Today, 30 March 2021, is the fifth National Indexing Day – hooray! For NID 2021, here is an A to Z of indexing, covering some key terms, figures and other observations from my near two decades of experience as a professional book indexer.
aboutness: Indexing in a nutshell. What is this sentence, paragraph, chapter or book about? And where can you find that information? That’s what should be in the index.
acknowledgements: The indexer is not often credited or thanked in the acknowledgements section of a book, but it makes our day when we are, so if you love your index, please do this.
acronyms/abbreviations: If an organisation or concept is known as an acronym or abbreviation as well as its full name, the indexer should decide which will be the preferred term and cross-reference from the other – e.g. SI see Society of Indexers / Society of Indexers (SI) 56, 73, 92. But see also double-posting.
alphabetisation: Indexers know their alphabets. Indexes can be sorted in letter-by-letter or word-by-word order. In letter-by-letter order, spaces and punctuation are ignored for filing purposes (so Bluebeard, blueberry, blue moon). In word-by-word order, a space/punctuation mark files before a letter (so blue moon, Bluebeard, blueberry).
articles (a/an/the): These should usually be disregarded for sorting purposes, and some publisher styles require them to be placed at the end of an index entry, so A Tale of Two Cities would then be indexed as Tale of Two Cities, A. The entry will be found at Tale, not A.
authors: Some authors are also very good indexers. However, writing a book and indexing it are very different prospects. For a guide on why you may need a professional indexer and how to find, hire and work with one, see my recent blog post ‘Why you need and how to work with a book indexer: a guide for authors’.
Bell, Hazel K.: Indexer, author and past editor of The Indexer journal. I highly recommend her books Indexing Biographies and Other Stories of Human Lives and Indexers and Indexes in Fact and Fiction, along with her many articles for The Indexer. Hazel’s website is at https://www.aidanbell.com/hazelbell.
bibliography: The bibliography or references section is not included in the indexable text itself, but indexers refer to it when indexing citations in the main text.
biscuits: A useful addition to the indexing thought process. (see also tea)
Booth, Pat F.: Indexer and author of Indexing: The Manual of Good Practice. An excellent guide to the principles of indexing, but unfortunately now often difficult and costly to track down
books: Non-fiction books nearly always benefit from having an index. Some fictional works do too. (see also fiction)
brain fog: A phenomenon occurring after too many hours of intense indexing. Take a break. Eat or drink. Move.
Chartered Institute of Editing and Proofreading (CIEP): The Society of Indexers and CIEP share a common heritage and continuing close ties. The Society of Freelance Editors and Proofreaders (SFEP), as it was then, was founded by a small group of volunteers led by Norma Whitcombe after informal conversations at the SI conference in Cheltenham in 1988.
Chicago Manual of Style: This contains a very useful chapter on indexing, which forms the basis of the indexing style guide of many US publishers. The indexing chapter can be purchased as a standalone publication.
citations: The general rule for citations in the text is to index only those that are substantially discussed, not just those listed in parentheses with no other information. Publisher requirements can vary on this though.
comedy see humour
computers: Indexers work with them and they help us considerably. However, they don’t read like a human, so they can’t index like one.
concordance: A concordance is an alphabetical list of every occurrence of the principal words used in the text. That is not the same thing as an index. Computers are good at concordances, but not so much at indexes.
cross-references: A see cross-reference leads from a heading with no locators to the preferred entry term where the information is located. A see also cross-reference leads from an entry with locators to another entry where further information can be found. For example, book editing see editing / copy-editing 15, 31 see also editing / editing 15, 20–5, 31, 52.
cross-references to make you cross: A blind or broken cross-reference sends you to an entry that doesn’t exist. A circular cross-reference leads to an entry that sends you straight back to where you started (goose chase see time wasting / time wasting see goose chase). A serial cross-reference points to a heading that leads to another cross-reference, rather than to the preferred term. Your reader will not appreciate any of these.
Ctrl-F: A Ctrl-F ‘find’ search has its place, but should not replace an index. It does not distinguish between significant discussion and passing mentions, suggest synonymous access points, show the coverage of topics, or highlight connections between concepts. (see also ebooks)
deadlines: Any freelancer will be well used to dealing with deadlines, indexers especially, as the index is usually done at the last minute before publication. As Douglas Adams said, ‘I love deadlines. I like the whooshing sound they make as they fly by.’
density: This refers to the depth of indexing or length of index. There are no set rules here, as each book requires its own indexing solution. A reasonable average is a 5% index, so a 10-page index for a 200-page book. Some publishers set a maximum index length, but this is often more due to the space left before incurring extra paper costs rather than what the book ideally requires.
double-posting: If a see cross-reference leads to an entry that only takes up a single line, or the same or less space as the cross-reference, it is often best to post the locators in both places instead as no space would have been saved and then the reader is not sent elsewhere unnecessarily. Indexers call this double-posting.
ducks: A walk to see and feed them works wonders for this indexer when her brain gets too full. Thanks, quackies.
ebooks: Ebooks need indexes too, but they are often not provided with them. Indexers have the knowhow and the technology to do our bit to make this happen to enable active linked indexes through the techniques of embedded indexing. Not many book publishers have the workflow processes to enable it though.
elision: Page elision or elided locators refer to the style of page ranges in the index and how many figures are included – for example, is the style 100–105, 100–05 or 100–5. This is often set by the publisher house style. Maximum elision (aka minimal figures) is often preferred as it means the index takes up less space.
embedded indexing: An indexing method where the index entries are anchored (or embedded) into the text in as precise a location as required – page, paragraph or word level. Then if the pagination or format of the book changes, the index locators can update automatically. It is most often used for Word manuscripts, and so can occur at an earlier publishing stage than page proofs, but can also be used with InDesign or XML files. It adds an extra level of complexity and not all indexers work in this way. Professional indexers tend to use additional programs rather than try to index in Word or InDesign directly, as that can be downright painful. (see also software)
entry: An index heading with locators and/or cross-references. There will usually be main entries and additional subentries for larger entries, with some indexes also requiring sub-subentries.
estimates: An indexer will require certain information before providing a cost estimate. These include the deadline, format of index, length of book (ideally word count), publisher style and subject of book.
fiction: It’s usually non-fiction that has an index, but some novels and fictional works do have them. Famous ones include Woolf’s Orlando, Tolkien’s The Lord of the Rings, Nabokov’s Pale Fire and the annotated edition of Jerome’s Three Men in a Boat. Ballard’s short story ‘The Index’ is entirely in the form of an index. There is an indexer character in Vonnegut’s Cat’s Cradle (Chapter 55: ‘Never Index Your Own Book’) and an index central to the plot of Susanna Clarke’s Piranesi (which I recently reviewed for The Indexer; open-access version now on my blog). Rarely seen, and tricky to do, but they have their place.
freelancing: Most indexers are freelancers and self-employed sole traders, working from home, so that at least has been one constant for many in recent strange times. The number of people also in the house while doing that indexing has however changed for several of us.
glossary: Glossary terms are a good source of key entries that should be in the index. The glossary itself may be indexed or unindexed, depending on how much information is given there.
granularity: A good word for indexing, which means the level of detail in something. With a tight maximum index length, the level of index granularity will necessarily be reduced.
hard copy: The paper printouts of book pages. Indexers used to index from hard copy, and many still do, but increasingly indexers are just emailed a pdf of the book and then work entirely on screen.
header letters: This index has header letters for the beginning of each alphabetical section, A, B, etc. Most indexes do not, but this can depend on the publisher’s house style.
headings: The words used in the index list. The heading may also require a breakdown into subheadings if it is a frequent term in the book.
house style: The publisher’s conventions on layout, format and so forth for indexes in their books. If the publisher has an indexing house style document, the indexer will need a copy of it.
humour: Indexes can be funny too, if the book content is appropriate. My blog currently contains review of several comedy book indexes, including the Alan Partridge and Steven Toast ‘memoirs’, and titles by Richard Ayoade, Charlie Brooker and Francis Wheen, with more to follow this year.
index: The list of words and numbers at the back of the book, showing what is in the book and where to find it. That was an easy one.
index cards: What indexers used to use to write their index entries on, hence the name. (see also shoeboxes)
#indexday: The hashtag for National Indexing Day, 30 March. (see also National Indexing Day)
The Indexer: The international journal of indexing, launched in 1958, and now published by Liverpool University Press: www.theindexer.org.
indexes: This is the plural you need for more than one book index, not indices. (Trust me. I’m an indexer.)
introductory note: If an index contains any particular typographical features or other conventions, it can be helpful to include an introductory note at the beginning. This can advise the reader, for example, that locators in italic denote illustrations, locators in bold indicate definitions, locators in the form 20n1 refer to information in the notes, or locators represent paragraph numbers.
J: This is often a missing section in an index as it has no entries at all. It is also a letter that traditionally didn’t exist in the Welsh alphabet. Neither did K, Q, V, X or Z. Welsh does however have the eight bonus digraphs (double letters) of CH, DD, FF, NG (filed after G), LL, PH, RH and TH.
jigsaw: I have seen the process of indexing likened to solving a jigsaw, crossword or puzzle. Each index is a new challenge and it is most satisfying to find the solution that works best.
kettle: An item frequently in use during the creation of an index. Brew, anyone? (see also teapot)
key words: An index is like a list of key words to the book, but indexers do not normally need to be sent a list of them to include. We don’t start with a list of key words and search the book for where they occur. We read from page one and build the list of index terms as we go along.
Knight, G. (Gilfred) Norman (1891–1978): Founder of the Society of Indexers in 1957 and its first Honorary Secretary and Chairman. Author of Indexing, the Art of. See his obituary in The Indexer.
length of index see density
letter-by-letter order see alphabetisation
locators: This usually means the page numbers next to an index heading, but locators can also be in other forms, such as paragraph or section numbers.
locator strings: An index entry with a long list of undifferentiated page numbers following is termed a locator string (or string of locators) in indexer parlance, and indexers are trained to avoid them by not including passing mentions and breaking down large main headings into useful subheadings. An index with many lengthy locator strings is not a good or helpful one.
manuscript: The book manuscript (or, strictly speaking, typescript) is the copy the author submits to the publisher for typesetting. The copy-editing stage happens on the manuscript, as can embedded indexing. This is usually not the final format of the book and so does not have the final page numbers for the index, so indexing is more usually done on the typeset proofs. (see page proofs)
maps: An index is a map of a book. It shows you what things are there and shows you where to find them.
Matthews, Douglas (1927–2021): The prince of indexers. An indexer extraordinaire, who was still indexing into his nineties. The Guardian obituary is here. A longer obituary is in the March 2021 issue of The Indexer (vol. 39, no. 1).
metatopic: This is the main subject of the whole book. For a biography, it will be the primary person discussed. There are various arguments over how much should be indexed under a metatopic heading in the index, ranging from nothing (as the whole rest of the index is about it) to everything. I tend to somewhere in between, with a general overview under the metatopic, but judging what works best for each index.
Mulvany, Nancy C.: Author of Indexing Books, another excellent guide to indexing practice. As she makes very clear from the first page of the preface, ‘Book indexing is something you will either enjoy or detest; there is little middle ground. You will have a knack for it or you won’t.’
music: I love music, but I can’t listen to it when I’m indexing. Something to do with the phonological loop and working memory inhibition, if my past psychology studies serve me correctly. If I can hear music from elsewhere, I use a noise-masking site such as myNoise, and disappear off to an Irish coastline or Japanese garden or the like.
mynegai: This is the Welsh word for ‘index’. An indexer is a mynegeiwr (male) or mynegeiwraig (female) and the verb to index is mynegeio. I am learning Welsh. I can’t index in it. One day, maybe. (Un diwrnod, efallai.)
names: These seem to be the ‘easy’ (ha!) bit of indexing, as they’re quick to spot, but they can cause all sorts of issues – different country and language rules, geographic names, maiden and married names, medieval names, name changes, nobility and royalty, patronymics, pseudonyms, etc. My copy of the textbook Indexing Names, edited by Noeline Bridge, is extremely well thumbed.
National Indexing Day (NID): A day launched in 2017 by the Society of Indexers to celebrate indexes, indexers and the profession of indexing. The first one was invented by SI members Ruth Ellis and Paula Clarke Bain (that’s me). The date commemorates the founding of the Society of Indexers by G. Norman Knight and colleagues on 30 March 1957. In 2018 and 2019, SI held events for publishers in London and Manchester. For now, the NID is back to online only, but the publishing events will return in time. A Wakelet summary of the social media activity for NID 2021 is available at National Indexing Day 2021 #indexday.
notes: The general rule with endnotes and footnotes is to only index substantive new information, not just bibliographical references. Some publishers have different policies about this. To distinguish information in the notes, the locator in the index can be marked in some way, e.g. 20n1 would be note 1 on page 20.
oldness: Indexing is an ancient art. Handwritten indexes are said to date back 3000 years to the Chinese I Ching. Records from the papal court at Avignon show that by the early 1300s people were being paid to compose indexes. The oldest western printed indexes are thought to be found in two editions of St Augustine’s De arte praedicandi (‘On the art of preaching’), published in the 1460s soon after the invention of the Gutenberg printing press.
organisation see alphabetisation
overanalysis: To be avoided. The indexer should not rewrite the book in the index.
page one: Where the indexer starts reading (or page i where there are prelims). Yes, we do have to read the whole book. We get asked that rather a lot.
page proofs: The typeset book file. Indexing is usually done on the proofs when the page numbers are final. Indexers can work from a hard copy of the proofs or from a pdf file on screen.
passing mentions: A term used for the mention of a name or concept in the text without any useful information given. Indexers are advised to avoid including these in the index, although distinguishing between what is and isn’t a passing mention can be debatable.
quacks see ducks
qualifiers: Some index entries may require extra qualifying information if there is likelihood of confusion without it, for example if there is more than one person with exactly the same name (quite common in biographies). This could take the form of birth/death dates, relationship to the main figure in a biography (e.g. mother of, daughter of), title, profession, etc. – for example, McQueen, Steve (actor) / McQueen, Steve (director).
qualifications see training
quotes see estimates
readers: The ones an indexer bears in mind when indexing a book. Who are they, what will they need or want to find, and how will they look for it?
reading: Indexers do a lot of this and need to enjoy it.
references see bibliography
run-on style: In run-on style, subentries run on the same line from each other, separated by semi-colons. It can save on space and is often the best choice for more narrative texts such as biographies.
set-out style: In set-out style, subentries are indented under the main entry, each beginning a new line. This style can be clearer to read, but also takes up more pages.
shoeboxes: A vital piece of indexing equipment from the not-too-distant past. Indexers used to write their index entries on index cards, sort them into order in boxes – shoeboxes being a good size for this – and then either send the box of cards to the publisher for typesetting, or type the full index and post the typescript. We no longer have to do this.
Society of Indexers (SI): The UK professional body for indexers at www.indexers.org.uk. The Society runs a training course in indexing and maintains the online SI Directory of Professional Indexers, searchable by subject and skill.
societies, indexing, other: There are other indexing societies around the world including American Society for Indexing (ASI), Association of Southern African Indexers and Bibliographers (ASAIB), Australian and New Zealand Society of Indexers (ANZSI), China Society of Indexers, Deutsches Netzwerk der Indexer (DNI), Indexing Society of Canada (ISC) and Nederlands Indexers Netwerk (NIN).
software: Indexers use dedicated indexing software, which helps speed up certain formatting and sorting tasks, but doesn’t replace the intellectual work of a human. The main programs for standard back-of-the-book indexes are Cindex, Macrex, and SKY Index. Some main embedded indexing programs are DEXembed, Index-Manager and WordEmbed. I use SKY Index and Index-Manager but it’s important to find the right ones for you.
sorting see alphabetisation
Stauber, Do Mi: Indexer and author of Facing the Text: Content and Structure in Book Indexing, which is a great guide to indexing practice, especially for arts, humanities and social sciences texts.
strings see locator strings
subentries: The subheadings plus locators used to break down a large main entry. These are usually sorted in alphabetical order, as the rest of the index, but sometimes another order might be more helpful, such as chronological (earlier to later events), by page order, or by themes in a biography (e.g. character, education, relationships, works).
swearing: This can be used to good effect within an index, if it has been used in the main text. It can also help a stressed indexer while doing the indexing work.
tagging: A step between standard indexing and embedded indexing. Some publishers require the indexer to add and use location codes when creating the index. These are then converted by the typesetter to produce the correct page numbers in the final document. This is not embedded indexing as here the index copy stands in its own right and the codes are entered by the typesetter.
teapot: A pot for brewing tea. Another key piece of equipment for this indexer. (see also kettle)
The: The definite article ‘The’ should be ignored for sorting purposes, so The Lord of the Rings should be filed in the ‘L’ section, and might appear as Lord of the Rings, The if that is the publisher house style. If you find all the main entries beginning with ‘The’ in the ‘T’ section, that’s an early bad sign of what might be wrong elsewhere in an index.
time: Or how long will the index take? Again, it all depends on the book, but a rough guide for an experienced professional’s indexing speed is 10 pages per hour, so maybe 30 hours for a 300-page book. The number of words per standard page can vary widely though, which is why many will prefer to base estimates on word count.
training: Indexers can train via the distance-learning course run by the Society of Indexers leading to the qualification of Accredited Indexer and an entry in the Directory of Professional Indexers. An overview of the course is here.
turnover lines: If an index entry runs over more than one line, standard guidance is to indent the turnover more deeply than the lowest subentry indention.
typescript see manuscript
undifferentiated locators see locator strings
user needs see readers
value of indexes: Have you ever tried to use a book with a bad index? Or one with no index at all? The horror…
vocabulary: The index will largely follow the vocabulary of the main text, but the indexer will also consider extra access points and synonyms which might not be mentioned in the book at all. For example, if the index has a ‘food’ section, the indexer may add cross-references from headings such as diet, nutrition, etc., depending on the readership.
Washington Read: This is the term given to checking an index for references to oneself and then only reading those parts of the book.
Welsh: Dw i’n dysgu Cymraeg. Mynegeiwraig dw i. (I am learning Welsh. I am an indexer.)
Wheatley, Henry B. (1838–1917): The ‘father of British indexing’, founder of the Index Society in 1877, and author of What is an index? (1879) and How to make an index (1902). See more at ‘The father of British indexing: Henry Benjamin Wheatley’.
word-by-word order see alphabetisation
word count: The number most useful to me when estimating the cost of an index. The number of pages is handy, but because the number of words per page of text can vary so widely, the word count will enable a more accurate estimate.
X often stands for nothing at all in an index, as few indexes will need an X section, unless there are certain names beginning with it. You may see the odd xylophone or x-ray and sadly far too many instances of xenophobia.
yesterday: A deadline that cannot be met.
Z is also often a blank section in an index, unless the names require it or you’re working on something particularly zebra, zoology or zombie related.
zeds: Let your tired brain do its thing overnight and you’ll be a better indexer tomorrow. Get some sleep. Nos da (good night).
Paula Clarke Bain is an Advanced Professional Member of the Society of Indexers and the Chartered Institute of Editing and Proofreading and a new Member of the Society of Authors. She tweets as @PC_Bain and her website (with comedy book indexes blog) is baindex.org. Find the Society of Indexers on Twitter @indexers and www.indexers.org.uk.