The New Edinburgh Edition of the Collected Works of Robert Louis Stevenson

Writing Explanatory Notes

with 4 comments

I don’t know about you, but when I pick up a new annotated edition I go straight to the explanatory notes—the salted peanuts of the volume as far as deliciousness and difficulty of stopping are concerned. Unlike salted peanuts, however, they are all different: more like a series of entries in that fascinating publication Brewer’s Dictionary of Phrase and Fable. It would seem that we are hard wired as a species to like random information—which might explain, to a puzzled observer like myself, the addictive nature of mobile phones.

When it comes to writing such notes, however, you are immediately aware of a series of problems, and it was to share ideas about this that Prof. Burkhard Niederhoff kindly invited fellow essay-editor Lesley Graham and myself to speak for a morning in June this year at the University of Bochum. This gave us an opportunity to think about our experience in writing notes; what follows are a few notes about what was said, followed by a summary of a recent article and information about a conference on this very subject.

General considerations (Lesley Graham)

1. What we annotate. 1. literary, biblical and historical allusions; 2. words that are not immediately understandable; 3. words in a foreign language; 4. proper names; 5. cross-references to themes covered by Stevenson elsewhere; 6. biographical details; 7. facts that can be checked.

2. The imagined reader. Someone a little like ourselves. Not a school child, but not a literary expert either. Not an expert in either essays, history, philosophy, the life of Stevenson or 19th Century philosophy, but someone with a curious mind. Not necessarily a Westerner, but someone with at least a basic knowledge of the Western literary canon. Not a fluent speaker of French or Italian or Scots, or German, nor a reader of Latin, but someone that knows what voilà and al fresco mean. And probably someone who will occasionally like to take some of these notes a little further.
[RD: 1. For the essays in particular, we imagine a range of readers: the notes here, apart from as a way of understanding the text, are going to be read for a wide range of documentary reasons, so, for example, it’s a good idea to provide relevant background biographical information.]

3. Wouldn’t digital annotation be better? Hyperlinks can be detrimental to a profound reading experience; the affect is engaged in different ways when we see an essay as a whole self-standing thing with its own current and internal logic rather than an organic, pulsing jump off point. Our job in the volume is to accompany the reader, without allowing him/her to go off on long detours, in a reflective reading experience, to facilitate the reading of the work, to accompany the reader’s understanding and appreciation of it but doing some of the hard work of establishing context, definition, and allusion but none (or very little) of the joyful, heuristic work.

4. Restraint. There are so many fascinating facts to be found and that may seem highly relevant but when you step back and look at the whole thing again, you realize you have to cull the extraneous material if you want to preserve the joy of discovering the essay for the reader.

5. Some problems found while annotating the essays.
1. Stevenson writes either two or three parish churches: should we supply information where the author is deliberately vague? (in this case, no).
2. Obscure terminology, e.g. travellers, and headers, and rubble, and polished ashlar: should we define terms that Stevenson did not expect his readers to know? (we decided, yes in this case).
3. Avoid self-indulgent additional information, e.g. Turnberry Point: should we mention the Trump golf-course there now? (no; restraint required). Dunfermline, in whose royal towers the king may be still observed (in the ballad) drinking the blood-red wine: in supplying the allusion to the ballad should we add historical information about royal connections of the town (we thought not).
4. And a lord he was (reading a Greek New Testament on the beach of Fair Isle): do we need to identify him? (yes;  Stevenson does so in his letter from Fair Isle).
[RD: Item (iv) is a case of Stevenson’s use of allusions that are known to only a small number of readers or often only by himself. The essay editors in their discussions have called these ‘bald allusions’ and Barry Menikoff, with reference to Kidnapped, has called the phenomenon ‘subtextual meaning’: ‘Stevenson actually defies rather than helps his real readers. He forces them to uncover the allusions for themselves, but nothing is lost if they do not. For the surface prose is sufficient unto itself; the literal meaning of the text can be followed with no difficulty whatever’ (Narrating Scotland, 60; see also 59-60, 91-2, 109). I think it’s clear that the reader of an annotated edition will want to have these allusions explained.]
5. My business lay in the two Anstruthers: do we provide date and Stevenson’s reason for being there? (yes: the essays will be used by those interested in Stevenson’s biography.)
6. Shell House: should we supply more information about this place? (yes: the essay will be used by those interested in local history, so we need to explain the ‘snatches of verse’ and to point out that its location as an ‘outpost’ to Anstruther Wester is the result of a confusion with another shell-decorated house).

6. Excluded from Explanatory Notes in the essays.
1. any analysis of the structure of the essay, of its internal logic.
2.the editor’s personal reading, irrelevant in a scholarly edition. This is at once frustrating (I would love to tell you about how I believe ‘An Education of an Engineer’ is really about the difficulty of communication and the risk of miscommunication, but also humbling and affords freedom for follow up studies and analyses.)

7. Lightness (RD). 1. put first the most important and the relevant things; 2. put things in chronological (and other ‘natural’) order; 3. don’t use complicated series of subordinate clauses etc.; 4. give the most probable explanation without too much hedging.

Writing definitions (Richard Dury)

1. Place the gloss or definition first.
Scots law the legal system of Scotland: under the 1707 Act etc. [more information]
A medlar the fewer on the three-legged medlar-tree! one rotten fruit the fewer on the gallows tree; medlar: etc. [glosses and commentary on ‘medlar’ and ‘three-legged’].

2. Make the gloss syntactically equivalent to the lemma. Examples: check-string cord etc (not ‘a cord’); doubled … with played by the same actor as. The definition should ideally be able to replace the lemma in the text.

3. Don’t copy-and-paste the OED definition. This is difficult in our edition anyway, where the note starts with a lower case letter if preceded by an elided ‘is’ or ‘means’ (e.g. ‘Scots law [is] the legal system of Scotland’) and OED definitions always start with a capital letter; difficult anyway because you will often be glossing a noun or verb not in the dictionary citation form, so the ‘lemma’ followed by the OED definition would not make a coherent sentence (see previous item). Other reasons for adapting the OED definition or writing your own definition are given in the following points. [LG It may be useful to compare the OED definition with other dictionary definitions]

4. Make OED definition clearer and more concise if necessary. OED definitions will cover many cases, for the Explanatory Notes only the relevant parts should be included. You may also be able to make the formulation less wordy than in the OED: e.g. : check-string a string by which the occupant of a carriage may signal to the driver to stop (OED) / cord inside a carriage, pulled to tell the driver to stop.

5. Look critically at OED definitions and the citations. The OED, like all sublunary things, is not perfect; you may need to write a definition not found there, e.g. the entry for hold the candle does not mention its use (from French) of ‘assist in a love affair’, though Stevenson uses the phrase alluding to this meaning on at least a couple of occasions. In other cases, reading the citations carefully will reveal a meaning not listed. [LG: When the citation is the very sentence you seek to elucidate, give yourself a clap on the back]

6. Look elsewhere for help in defining a word or phrase. 1. Look elsewhere in Stevenson’s works, e.g. a puzzling use of motive may be resolved by finding that he sometimes uses the word to mean ‘motif’ (for this you will need a corpus of Stevenson’s writings; I’ll try and provide this asap). 2. Look in related entries in the OED, e.g. a reference to Henry James and his humorists of ordinary life may be solved by looking at ‘humour’ meaning ‘a particular disposition, inclination, or liking’. 3. Search internet (including using Google Advanced Book Search) so that, instead of the general OED definition, sinnet, for example, can be defined with relevance to its use in the text: ‘braided, rather than twisted, cordage, (here) the typical flat, plaited coconut-fibre cords of the Pacific islands’. 4. Take into account what the reader needs to understand, phenomena possibly assumed as known by the OED, e.g. aspects and connotations of vanished Victorian domestic life such as pass-key and area. 5. Inspect the cognate word in French (see next point).

7. Be aware that Stevenson often invents new uses of words. The context is of more importance than the OED in determining Stevenson’s ‘nonce’ meanings, e.g. a generic in the following: ‘Boswell’s is, indeed, a very special case, and almost a generic’. After studying the whole context very carefully it seems clear that Stevenson is using ‘a generic’ to mean ‘a case apart, i.e. a genus on its own’, not found in the OED. Stevenson’s nonce-words or -uses are often calques from French: checking in the online Trésor de la lange française is a good strategy in doubtful cases. Check with Google Advanced Book Search to confirm a suspected original use by Stevenson.

8. The OED may only provide negative information (show what is not possible). For example blowing in the key cannot mean ‘thrusting in the key’ as there are no examples of ‘to blow’ as a verb derived from (the etymologically unrelated) noun ‘a blow’ (it means ‘blowing to remove any dust from the key before inserting it’).

3. An article on annotation

Matthias Bauer and Angelika Zirke (2017), ‘Explanatory Annotation of Literary Texts and the Reader: Seven Types of Problems’, International Journal of Humanities and Arts Computing, 11.ii (2017): 212-232.

The second part of the article describes a model of layered annotations for digital editions, but the following notes are only on the first part, which examines seven ways that existing annotations  do not take into account readerly needs. The comments there are of interest to both print and digital annotation.

The authors propose not to attempt guidance by an imagined readership but to provide annotations that are of use to a variety of readers and do not think that interpretation should be discarded from annotations altogether. They then analyze a series of actual annotations by asking two questions: (A) What knowledge does the annotation presuppose? (B) What knowledge does it provide? Here are the seven types of annotation problems.

1. Stating the obvious: explaining something that any potential reader will know already or can gather from the text itself.

2. Inconsistent assumptions and unclear functions: where the explanation assumes lack of knowledge in an area but uses unexplained terms from the same area; or provides information which does not have a clear function in understanding the text.

3. Presupposing (expert) knowledge: the assumption of knowledge that is never made explicit, as for example in a quotation from the same author that supplies a wider context (an enriching rather than explanatory annotation), when this is not clearly related to the text being annotated (just introduced by something like ‘Compare’).

4. Sending the reader on the wrong track: for example, using a general definition from the OED that omits important contextual meaning in the text being annotated or meanings supplied by the author’s personal use elsewhere.

5. Delimiting interpretation: giving one definition/explanation where more than one is plausibly present.

6. Offering intuitions without evidence: giving personal reactions (in an essay-like fashion)—the text annotated ‘becomes an occasion to think about one’s experiences and feelings’

7. Missing annotations: a missing annotation tells us either that nothing is to be explained or that explanation is impossible—in the latter case the difficulty should be dealt with in a note anyway.

The second half of the article gives information about digital annotation using the ‘Tübingen Explanatory Annotation System’ (TEASys), using three levels of information and eight categories that classify the content. As NEd is not using digital annotation, this part is less directly relevant to us. But here are the eight categories of annotation content which could well be of interest:

A   linguistic (lexicon, syntax etc.)
B   formal (verse, narrative structure, iconicity etc.)
C   intratextual (motifs, recurring structures etc.)
D   intertextual (relations to other texts)
E   contextual (biography, history, philosophy, theology, etc.)
F   interpretative (synthesis of A–E)
G   textual (variants relevant to the understanding of the text)
H   questions (items that require annotation; comments on research already done relating to an item).

A conference on annotation

‘Understanding (through) Annotations’ (15th International Connotations Symposium): July 28 – August 1 2019, Eberhard Karls University Tübingen (Germany).

We invite papers that are concerned with annotations to specific literary texts written in English and address their functions. Papers may also reflect on the speakers’ own  annotation projects, analyse existing annotations, offer suggestions as to a more systematic approach to the practise of annotating texts, and/or discuss historical and theoretical dimensions involved, such as the relation of lemma and context, part and whole, the envisaged reader of annotations, etc.

Please send an abstract (300 words max.) to the editors of Connotations by October 15, 2018 at symposium2019@connotations.de

See also Writing Explanatory Notes/2

Written by rdury

27/09/2018 at 1:55 pm

4 Responses

Subscribe to comments with RSS.

  1. Here are a few follow-up notes about the possibility or practical usefulness of (i) determining rules for writing explanatory notes and (ii) identifying a target reader and adapting the notes accordingly. I jotted these down as a contribution to an email exchange on the same subject.

    1. I think that we have to concede that there is no ideal reader, in that no real person will ever correspond to the identikit picture we have in our heads. Knowledge and experience profiles are irregular and unpredictable. But perhaps we can hope that our un-ideal reader is a tolerant type who will skip over the notes that are superfluous for him/her without feeling that their intelligence has been insulted.

    It has been suggested that notes could be tested on a group of readers of the type we are aiming at, but this seems difficult since the characteristics of the group will be just as inconsistent as those of the individual. Take any class of undergraduates and you’ll find that a certain percentage of them studied Latin at school, another set learned a lot from a television series about Roman Britain, others were brought up in Scotland, some hated Geography classes and can barely name a capital city … you get the idea. I don’t know what smoothing out all that deviance would do to the data set. Anyway, the Latin-readers can always skip the Latin glosses, the Roman experts can skip the dates of Augustus etc. And they may even get a little confidence-building thrill from the fact that they don’t need the note.

    2. However, putting the notes through the filter of several pairs of eyes of colleagues and fellow-editors is definitely an advantage. That word with a meaning that is transparent for one editor, may be completely unknown to another. After that, it’s a question of negotiation, whose vocabulary is closest to that of the average reader of a scholarly edition? Is one of the editors an outlier? If in doubt, lean towards under-annotating rather than over-annotating lexical items. We are all capable of extrapolating the meaning of some unknown words. I spent a lot of my childhood guessing what galoshes and gymkhanas were, but it didn’t spoil my enjoyment of boarding school literature, in fact it probably heightened the impression of an encounter with the exotic.

    3. The list of things that should/could be annotated is not set in stone. Full disclosure: I came up with the list in the blog post a posteriori, having already done the first draft of annotations for a good number of essays. The real process is, I’m sure you will all agree, much more intuitive than following a list. I find the technique set out in the article referenced at the end of the blog post, with its layered system of annotation, rather alarming. It sounds as if they’re moving towards annotation by artificial intelligence.

    It seems to me that the greatest danger lies in self-indulgence; in wanting the reader to admire how clever we have been in digging up an interesting fact, or cross-reference, or some random correspondence. From this point of view, I admire the self-restraint exemplified in Mehew’s notes. This week I came across a note in which he tells us that the very naughty child who said something that made him laugh was Bertrand Russell. And that is all he says: no knowing commentary, no exclamation mark, just the fact. Such self-restraint!

    4. On the subject of the irrelevance of the author’s biography, I would just say that if our aim is to provide today’s reader with the knowledge that the late 19C reader had at his/her disposal, we have to recognise that Stevenson’s life story was well-known to most readers and that we may have to explain some of the biographical allusions to give our readers that same advantage. Having said that, one does wonder just how familiar the typical Scribner’s reader was with the intimate relevance of the topography of Edinburgh for Stevenson, and with that admit that some of the biographical detail is not deliberately obscure exactly, but that Stevenson didn’t really care if 99% of his readers didn’t “get it”. But we have to care.

    5. I have found it useful to look at the notes for the few essays in my volume that have been annotated in the past, realising that their extent and content is defined not by the intended audience but the editor’s agenda. Some are inaccurate. Some are so erratic and random as to be of no use whatsoever. Some are so overblown that they are tedious to read. Some only address one category of information (Scottish history, literary criticism …) and that is fine if it is the declared rationale for the book, but it isn’t our remit. The occasional note is breathtakingly judgemental and condescending, like this one on a popular French novelist in which the editor adds “He …testifies to Stevenson’s somewhat indiscriminate taste and his inability to put childish things behind him at a literary level”.


    17/10/2018 at 5:01 pm

  2. The following were some thoughts stimulated by the same email exchange mentioned by Lesley, with some reference to Lesley’s notes too:

    1. It’s useful to distinguish between rules and guidelines. Rules are intended to be followed always and will be corrected by the copy-editor. Examples are ‘start the note with a lower-case letter if you could insert an ellipted “is” or “means” after the lemma’. Guidelines are intended to give an idea of a guiding spirit that will help in creating uniformity in selection and expression (i.e. in what is considered relevant and in compositional style). We agree that guidelines cover most of the activity of writing notes. A line-editor will suggest changes if a note appears too far from these.

    2. Guidelines for a single editor will be an internalized feeling of what is right; editors working on a coordinated project will be guided in the same way, but some written guide is also necessary. So I agree with Lesley: ‘The real process is […] much more intuitive than following a list.’
    It is also possible to make guidelines that could cover what is considered best practice in annotations (for a certain type of edition), based on notes that have been judged successful: this is a question of opinion, but some notes will be better than others and it is useful to try to learn from them.

    3. An editor can invent what rules and guidelines he wants, but then has to follow them. This invention, however, has limits, i.e. the broad consensus of what notes should contain and look like. Going beyond these limits, the notes risk looking eccentric. Following no rule or guidelines, the notes will be too varied in form and will leave the reader nonplussed. (As Lesley says, ‘so erratic and random as to be of no use whatsoever’).

    4. In the original blog posting we proposed guidelines for our edition and analysis of what we consider to be best practice, not proposing objective rules to be followed in each and every case. We are all in agreement that the latter endeavour is an illusionary goal. As Stevenson says in ‘Lay Morals’, reality is too complex and unpredictable to be guided by precise rules, which either do not fit well most actual cases, or are multiplied, trying to cover all cases, until they are useless.

    5. The imagined reader has to be borne in mind but only as a general guide. The situation is perhaps similar with children’s literature: the best writers repeatedly say that while they bear in mind their audience, they otherwise write what they themselves find interesting or want to write. So I agree with Lesley: ‘our un-ideal reader is a tolerant type who will skip over the notes that are superfluous for him/her without feeling that their intelligence has been insulted’: i.e. readers adapt, learn some things, and (within limits) accept things they don’t understand, don’t object to things they already know, skipping some of these if they like.

    6. We have to be guided by ‘tact and restraint’, i.e. an internalized sense of what is relevant and what is sufficient.
    The opposite of tact and restraint is self-indulgence, as Lesley points out: desire to make the reader admire what you’ve found or reluctance not to include what took you time to acquire. This can be partly solved by repeated drafts and readings by yourself and others. As Hemingway said, ‘You know you’re getting near the end when you’re throwing away good stuff’.
    Some writers of notes will be better than others and not because the notes are tailored precisely to the judging reader, but because they are useful, interesting and well-formed.


    18/10/2018 at 4:42 pm

  3. The following may be of relevance to the writing of explanatory notes.
    Wittgenstein thought that the information in Frazer’s Golden Bough could be better presented without a theoretical gloss ‘by arranging the factual material so that we can easily pass from one part to another and have a clear view of it’. ‘This perspicuous presentation presentation makes possible that understanding which consists just in the fact that we “see the connections” ‘.
    I understand that the two cases are very different (and I have probably not fully understoof Wittgenstein’s ideas) but the idea of perspicuous presentation and seeing connections appealed to be as a possible guide in a simple presentation of information in an explanatory note.


    22/10/2018 at 11:28 am

  4. […] See also Writing Explanatory Notes. […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: