (250). I added these definition glosses in the most inconspicuous and non-intrusive way possible: as a mouseover event, hidden in a HTML span tag around the word or phrase in question.
Kline and Perdue generally follow the line of thought established by editor G. Thomas Tanselle in “Some Principles for Editorial Apparatus.” Here, Tanselle argues that “the kind of apparatus presented is an indication less of the nature of the text than of the type of audience for which the edition is intended. Just what apparatus is appropriate for a particular audience is a matter about which opinion naturally changes over the course of time” (“Some Principles” 42). For this audience, presenting my notes and glosses near the text is important to a clear understanding of the text. However, because of the facsimiles already present on the page and the anticipated nature of my audience, noting emendations on the page itself is not important.
The MLA Committee on Scholarly Editions’ Guidelines for Scholarly Editions is a series of procedures editors generally follow to obtain CSE approval of their edition. One parameter presents the following question: “Is it important, and is it feasible, to reproduce the material sources in facsimile as part of the edition?” (“Guidelines” 1.2.1.8). In the case of Robert Tannahill’s work, reproducing facsimiles of the texts in question provides a substantial benefit by showing readers the original, unaltered text. Comparing this photofacsimile to my clear text edition allows the reader, with the help of my emendations list, allows the reader to deconstruct the editing process and comprehend the editorial decisions I made to the copy-text.
The photofacsimiles used in “An’ Nature All Is Cheery: The Integration of Nature Imagery in Tannahill’s Poems and Songs” are .jpg picture files converted from .pdf files of Shelfmark Ry.III.f.19 in the National Library of Scotland Special Collections. I chose to use .jpg files instead of the original PDFs because .jpgs can be viewed in any web browser, including those on mobile devices. The MLA Committee for Scholarly Editions also approves of the use of .jpg images because they conform to their ISO standard (“Guidelines” 2.25.2). Furthermore, PDFs generally require a plugin such as Adobe Reader to function correctly. These plugins consume system RAM and take a longer time to load than .jpgs. I did not enhance the .pdf or .jpg files in any way; they are unmodified professional scans by National Library of Scotland staff. In terms of accuracy, both the .jpg files and the PDFs provide very clear, viewable recreations of Tannahill’s work. Clicking on any document image will bring the image fullscreen for closer scrutiny, providing readers with poor eyesight or perusing the site on mobile devices the same experience as an average user on the desktop site.
A reader may question why creating a clear text edition of Tannahill’s work is necessary at all if the original text is highly readable. Presenting both a clear text and the original text provides the reader with options. In “The Varieties of Scholarly Editing,” G. Thomas Tanselle refers to these two options as “literal (“diplomatic”) transcriptions … [and] critical editions” (“Varieties” 17). The first, photographic facsimiles, literally reproduces the document, and is thus very reliable, but not as readable. The second, a clear text, is so named because “their texts are the products of the critical judgment of editors” (“Varieties” 17), and is thus less reliable but generally more readable.The clear text is still central to the project because some emendations to the source text must be made, such as correcting misspellings, in order to “focus on … [the] work as intended at some past moment” (“Varieties” 17). This would be impossible to do in the medium of photography. The clear text is also web searchable, unlike .jpg images. Furthermore, the images do not provide line numbers, a helpful resource when dealing with poems and songs.
When adding line numbers, I chose to begin numbering with the first line of the poem’s body text and exclude the title, subtitle, and any other prior information from the numbering sequence. Stanza breaks are also not included in the numbering sequence. Choruses are included in the line count, but only if they are full instances. Abbreviated chorus lines that employ the &c. symbol are not included. These decisions can be justified because they reflect poetic and musical conventions, which do not count blank lines and stanza breaks as lines. Because of the relatively clean nature of the copy-text, no information outside the text is lost by not having line numbers that reflect the entire space consumed by the poem. Ignoring repeated chorus lines also helps to hold the integrity of the line count and preserves stanza counts (e.g. if stanzas or verses are four lines, this groups them as four lines with a chorus repeat line instead of five lines).
For information on my design process, please see the Design Principles subsection of the Project Rationale.