Tools for encoding manuscripts

Kinds of titles and where to put them

Each poetry manuscript will have three different kinds of titles; these titles may be identical, they may be different. Please see the rules below for details.

Title in the <titlestmt>
This title, which occurs inside the TEI header, names the electronic file you are creating and should therefore be distinguished from the title of the source material. Do this by adding the phrase "a machine readable transcription" as a subtitle, as in the following example.
<titlestmt>
 <title level="m" type="main">Death dogs my steps</title>
 <title level="m" type="sub">a machine readable transcription</title>
</titlestmt>
If you are transcribing and encoding a manuscript that doesn't have a title written on it, as the main title use the title derived from the first line, as described below. Also, include the attribute rend with the value "bracketed." This will allow titles we assign based on the first line to be bracketed when displayed. An example:
<titlestmt>
 <title level="m" type="main" rend="bracketed">And to me each minute of the night and day is vital and visible</title>
 <title level="m" type="sub">a machine readable transcription</title>
</titlestmt>
For manuscripts that contain more than one poem, follow the above procedure for each poem, but for the value of level use "a" (to indicate individual items within an item). Then wrap all of these individual titles in another <title level="m">. The following example imagines a manuscript with two poems, the first of which Whitman has given a title and the second of which he hasn't.
<titlestmt>
 <title level="m">
 <title level="a" type="main">Title Written on Manuscript</title>
 and
 <title level="a" type="main" rend="bracketed">Title derived from first line</title>
 <title level="m" type="sub">a machine readable transcription</title>
 </title>
</titlestmt>

Title in the <sourcedesc>
This title should be the title which is given to the artifact by the holding institution. This is basically a bibliography, so the information should be sufficient for a user to track down the location of the item, were they to visit the holding institution. In some cases, therefore, the "title" in this context may bear little relation to the poem contained in the manuscript -- for example, this "title" might be the title of the folder which holds the item rather than the title of the item itself.

Title in the <head>
This is the location which the stylesheet will look to pull titles of poems, etc. for indexing and display. For more detailed rules on how to formulate the titles themselves, please see the section below. The rules for the use of <head>:
- Each <lg> and <div> can have its own <head> (and thus its own title).
- Use of the "type" attribute on this element is required and the available types are listed below. The first portion of each value (before the hyphen) is meant to indicate if the title is the main title or a subtitle.
 main-authorial (title given by Whitman on the page)
 main-derived (title assigned by us, derived from the formula for titles (see next section below)
 sub-authorial (subtitle given by Whitman on the page)

Naming poetry manuscripts

We have developed a simple set of rules for giving names to Whitman's manuscript poems. Note that this naming is IN ADDITION TO the adoption of a unique identifier. The rules are listed here in the order of priority:

First priority: a given name written by Whitman. (<head type="main-authorial">)

Disregard ALL numbers, roman or arabic AND punctuation which precedes words (like "?")
Invert any titles which begin with articles such as "a," "an," and "the." For example, "The Centenarian's Story" is "Centenarian's Story, The"

If mss is not titled by WW, create a <head type="main-derived"> and use the first words not struckthrough. If you're using the first line, go up to (but do not include) the first punctuation mark OR the end of the line OR the end of the segment, WHICHEVER COMES FIRST.

For poems with common titles (like "Leaf") use the title followed by the first line (following the above rule). So: Leaf [A promise to Indiana]

Don't worry if two poems have the same title. Our stylesheet will pull the identity of the holding institution from the appropriate place in the header. To see the rules for attaching id's to holding institutions in the source description, click here.

Distinguishing poetry and prose

Whitman almost always indicates verse by his use of space on the page: new lines are left-justified and segmentation is indicated by indentation away from the left margin. These should be tagged accordingly with <l> and <seg>. Our general rule: unless a manuscript declares itself verse by using the conventions described above, tag it as prose by simply enclosing it in . Once you have chosen to use a , however, you cannot mark line breaks. To encode a manuscript which seems to include both prose and verse, see the section which follows.

Mixed-genre mss

Often Whitman's manuscripts are not simply drafts of poems, but collections of jottings in prose and verse, lists, even pasted clippings from newspapers. First, as fully as possible, map out the structure of the manuscript and decide if and how each part of the manuscript is related to the rest, (for a fuller description of how to mark structure, click here.). Enclose the entire mss in a <div1 type="poem notes">. Then the following elements are available:

for all prose

<l> need not be in a <lg> if you don't think it qualifies as a poem. Mark line segmentation <seg> only when it's clearly employed by Whitman.

<lg> for groups of lines

<list> must contain one or more <item>;can contain a <head>

<dateline> occasionally Whitman will give the date and/or occasion

<byline>&bnsp;occasionally Whitman will "sign" his poems after the title

<signed> Whitman's signature

 for all writing which is not explicitly marked as verse on the page (with Whitman's usual conventions)

<note> for any notes on the text, by Whitman or others. For a complete description of this issue, see notes and marginalia.

For example, see "Ashes of Roses," Myerson, v.1, pt.1, 153. To view a marked-up transcription, click here.

Simple additions and deletions

For simple additions and deletions, the <add> and <del> elements may be used, as described in detail in the prose section. But for anything more complicated, you must use the <app> element, as described below.

Allowable types and places on <add> and <del>

Our DTD limits the values allowable in the type and place attributes. Available values are:

For <add>:

type
insertion: an insertion above the line, marked by a caret (which often looks like an "x").
unmarked: added either surpralinearly (above line) or infralinearly (below line) with no caret.
overwrite: change is written over base text
pasteon: new text is literally glued over old with a new piece of paper
place
inline: in space left by scribe
infralinear: below line
supralinear: above line
marginbot: in bottom margin
margintop: in top margin
marginleft: in left margin
marginright: in right margin
over: over the existing letter, word, or phrase

For <del>:

type
erasure: deletion indicated by erasure of text
bracketed: deletion indicated by brackets in the text or margin
hashmark: when a large chunk of text (often the whole manuscript page) is struck through with vertical or diagonal line(s)
overstrike: deleted portion is struck through or crossed-out
overwrite: when a substitution is made by an overwrite, type on both <add> and <del> use this same type. For an example of how this works, click here.
pasteover: the old text which is deleted when an <add type="pasteon"> occurs.
Please note there is no place attribute for <del> because the idea of deletion implies the erasure of something from its current location.

Recognizing and determining the extent of deletions

Whitman's use of overstrikes for deletion are usually emphatic and easily recognizable, but occasionally one sees a mark which may be either an overstrike OR a stray pen mark. In these cases, be sure to check with Ken and/or Ed and then use the cert attribute indicate how sure you are about the mark's status as a <del>.

For example, in Myerson, v. 2, pt. 1, p.23 (a few lines from the bottom), the word "editor" appears to be struckthrough:
<del type="overstrike" cert="80%">editor</del>

Regarding extent: use your common sense when marking which words and characters have been deleted. If an entire line has been deleted, for example, but the punctuation is not touched by the line which strikes through, assume the entire line, punctuation included, has been signaled for deletion. When in doubt, please consult Ed or Ken for their reading of the passage.

Nesting <add> and <del>

The rules for combinations of the <add> and <del> elements, and for the interpretation of such combinations, are similar:

When one <add> contains another <add>, it indicates that, first, an addition was made to the base text and that something was later added to the first addition:
This is the text
 <add type="insertion" place="supralinear">with some added
 <add type="unmarked" place="supralinear">(parenthetical!) </add>
 material</add>
as written.
When one <del> nests within another <del>, it indicates that the author wrote a passage, deleted part of it (the text contained within the nested <del>), and then later deleted the entire passage (everything contained within the outer <del>):
<del type="overstrike">This sentence contains some
<del type="overstrike">redundant </del>
unnecessary verbiage.</del>
When an addition nests within a deletion, the normal interpretation will be that an addition was made within a passage later deleted in its entirety.
When a deletion nests within an addition, it indicates that a deletion was made within a passage added earlier.

***Please note that when a word or phrase is deleted and replaced with an addition, this qualifies as a substitution; rules for tagging substitutions follow.

Deletions with an addition:

In many cases Whitman deletes certain words or phrases and adds others in their place. Technically speaking, this is a substitution which is covered in section 18.1.5 in the TEI Guidelines. The most unambiguous and powerful way to represent a substitution (a deletion <del> followed by an addition <add>) is to use the <app> structure. In recording readings within an apparatus entry, the <rdg> element must always be used; each <app> will contain at least two <rdg>s. On simple changes, it may seem overly burdensome, but it allows for much more flexibility when the changes become more complicated. So, for example:

Calamus=Leaves.
~~Live Oak, with Moss.~~

is encoded as:

<app>
 <rdg varseq="1">
 <del type="overstrike">Live Oak, with Moss.</del>
 </rdg>
 <rdg varseq="2">
 <add place="supralinear" type="unmarked">Calamus&hyphen;Leaves.</add>
 </rdg>
</app>

The varseq attribute is used to give the best representation of what has occurred to the text. If the number of changes makes it difficult to determine what was done in what order, a resp attribute is available, which can identify the editor responsible for asserting a particular reading or an order in which things occurred.

A NOTE ON NESTED SUBSTITUTIONS: even though this section requires that <app> be used to deal with all substitutions, there is one scenario which is an exception to this rule. If there are several substitutions which are nested (for example, a word or phrase is deleted and replaced supralinearly with a change; and within that change the same sort of substitution is done), represent only the higher-level (or first to occur) substitution with the <app> element. The reason for this is the difficulty that nested <app> structures would pose for later processing. If you encounter this situation, the second, lower-level substitution should be encoded as faithfully as possible with a sequence of <add> and <del>.

Deletion with an over-written addition

For an example of what this looks like on the page, click here. The letter "e" has been changed from upper- to lower-case by writing the change over the original. In this case the type on both the <add> and <del> is the same and the <add> also has a place attribute. The word "each" from the manuscript page referred to above is encoded as:

<l><seg> . . .
 <app>
 <rdg varseq="1">
 <del type="overwrite">e</del>
 </rdg>
 <rdg varseq="2">
 <add type="overwrite" place="over">E</add>
 </rdg>
 </app>
ach . . .</seg></l>

Use of <addSpan> and <delSpan>

<addSpan> and <delSpan> are used when larger chunks of text are added or deleted, and more specifically, when an addition or deletion cuts across structural boundaries, like lines. TEI does not allow the use of <add> and <del> across the boundaries of <l>. The type and place attributes should be used as on <add> and <del>, and all of the values described above are available. One other value is available for the place attribute on <addSpan>: "intralinear". Also please note that <addSpan> and <delSpan> are empty elements which require an anchor element to indicate when the addition or deletion ends. Name these additions and deletions simply with an "a" (for an addition) or a "d" (for a deletion) and number them consecutively (each new document will begin again at 1).

<addSpan>
For example, on a particular manuscript page (Myerson, v.2, pt.1, p.23), two lines were added after the first draft, which is indicated by the way in which the lines are squeezed in between the otherwise consistently spaced poem. The parts of the poem immediately preceding and following have also been represented. Note the ellipses are in the original and do not indicate a deletion:

</add></l>
<addspan type="unmarked" to="a3" place="inline">
<l>I see the wife.... and she is not one jot less than the husband,</l>
<l>I see the mother .... and she is every bit as much as the father </l>
<anchor id="a3">
<l><seg>¶ I <del type="overstrike"> see you </del> etc.
<delSpan>
Whitman often strikes through several lines at once, and sometimes strikes though an entire page. Again, these are instances where <delSpan> should be used over <del> (and, in fact, the DTD will not allow a <del> to cross structural boundaries). Example: in the manuscript poem, "I see you who no one else sees," there is a single vertical line over the entire page. This would be encoded as:
<teiheader>

</teiheader>
<text type="manuscript">
<body>
<pb corresp="duk.000001" type="recto" id="leaf01">
<delspan to="d1" status="unremarkable" type="hashmark">
<lg1 type="poem">
<l> text of poem here </l>
</lg1>
<anchor id="d1"></body></text>

Cancellation of deletions and other markings TEI Guidelines

An author or scribe may mark a word or phrase in some way, and then on reflection decide to cancel the marking. For example, text may be marked for deletion and the deletion then cancelled, thus restoring the deleted text. Such cancellation may be indicated by the <restore> element:

<restore> indicates restoration of text to an earlier state by cancellation of an editorial or authorial marking or instruction. Attributes include:

type indicates the action cancelled by the restoration.
desc gives a prose description of the means of restoration.
resp signifies the editor or transcriber responsible for identifying the hand of the restoration.
cert signifies the degree of certainty ascribed to the identification of the hand of the restoration.
hand signifies the hand of the agent which made the restoration.

Presume that D.H. Lawrence decided to restore "my" to the phrase first written ``For I hate this my body'', with the "my" first deleted then restored by writing ``stet'' in the margin. This may be encoded:
For I hate this <restore hand='DHL' desc='marginal "stet"'><del>my</del></restore> body

Illegible text TEI Guidelines

We have decided not to encode ink blots or stray pen marks. However, anything that looks like text, no matter how indecipherable, should be encoded. The <gap>, <damage>, <unclear>, <supplied> and <del> elements may be closely allied in their use. For example, an area of damage in a primary source might be encoded with any one of the first four of these elements, depending on how far the damage has affected the readability of the text. Further, certain of the elements may nest within one another. The examples given in the last sections illustrate something of how these elements are to be distinguished in use. This may be formulated as follows:

When a manuscript page is simply missing:

This is addressed by use of the empty <gap> element. A reason= attribute is available and for this situation should be: reason="ms page missing".

When text is completely unreadable:

The text must be at least partially legible in order for the encoder to be able to transcribe it. If it is not legible at all and no text is supplied by the editor in place of what is lost: place an empty <gap> element at the point of deletion or damage. The reason attribute can give the cause of the omission from the transcription as "deletion, illegible."

When text is unreadable but is supplied by another source

Where the text has been rendered completely illegible by deletion or damage and text is supplied by the editor in place of what is lost: surround the text supplied at the point of deletion or damage with the <supplied> element. Use the reason attribute to state the cause (damage, deletion, etc.) of the loss of text leading as well as the source of the supplied portion. For example, this excerpt is from our transcription of "Ashes of Roses":

Are we to have a National Hy<supplied reason="this portion of clipping cut off" source="Library of Congress transcription"> mn by</supplied> Centennial time?

When text is readable with perfect or reasonable confidence:

In many cases, despite damage or illegibility, the text may yet be read with reasonable confidence. In these cases, the following elements should be used:

<damage>
Where there is deletion or damage but the text can be read with perfect confidence: transcribe the text and surround it with the <del> element (for deletion) or the <damage> element (for damage). Use appropriate attribute values to indicate the cause and type of deletion or damage. Observe that the degree attribute on the <damage> element permits the encoding to show that a letter, word or phrase is not perfectly preserved, though it may be read with confidence. Attributes include:

type classifies the damage according to any convenient typology.
resp indicates the individual responsible for identifying the area of damage.
hand in the case of damage resulting from an identifiable cause, signifies the causative agent.
agent In the case of damage resulting from an identifiable cause, signifies the causative agent.
degree Signifies the degree of damage according to a convenient scale. The <damage> tag with the degree attribute should only be used where the text may be read with some confidence; text supplied from other sources should be tagged as <supplied>.
extent indicates approximately how much text is in the damaged area, in letters, minims, inches, or any appropriate unit, where this cannot be deduced from the contents of the tag. For example, the damage may span structural divisions in the text so that the tag must then be empty of content.
<unclear>
where the text has been rendered partly illegible by deletion or damage so that the text can be read but without perfect confidence: transcribe the text and surround it with the <unclear> element. Use the reason attribute to state the cause (damage, deletion, etc.) of the uncertainty in transcription and the cert attribute to indicate the confidence in the transcription. Attributes include:
reason indicates why the material is hard to transcribe.
resp indicates the individual responsible for the transcription of the letter, word or passage contained with the <unclear> element.
cert signifies the degree of certainty ascribed to the transcription of the text contained within the <unclear> element.
hand Where the difficulty in transcription arises from action (partial deletion, etc.) assignable to an identifiable hand, signifies the hand responsible for the action.
agent Where the difficulty in transcription arises from an identifiable cause, signifies the causative agent.

Misspellings or similar "mistakes," like incorrect capitalization, etc.

The tags <sic> and <corr> are used in tandem to represent a mistake by the author and a correction. The TEI guidelines say, "If the highest priority is to present an uncorrected transcription while noting perceived errors in the original, the choice will be typically to use <sic> throughout."

Example:
If the original text or manuscript includes the sentence: "This word is mispelled." This would be tagged in the following way:
This word is <sic corr="misspelled">mispelled.

This preserves the way the text appears in the original, but notes that it contains a mistake and offers a correction.

Manuscript leaves

To represent Whitman's own page breaks into manuscript leaves, a pagebreak element should be used <pb> which can occur anywhere, even within a line group. This is an empty element and therefore does not need a close tag. For more information on the required attributes on this element, go to corresponding page images

Reverse side writing

Whitman's manuscripts present a variety of challenges to encoding. Besides the heavily marked-up drafts and the cutting and pasting he often does with pages, he often writes on the backs of manuscript pages as well. To represent this, we have simply borrowed a technique from scholars of medieval manuscripts: each manuscript page will have a designated "recto" (front) and "verso" (back) side, which will be indicated using the required type attribute on the <pb> element.

For example:
<pb corresp="uva.00050.001 id="leaf01" type="recto">

Cutting & pasting of manuscript pages

There are two distinct scenarios for pasted manuscript pages. Please look over both options and decide which best describes the instance you're trying to encode.

Pages are pasted together by Whitman simply to give himself more room to write
One page pastes over another, deleting old material and adding new

Scenario #1

To simply represent the way the manuscript looks to the eye, use the <milestone> element and the available unit attribute. For an example of this, in Myerson's Walt Whitman Archive, Garland, 1993, see v.2, pt. 1, p.53, "?Gliding." The seam of the two pages is encoded thus: <milestone unit="glued">

Scenario #2

Whitman sometimes pasted one page over another in order to add new lines and delete previous lines (the ones underneath the pasteover). There is an apparatus <app> to deal with this difficulty. The first reading <rdg> is what got pasted over, (the deleted portion), is referred to as "pasteover". What was added <addSpan> is referred to as "pasteon."

Please note that this occurs within a linegroup <lg> but outside of a line, so the <delSpan> and <addSpan> elements MUST be used instead of <del> and <add>.

For example, see leaf 2 of [When I heard at the close of] here.
<app><rdg varseq="1"><delSpan type="pasteover" to="d1">
<l><seg>And that night O you happy </seg>
<seg>waters, I heard you beating</seg>
<seg>the shores – But my heart</seg>
<seg>beat happier than you – for</seg>
<seg>he I love is returned and </seg>
<seg>sleeping by my side,</seg></l>
<l><seg>And that night in the stillness</seg>
<seg>his face was inclined toward</seg>
<seg> me while the moon's clear</seg>
<seg>beams shone,</seg></l>
<l><seg>And his arm lay lightly over my</seg>
<seg>breast – And that night I</seg>
<seg> was happy.</seg></l>
<anchorid="d1"></rdg>
<rdg varseq="2"><addSpan type="pasteon" to="a1">
<l><seg>And that night, while all</seg>
<seg>was still, I heard the</seg>
<seg>waters roll slowly continually</seg>
<seg>up the shores</seg></l>
<l><seg>I heard the hissing rustle of</seg>
<seg>the liquid and sands, as directed</seg>
<seg>to me, whispering, to congratulate</seg>
<seg>me, – For the friend I love lay</seg>
<seg>sleeping by my side,</seg></l>
<l><seg>In the stillness his face was in-</seg>
<seg>clined towards me, while the</seg>
<seg>moon's clear beams shone,</seg></l>
<l><seg>And his arm lay lightly over my</seg>
<seg>breast – And that night I was happy.</seg></l>
<anchorid="a1"></rdg></app></lg>

Collage-like additions

Sometimes Whitman pastes other material to his manuscripts. For example, in "Ashes of Roses" (Myerson, v.1, pt.1, 153), Whitman has pasted a newspaper clipping in the lower left-hand corner of the first leaf. In these cases, use the <addspan> element and its available "type" attribute. In this particular case, the author of the newspaper clipping is "unknown."

<addspan to="a1" place="marginbot" hand="unk" type="pasteon">
Are we to have a National Hy<supplied source="Library of Congress transcription">
mn by</supplied> Centennial time?
<anchor id="a1">

Please remember that hands other than Whitman's must be declared in the header (see the discussion of "notes" below).

Notes

For writing that is not part of the text -- either is wholly unrelated to it or conversely, comments on it -- use the element <note>. When the <note> occurs inside <text> (as in the first two scenarios below), it should also

include a place attribute with one of the following values: "margintop," "marginbot," "marginleft," "marginright," or "interlinear";
contain (paragraph) tags inside: <note></note>.

First, make sure that the writing is not more appropriately tagged with <add>. The <add> element is used for writing in Whitman's hand that is clearly indicated as having been added to the base text. If an addition occurs in the margin, use one of the following values for the place attribute: "margintop," "marginbot," "marginleft," or "marginright." Do not change these values or add new ones to make the location more specific.

Use <note> for basically all other situations. The type attribute is required, and available values are "authorial" "editorial" and "project."

Use <note type="authorial" place="xxxx"> when Whitman comments on the text or seems to be writing a reminder to himself (and this can occur inline).

Use <note type="editorial" place="xxxx"> for writing on the manuscript that is not in Whitman's hand, and include the resp attribute. For example, on some of the pages of Leaves of Grass in the UVa collection, Fredson Bowers has added numbers in the left-hand margin. Click here to view one of these manuscript pages. This writing should be encoded as follows:
<note type="editorial" resp="fb" place="marginleft">

#14
<lb>p. 360

</note>

In both of these situations the <note> should be encoded as close as possible to the structure it refers to. So if a marginal comment refers, say, to a specific line, the <note> should go there (with its accompanying place= and type= attributes). However, if the <note> refers to the whole poem, as in "This poem is terrible," it should go outside the <lg type="poem"> (but inside the <body>).
Use <note type="project"> to describe (not transcribe) any writing on the manuscript that pre-dates Whitman's--that is, when he has scribbled notes or drafts of poems on things such as opened-up envelopes from fans or forms from various government agencies. In these cases create a note in a <notestmt> in the TEI header, within the <filedesc> and just before <sourcedesc>. To point to the part of the manuscript being described, include the attribute target, the value of which should be the same as for the id attribute of the <pb> page break. (Note that is unnecessary for such notes, unless you want to write more than one paragraph of description.) Here's an example:
<teiheader>
<filedesc>
. . .
<notestmt>
<note type="project" target="leaf01v">Verso of manuscript leaf is addressed to Walt Whitman, Camden, New Jersey, postmarked September 25, 1890.</note>
</notestmt>
<sourcedesc>
<bibl>
<author> Walt Whitman </author>
. . .
</teiheader>
. . .
<pb corresp="loc.00046.001" id="leaf01v" type="verso">
. . .

Resp Values

Anytime you use the resp attribute (whether on <note> or <unclear>, etc.) you must also put into the header information that glosses the value(s). This is done by including the following structure directly before the <revisiondesc> (revision description):

<profiledesc>
<handlist>
<hand scribe="Horace Traubel" id="ht">
</handlist>
</profiledesc> Current list of resp values. To add to this list, email Brett and Peter.

Name primary location of mss code

Bowers, Fredson UVa fb

Traubel, Horace Library of Congress, Feinberg Collection ht

unknown n/a unk

Whitman, Walt n/a not necessary to declare Whitman's hand

Whitman's editorial marks

Whitman was highly versed in printing technology and therefore often had specific ways of indicating on the printed page how he wanted his texts to appear in written form. For example, he uses all the standard editorial characters and abbreviations to indicate changes and substitutions. For galley proofs (i.e. handwritten changes on printed proofs), click here.There are two different scenarios which must be addressed in order to make markup decisions:

A mark which indicates a change should be made from what is on the page. For example, the use of a caret to indicate an addition. These sorts of changes should be encoded as changes with the relevant elements. See the section on "deletions, additions and substitutions." Do not worry about representing exactly the way the mark looks on the page; instead, the priority is to encode the function of the mark.
A mark which simply emphasizes the way a certain item should appear in print. For example, Whitman might triple-underline the first letter of a word to emphasize he wants that letter capitalized; in the case of hyphenated compound words Whitman sometimes includes horizontal curved brackets between each of the two words and their hyphen, to indicate there should be no spaces in the word. The difference from scenario #1 (above) is that no changes are actually made, but the desired appearance of the character or word is emphasized by Whitman's marks. DO NOT ENCODE THESE CASES. We may, in the next stage of our project, choose to go back over some documents and represent these marks, but for now, don't attempt to encode them.

Written pen lines

Whitman fairly often draws a long line to signal the beginning or the end of a unit of text. You can see examples at the bottom of this manuscript and at the top of this one. We take these lines to indicate some kind of division, though we make no claims about the sort of unit(s) they define. They are encoded using the empty milestone element, with "undeclared" as the value of the unit attribute and "horbar" as the value of the rend attribute. The first exampe above should be encoded as follows:

<seg>We never separate again.—</seg></l>
<milestone unit="undeclared" rend="horbar">
</lg1>

Bylines

Occasionally Whitman will write the title of a poem and then follow it with "by Walt Whitman." For these instances, please enclose this information in the <byline> element within the <lg> but before the <l>.

Signatures

Sometimes Whitman signs a poem with his name and we don't consider this a line or part of the poem. Assuming the signature occurs at bottom, insert a <signed> element after you close the <lg> and before you close the <body>.

Example:
<seg>blah blah blah</seg>
</l>
</lg>
<signed>Walt Whitman</signed>
</body>
</text>
</tei.2>

Bracketed passages

This section covers Whitman's use of brackets to group lines and bits of text. These should be marked up with the element, which although also an empty element, works slightly differently from <addspan> and <delspan>. The "value" "from" and "to" attributes are all required, and so it would work like this:

<l>blah blah</l><anchor id="s2">

Note that here the "value" attribute, defined in the guidelines as "identifying the specific phenomenon being annotated" works kind of like type. And instead of simply having a "to" and an anchor (as in <addspan> and <delspan>), you must include a "from" and a "to." So simply assign the span an id, here "s1" and give the "from" the same id (which implies "start from here.")

Materials which accompany manuscripts (notes, transcriptions)

For now, we have decided not to transcribe and encode any material not written by Whitman. For example, various materials often accompany original manuscripts, such as notes on the text or transcriptions. Eventually, these will be encoded as separate documents and linked to the relevant manuscript, but this work will happen at a later time.

Naming and genre

Deletions, Additions and Substitutions

Gaps and mistakes

The Artifact of the Manuscript Page

Other Written Phenomena

Deletion with an over-written addition

Resp Values

Written pen lines