Something which I see all the time in (popular) science writing is the use of abbreviations to indicate concepts. For (a made up) example:

So when we're dealing with Anachronistic Meta Mechanics (AMM) we have to take a wholly different approach than before in the case of Amniotic Uber Psychotics (AUP). Those in favour of AMM in fact frequently disagree with the conclusions reached by applying AUP mechanisms to the same data set...

Since I'm not familiar with these abbreviations from prior experience I always end up looking back to where the abbreviations were first introduced. This severely impedes the progress I can make across a text like this. However, since it is an extremely common practice I figured I'd ask if this is a problem more people have and whether there are good alternatives.

  • Is this question different from this one: writers.stackexchange.com/questions/878/… – John Smithers Feb 20 '11 at 17:26
  • 1
    There's a difference between using technical terms and using abbreviations for technical terms. You need to explain what the terms you're using mean anyway, but you don't necessarily have to use abbreviations. – David Rutten Feb 20 '11 at 19:30

If they are abbreviations which are extremely common to the field, once per work is often enough to define them. If they are rare, invented for the piece, or really jargon, I would say once per section (once per chapter, once per web page).

Alternatively, a list of acronyms at the beginning or end of a piece might also be helpful.

  • 1
    If it's popular science writing your audience is probably not well versed in 'the field'. So even if the abbreviation is common amongst experts, you cannot expect your readers to memorize them that quickly. – David Rutten Feb 20 '11 at 14:56
  • @David: good point. "Know your audience" applies to practically everything. – Lauren-Clear-Monica-Ipsum Feb 20 '11 at 19:24

IBM is a bastion of abbreviation use. Three-letter acronyms (TLAs) are literally the foundation of the company -- along with the misnomer calling all abbreviations acronyms (even ones that don't spell out a new word).

In IBM style, even cutting edge research papers and journal articles used the "define on first reference" rule for new terms and abbreviations. The argument was that, once ingrained, an abbreviation speeds comprehension. My feeling always was that authors used the technique to shorten writing time more than as a way to help their readers. Sentences ended up looking like alphabet soup.

In this age of digital content, it is a little easier to search back for the first reference, but this still "impedes progress" as David points out. The idea of a list of abbreviations used in a piece was never common practice, but it absolutely should have been. (Of course, in HTML each reference to an abbreviation can be tagged so that, in many browsers, a pop-up appears with the abbreviation, obviating the need for a search. Authoring can be done using software that matches the term against a global taxonomy so that the accurate description of even the most uncommon abbreviation is always used.)

  • 1
    "software that matches the term against a global taxonomy so that the accurate description of even the most uncommon abbreviation is always used" You mean that the authoring software should be RFC 5513 compliant? – user Dec 19 '12 at 9:36

If you use a lot of abbreviations that may not be familiar to the reader, then readers may find they have to continually refer back to see what they mean. But if you don't use abbreviations, the text can get tedious and very repetitive-sounding.

Like -- let me see if I can make up a couple of sentences to illustrate:

The Warp Drive Autonomous Decontamination Peripheral (WDADP) can prevent your ship's engines from failing. A WDADP keeps foreign particles out of the engine. Most WDADPs are made from titanium alloys. Be sure to check your WDADP monthly. A well-maintained WDADP ... etc

Try re-writing that paragraph, but spelling out WDADP. It becomes terribly wordy and awkward.

On the other hand, every now and then I come across a sentence that's practically one long stream of abbreviations. "FTP your ASPX files to the IIS server before starting the CRON task on the LINUX box" and the like can be very intimidating to someone who doesn't use these abbreviations every day. (Every now and then I read a sentence filled with such abbreviations and get a chuckle from the fact that I understood it completely despite its obscurity.)

Personally, when I have a long technical term that is used repeatedly in a document, I'll abbreviate it with an explanation on first use. If I expect an abberviation to be very familiar to my target audience, I'll use it without hesitation and often without explanation. But if something is likely to be unfamiliar and only turns up a couple of times, I spell it out.

  • Strictly speaking, the only two abbreviations that qualify as such in your second example would be FTP and IIS. "ASPX files" specifies which "files" the "FTP" applies to (it expands to "files matching *.ASPX"). CRON and LINUX aren't even abbreviations; cron is the name of a class of service, and Linux is a proper name little different from Windows or Ford. And in most cases, such precision is not needed; simply saying "transfer your ASPX files to the web server" will work just as well, if you have established that the web server must be running IIS. – user Dec 19 '12 at 9:32
  • Well, this isn't really the point of the question so I don't want to get into a debate over a flawed example. I think my point stands. ASPX is an abbeviation for "Active Server Pages eXtended". The fact that this abbreviation is used as a file extension doesn't make it any less an abbreviation. Linux and cron ... okay. I drifted from "abberviation" to "technical terms". (Linux is a compaction of "LINUs Torvalds" + "UniX", so in a sense it's an abbreviation. And I always thought of "cron" as an abberviation of "chronological". :-) I'm not sure what the point of your last sentence is. ... – Jay Dec 20 '12 at 21:19
  • ... The fact that in a particular case details could be left out of a sentence and thus the problem avoided doesn't really address the problem in the general case. Whether in this example it's important to say that the files are sent with FTP depends on the context. Ditto whether the server is IIS: this would be important if you have multiple servers with different engines and unnecessary if you have only one. As written the point of the sentence would appear to be that uploading the files must be done before running the cron, so I don't see how you could just leave that out. Whatever. – Jay Dec 20 '12 at 21:23
  • My point is that in many cases, sentences don't need to be run-on lists of abbreviations. Sure, there might very well be the occasional case where it helps readability, clarity and understanding to do it that way, but if you are using abbreviations that way a lot (and this is irrespective of your specific example), then you should consider whether it is really necessary, or if the text can be rewritten in some way that doesn't require use of those terms all the time. Few sentences must stand completely on their own; there is almost always plenty of context. – user Dec 20 '12 at 21:33

I was reading a history book recently that demonstrated this very problem - it was littered with abbreviations that were only explained in the first instance they were used. I constantly had to keep going backwards and forwards through the book to remind myself what they were.

The author really should (in my opinion) have put a complete glossary of abbreviations in the back of the book for ease of reference.

If a book uses a lot of unfamiliar abbreviations a glossary is the way to go.


Stanislaw Lem had a very nice method for that. Read his "Observation on the Spot" for it, although I'm not sure if translation captures the spirit.

In essence, the acronyms compose into meaningful, half-meaningful, humorous, horribly misspelled, rudely suggestive and otherwise very memorable words.

So when we're dealing with SuperChronistic Mechanics (SuC-Me) we have to take a wholly different approach than before in the case of Temporally InterTransmissive Area Oligarchs (TIT/AreOli). Those in favour of SuC-Me in fact frequently disagree with the conclusions reached by applying TIT mechanisms to the same data set...

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.