Footnote folly

One disadvantage of being an editor is the habit of reading pedantically. When most people come across misused words, grammatical mistakes or erratic punctuation they use their natural ability to infer meaning from the jumble (try reading the transcript of a conversation you understood perfectly well to see how good you are at finding order amidst chaos). But when editors come across the same problems they tend to fixate on the errors instead of what the author is actually saying.

This happened to me on Tuesday when I was reading the High Court’s industrial relations judgment. I was continually distracted by the wrong placement of footnotes. Take these two not untypical sentences:

The constitutional underpinning of the legislation was noted, but not questioned[8]. McHugh J said[9] that “[t]he corporations power provides a broader basis upon which s 170LI may operate”.

Which should have been written:

The constitutional underpinning of the legislation was noted, but not questioned.[8] McHugh J said that “[t]he corporations power provides a broader basis upon which s 170LI may operate”.[9]

With rare exceptions, such as with dashes or where ambiguity might otherwise be created, footnotes go after punctuation. Putting them in the wrong place is surprisingly common. Academics seem nearly as likely to think that footnotes go before punctuation as teenagers are to think that apostrophes are needed to create a plural (or apostrophe’s, for victims of the school system over the last decade or so). Most mistakes are made by people whose disciplines use the Harvard author-date system of referencing. In that system the author’s name and the date of publication go inside the punctuation (like this). But clearly even where footnoting is still widely used, such as in legal publications, some people still have the wrong idea.

Admittedly, the mistake rarely makes the meaning hard to discern. Its damage is to the process of reading. Good writers are often described as fluent, or as writing smoothly. There is nothing in their prose that trips us up. The [9] in “McHugh J said[9] that “[t]he corporations power provides a broader basis upon which s 170LI may operate” momentarily stalls the flow of words and makes reading a little more difficult than it needs to be. All referencing does this to an extent, but putting it after the punctuation takes advantage of a pause that would be there in any case.

The author-date system, to me, is worse when correct than footnotes or endnotes are when incorrect. It disrupts the flow of a sentence not just with a number but with names, dates, page numbers and brackets. Clunk. Clunk. Clunk.

The author-date system did once have a rationale. Older readers may remember the days when, while writing essays by hand or on a typewriter, careful judgments had to be made as to how much space to leave at the bottom of a page for footnotes. It was much easier to insert a (Smith 1975) or Smith (1975, p.46) and put the full reference at the end. When word processing came along, the footnote space problem was solved. Yet the author-date survived. One reason is surely that academic journals, unlike most other journals, don’t exist for readers. They exist for writers to get DEST publication points, removing the incentive to improve readability. Even so, there is no excuse for it. If referencing has to be inserted into the text, footnotes and endnotes are the way to do it.

41 thoughts on “Footnote folly

  1. Nice piece. Economists tend to use the Harvard system. When I write for the CIS I use endnotes (Andrew is the gate keeper). I have to say though I have a preference for footnotes and not endnotes – the footnote lets you glance down and skim the material. If it’s not important or interesting you can move on. Endnotes, however, break concentration – turn to back find note. Some books have them at the end of chapters, others at the end of the book. The latter are often poorly labelled adding to fraustration


  2. The main case for endnotes is that few people read them anyway, so the clutter of footnotes is moved up the back where people can easily skip over them. In the case of Policy, there is the added factor of space constraints. We often put endnotes online only, to create more space for text in the print version.


  3. There is a need to differentiate between citation notes and explanatory notes. My preference would be for citaions to go in footnotes and explanations to go in endnotes. However, within the citation footnotes, I would prefer to see the Harvard style of referencing used, with a complete reference list at the end of the paper. The use of latin terms in citation footnotes is a major distraction. Having slightly off topic explanations and the like at the end is much less distracting for the general reader. In general, explanations that are not incorporated in the main text should probably be kept to a minimum. Having them appear as endnotes encourages this.


  4. As an aside, some good advice that I have both read and been told about explanatory notes is that, if something is worth saying, it is probably worth saying in the main text. As such, explanatory notes should be kept to a minimum. However, I suspect that I don not follow this good advice as often as I should!!!


  5. Andrew, I wonder whether this problem may have arisen when the judgment was put onto Austlii (whatever conversion process it goes through). I don’t recall, off the top of my head, ever noticing errors in the official CLRs.

    In relation to the [9], is its placement technically wrong or merely undesirable? The Melbourne University published guide to legal citation states “The footnote number (or other identifier) should follow the portion of text to which the footnote refers. Ordinarily, the footnote number should appear after any relevant punctuation.”


  6. Without having any empirical evidence on this, I suspect that the vast majority of articles published in the vast majority of academic journals are written by academics at non-Australian universities. As such, I doubt very much that DEST points are a significant factor of the formatting of those articles.


  7. James: depends on what’s in the footnote, but that particular one is a reference to the place in which she said “..”. Desirably and technically, the footnote goes after the direct quote.

    Yes, I looked it up to check punctuation. I admit I’m sick. I actually had to request a pen to change a sign at the cafeteria yesterday. In my defence, it said that they had “glutton free” pasta available. The only thing necessary for awful writing to triumph is for pedants to do nothing.


  8. It may not be the Justice’s fault. As someone who spent a year wrestling with it, I can attest that the High Court style guide does not always accord with the Chicago manual of style.


  9. ‘The main case for endnotes is that few people read them anyway, so the clutter of footnotes is moved up the back where people can easily skip over them.’

    I don’t doubt this; yet it reflects poorly on both readers and writers. It also depends of the book and type of reader. In my uni days, we had a lecturer who often said ‘it’s not enough to know the law, you have to know where to find the law; the law is in the footnotes’. Hayek has lots of endnotes – often with substantial comments, yet they difficult to locate.

    Damien’s points is very good. Dierdre McCloskey has made that very point. If it’s important put in the text. But this is a style thing. My PhD supervisor believed that academic writing had lots of footnotes, and so my thesis and everything I wrote with him had lots of footnotes.


  10. Sinclair is right. Footnotes are a courtesy to the reader, balancing convenient information availability, and readability in the body text. Down with brackets!
    Damien, a system that had footnotes and endnotes co-existing would be the worst of both worlds. What’s wrong with discursive footnotes where they’re needed?
    I don’t know about the law, but in history the trend is being set by publishers, who don’t want to spend the typsetting time or the page-space on footnotes, and are more and more publishing with endnotes, or worse, nothing at all.


  11. James, Andrew – I did carefully avoid blaming the judges. The government style manual, contradicting the Chicago, does say put the number before the non-full stop punctuation. But it is quite clear that it goes at the end of a sentence or clause, so the examples quoted above are incorrect on both the American and Australian authorities. I think the Chicago version is preferable.



    Damien said: “Some good advice that I have both read and been told about explanatory notes is that, if something is worth saying, it is probably worth saying in the main text.”

    I have heard that advice too, but I would say that it is ‘simplistic’ rather than ‘good’. Material that is important/vital/of interest to one reader may be less important/vital/of interest to others. Footnotes, like boxes and appendixes, provide a mechanism for “layering” a document so as to meet the different needs of a diverse readership.

    Imagine, for example, that you are writing for an audience that includes some non-economists as well as economists. If you explain all your economics terms in the body of the text, you will bore and waste the time of the economists. However, if you do not explain your technical terms somewhere in the text, such as in footnotes, you will lose the non-economists.

    Another use of footnotes is to park proofs of otherwise unsubstantiated statements. Effectively this says to the reader: “if you believe this statement, read on, but if you need to be convinced of its veracity, take the time to read the footnote”.


  13. The Australian Institute of Judicial Administration publishes a “Guide to Uniform Production of Judgments”, which is replete with errors (apparently the one I have is the second “addition”). The style guide for references to books and journals is internally inconsistent. It encourages inline citation of cases, which can turn a paragraph into a dog’s breakfast.

    While I was at university, we used a modified version (to remove latinisms like ibid and op cit) of the Australian Guide to Legal Citation. The first thing it addresses is footnotes, and it’s pretty clear on where the number belongs — after the text to which it refers, and after any punctuation.


  14. “In relation to the [9], is its placement technically wrong or merely undesirable? The Melbourne University published guide to legal citation states …”

    “James: depends on what


  15. [Laura, Andrew’s blog thinks I am you now!!!]

    Sinclair, it was Deirdre McCloskey’s book in which I read that advice!!! It basically said footnotes are nests for pedants. I just looked it up then and that is actually the title of the relevant chapter (chapter 16, pp. 48-49). My PhD advisor also told me that some good advice about footnotes was that, if something ios worth saying, it is worth saying in the main text. I suspect that the reason I sometimes have trouble following this advice is that I am, at heart, a pedant!!!

    For anyone who is interested, the McCloskey reference is:

    McCloskey, DN (2000), Economical witing (second edition), Waveland Press, USA.


  16. DD – There is nothing inherently wrong with double negatives, if that is your complaint. In this case, ‘typical’ would not be my first choice, since many sentences had no footnotes. ‘Not untypical’ avoids that unintended meaning, while making clear that I am not just picking on two sentences that happened to have mistakes with their footnotes among others that were correct.


  17. Look, Dierdre McCloskey might be right that footnotes are nests for pedants. But then saying you shouldn’t use them is discrimination against pedants. For crying out loud, if you can’t be a pedant in academia or law, where can you be a pedant? I have always thought that if more than half the page is taken up with footnotes that it was a bit excessive though, and in my (limited but painful) experience you only tend to see that in legal articles.

    I’m guessing DD wants it to be ‘atypical’?


  18. Andrew, here is the voice of another editor, but also an academic (in a technical field).

    First of all, I sympathise with your initial point. I have developed a bad habit of not being able to get the meaning of badly written text, focusing too much on style rather then substance.

    However as a reader I totally disagree with your rejection of (author, year) referencing system. When I read, I can instally see what and who is cited, and that is great. In the number system I have no idea what [84] means, and thus have to look it up in the endnotes (footnotes are very expensive for the typesetter and therefore are not allowed in our journal).

    Mind you, with the current word processing software such as LaTeX it is as easy if not easier to create numbered references than (author, year) ones. So your cynicism is misplaced.


  19. Andrew, I’m happy with untypical in that case too.
    Sinclair, yes, of course. I was being just a tad facetious. My tendency to overuse parenthetical comments and to write sentences lasting a page is something I’m still struggling to overcome. That said, writing style does depend a bit on the intended audience; presumably lawyers are used to dealing with papers in which the footnotes take up more space than the text.


  20. My impression is that in mathematics there is no one style – it probably depends on the flow of the argument and attributing work to the correct authors. Writing nice mathematical argument is difficult – it takes a lot of work for it to flow nicely and understandably – students often underestimate the effort it takes to write mathematical argument well! (and I’m not saying that I’m a great expositor of mathematics at all!)


  21. Often in the mathematical and physics literature, a reference such as [BZ] is made rather than [84], and the bibliography will include something like

    [BZ06] Blumen, S. C. and Zhang, R. B. “A complete theory of quantum gravity.” 2006

    which can give you an idea of the authors and the specific paper (eg, the one published in 2006). It’s more informative than the numerical reference, but can look a little more ugly.


  22. Laura, I understand what you are saying. I think this is a specific problem of student essays and not of published studies. I have noticed this in my daughters essays.

    Having said that, if you think about it, the style of citation is actually irrelevant for your problem. I can say “Simpson (1966) said”, or I can say “Simpson [3] said” even with numerical style if I so desire.

    This point is most obvious in TeX and LaTeX where the style of referencing is determined by a single command at the begginning of the text. You can change the style of references throughout the text by changing one command. Thus the style of referencing does not affect the writer at all, only the reader – contrary to the points made by both Andrew and Laura.

    Of course few humanities people use TeX. But they probably should, as it encourages more organised writing. BTW one of the well known distributors of TeX is (or was) a professor of classical literature at Washington State University.


  23. I agree that LaTeX documents look beatiful. It is particularly good for work that contains equations, but I think it formats text nicely as well. I think it would be a good package even for english or history work. Its on my list of things to learn how to use. I have heard that MiKTeX is a particularly good implementation of LaTeX and that WinEdt is a particularly good text editor to use in conjunction with MiKTeX.


  24. LaTeX is freeware – and there is a wealth of material on it on the net. About its only downside is that you need a graphics program to draw good graphics and import the files into the latex file.


  25. The other downsides of LaTeX, are (1) the fact that you do not see the result of the formatting on the screen as you type (even though you can run a quick command to get it in a few seconds) and (2) you need to remember quite a few commands like in a programming language. I used to live with both of these difficulties, but as I grew older and the mind grew more lazy, I have got a commercial implementation which solves (2) completely, and solves (1) to a large extent. However (2) is mostly relevant to math equations.


  26. “LaTeX is freeware – and there is a wealth of material on it on the net.”

    This is another problem with LaTeX – so much info, you don’t know where to start.


  27. Agree Boris.

    Also, it’s hard for beginners to know where to start – like a lot of IT stuff, it’s hard to get into to and a lot of the info is designed for people who already who a fair bit about it.


  28. I think what Derrida Derider has in mind, above, is this quote from Orwell: “One can cure oneself of the not un- formation by memorizing this sentence: A not unblack dog was chasing a not unsmall rabbit across a not ungreen field.” Certainly this sort of double negative is not ungrammatical, but I think it is not unreasonable to say that either the writer should remove “not un-“, remove the entire not unsuperfluous phrase (“not un-X”), or some up with a not unoriginal way of making the point.

    The insight that Orwell has on this *specific* usage habit: the *real* reason people say “not un-” isn’t because, after careful thought, they decide it is the best way to express a concrete idea, but rather because “not un-” sounds smooth, rolls off the mind as easily as the most dead metaphors, and gives a false impression of reasonable moderation that insulates a writer from having to understand what he means.

    Against Damien, I would prefer my *footnotes* to be discursive, and the endnotes to be citations. I think the academic fetish with correctly and exhaustively sourcing one’s work is a bit overdone, and I almost never look at the sources until I am ready to sit down and look up all of the sources for a particular set of claims. (Obviously this depends on one’s field; in science, with more concise papers, and law, with all the precedents, the line between the purpose served by a discursive note (an addendum to the argument that will be critical to the concerns of some people, but less so to others) and by a citation (substantiating or sources ones factual claims) becomes blurry.)


  29. Asdas – Though we could equally say – as Orwell did – that rigid over-application of generally good sytlistic advice can also be harmful, losing as in this case a shade of meaning, or disrupting the flow of the sentence, as unsplitting infinitives can often do.


  30. I don’t find that the author date system breaks up the flow of the text at all. In fact, I find it less disruptive than constant ibids and loc cits. If I were to rationalize my preferences, I would note that constant footnotes require one to keep glancing down to the bottom of the page. But it may well be that it’s just a question of what system I use most.


  31. I personally find some endnotes invaluable. Example, Peter Watts’s latest book, Blindsight, has endnotes giving the science on which his fiction draws. I’m hoping he will eventually put the endnotes up on his website as working links to articles, where copyright allows. Endnotes to me are very, very primitive hypertext, the sort of connection to ideas that gets bigger the farther you go into it.

    Aside —
    I notice the forum software doesn’t use “that” to introduce comments, which reminds me to ask, wouldn’t the example given:

    McHugh J said that


Leave a Reply

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

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

Facebook photo

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

Connecting to %s