How to read a research article

At the University of Essex, where I teach, we have introduced a new assignment to help our second years with their preparations for choosing and researching a topic for their undergraduate dissertation. It involves each of them taking a research article and writing a short report on how the author would have gone about defining their topic and building up the evidence on which it is based. We appreciate that this could be a challenging task and so, in suggesting a couple of articles they could study from my reading lists, I have included one of my own articles with the offer of explaining how I came about writing the piece. I now realise that was rash: if the task of writing a report is challenging for them, it is all the more difficult for an author to think back to where they were intellectually a few years ago: the moment has passed and none of us, I imagine, would write now precisely how we did then.
What I can do which should be of some help is to give some hints of what an historian looks for when analysing a research article. So, here are some tips, meant to be of more general use, but which grow out of some reflections on an article entitled ‘Humanism across Europe: the structures of contacts’ which was published in 2012.

Not all research articles are equal. There is no one formula for a ‘research article’ – the term covers a range of styles of writing, and appear in a variety of publications. Some discuss the results of an investigation in a single archive, usually drawing out conclusions at the end. Others are more argument-led, with the research likely to be more eclectic. This is not simply a binary opposition: it is more like a spectrum, and my article falls closer to the ‘argument’ end than the ‘research’ end. As we shall see, however, no article can omit either of those two elements. You might well ask how you judge the article’s place on the spectrum – the basic answer is to pay attention to its content and how it is constructed but there is also another element which sometimes can help.

Be aware of the context. An article never appears on its own; it always appears in a larger publication. The classic format is for a research article to appear free-standing in a learned journal – in that case, it is unlikely for there to be dialogue between it and others in the same issue – but some journals increasingly produce ‘special issues’ where a set of essays discuss a related theme. This is similar to the collection of essays, like that in which my article appeared. In such a case, the author of the article you are reading may well be aware of what the other contributors have written and be engaging with them, even if it is by consciously bringing a different perspective. Look at where my article appeared and notice who was the editor of the volume: as it was the same person as the author, you can be sure that the chapter was written in full knowledge of the other contributions. You might even conclude that its purpose was to provide an overarching interpretation, placing the chapters that preceded it within a wider argument.

Check the date. Part of the issue of context is to understand when a work was written. We know that time’s arrow travels in only one direction: an article cannot be influenced by scholarship that appeared after it – and even the most assiduous academic may not be fully up-to-date with what has been published. So, the date of publication gives something like the cut-off point (the terminus post quem non) of possible influence – but it is only approximate. How long did it take the work to reach print? Was it written as a conference paper and the revised for publication, pushing the date of original composition back further? Are you using the original version or a re-print? Make sure you understand this element of the context: in a collection of essays, the preface will help; in an article, there is often a first footnote, listing acknowledgements but also, sometimes, explaining the origins of the work. Paying attention to these details will help you appreciate where the work sits in the historiography – that is, at what point it inserted itself in existing debates. That takes us to the next and perhaps most important point.

Find the debate. Historians are argumentative (and if one of them denies that this is the case, they have proven my point). Historians thrive by placing themselves in a debate, by detecting or creating disagreement so that they can then construct a new perspective. Any article, then, is more than its research: it forwards a position. How an article does this differs. Some might not even seem to be intent on making an argument, and some might be written as if what it has to say is not open to debate. In those instances, it is your essential task to detect how it fits into the historiographical scene and how it can be used to enlighten the big issues of scholarship: this creates the possibility that you, as an historian yourself, can use an article in ways the author might not have expected. But, as historians prize argument, it is unusual not to have the position being taken placed in the foreground. Admittedly, it is rare nowadays to find an academic in our discipline launching a full-frontal attack on another (‘Prof. X shows a shocking inability to do justice to the evidence…’) – and some would say we are better for that, even if it makes scholarship less entertaining for the readers. Some might even not mention an opponent by name in the text – my own style is often to talk about ‘the standard narrative’ or to remark ‘it has been argued…’. When that happens, how do find out who they have in their sights? The solution is to heed the next piece of advice.

Follow the footnotes. The text of an article and its footnotes relate to each other like the two hands on the piano. To read an article without paying attention to the notes is like hearing a melody without its base line – or, rather, an argument without its basis in evidence. The footnotes reveal what the text might hide: they reveal the author’s debts and disagreements – but they do more than that. They do not only help you more fully understand where the historian sits in a debate; they also make explicit the evidence – the research itself – on which the article and its argument is based. This is why footnotes matter so much to historians and why we disparage other disciplines which think that citations can be confined to a brief bracket at the end of a sentence: historians live by the demonstration of sound research their footnotes display. Or, sometimes, they fall because of their poor quality; more than one scholarly reputation has been damaged by the revelation that the author’s footnotes are inaccurate or, worse, borrowed – a kind way of saying plagiarised – from others. In my own area of study, footnotes most often cite unprinted primary sources (and so list a manuscript or document by its present location), though in the article I have set my references are more often to published editions.

Evaluate the evidence. By reading both the text and the footnotes, you can build up a fuller sense of the research that lies behind the article – and you can also begin to judge the strength of its argument. You will want to ask yourself what types of evidence are used and what range. Some articles will be tightly focussed on one set of materials, others will range more widely – but always with a sense of coherent whole to them. In my article, you will find several different types of primary material used, as well as what the social sciences call ‘secondary research’; that is, some statistics based on data-sets compiled by earlier scholars. As you think about how the author developed their article, you will also want to ask yourself: what is the most significant evidence used here? Remember that the way an article is written is very rarely an unfolding of the story of how the research was conducted. It is constructed after the research was completed and will order the evidence in a manner that makes it as convincing as possible, with the result that the most significant material is unlikely to be at the very beginning (or very end) of the article. So, what you are doing is understanding how the results of the historian’s time in the archives has been organised and deployed in order to make their argument persuasive.

Sense the sections. The consequence of this is also that an article is – like an essay you write should be – a set of steps. The author needs to lead the reader through the stages of the argument as the evidence is unveiled. Sometimes (but only in a minority of cases), these steps will helpfully be flagged up by sub-divisions in an article: these might involve sub-titles or numbers or (as in my article) simply an extra line-break. By being aware of these, though, you can get a better sense of how the author has arranged the information they have gathered. Of course, it is up to you as reader to consider whether the evidence could be organised differently or supplemented by other material, with changes to the argument as a result. When you do that, you move from being a reader to being a potential author, ready to do your own research and enter the sand-pit of historiographical debate.

Good luck.

Lord, save us from author date systems

If, as some would have it, academic disciplines are tribes, among our self-defining rituals — our handshakes or our body-art — is the manner of citations. We historians are guardians of the footnote. Publishers may force us at times to soil our handiwork with endnotes, but for us the gold standard of referencing is the full information of the work being cited — author, title, publication details, precise page — placed at the foot of the page in a smaller font than the text, and with a direction to the note placed surreptitiously in the body of our prose as a superscript number. Our tribe sub-divides: some prefer to gather their references together and place a single footnote at the end of a paragraph, a practice than can work but can also be a charlatan’s charter, allowing unfounded assertions to be smuggled into the argument under cover of the other citations. Another clan takes footnote numbers for stardust and sprinkles them across the page, accompanying most nouns, and some adjectives, with a superscript. Me, I belong to those who prefer one footnote per sentence, placed after the full-stop as part of the punctuation. No breaking the flow of the sentence but a gathering of sources close to where they have been mentioned.

Others in our family of tribes — the humanities — share this citation system but not those who see themselves as scientific. I have an archaeologist friend who mocks our manners, considering an ‘author date system’ preferable. To his way of seeing, the presence of footnotes breaks up the page. We, in our turn, find the practice of citing by surname and year, within the text, referring the reader to the bibliography for full details quite other — by which we really mean lesser. It is inelegant, interrupting the prose with phrases like Andrews and van Dyke 1964 in a way that we find much more intrusive than a slender number slid into a sentence-break. It can also, depending on the particular system (for these, too, are a series of peoples, not one tribe alone), be imprecise, failing to provide the page reference to pinpoint where the specific statement was made. It is also liable to error — I have found occasions when the work cited in the text has not been entered in the bibliography.

But I have been conversing recently with those who speak this language, as I have been reading some works in the discipline of educational studies. I might grace this with the grand title of ‘research’ but, in truth, I am only looking at recent writings (as opposed to the historians’ holy of holies, the primary source) and those on the internet. I have much to learn but I am also seeing practices that I find concerning: not in major publications themselves but in how the style of scholarship is being mimicked — or pastiched — elsewhere. I am not going to point fingers — it is a practice, not an individual instance that I am highlighting — so let me talk of an example that we can call Anon. 2014. It is a page on a British university website, discussing a topical issue in educational studies. It provides a paragraph of texts which bristles with author-dates. It is followed, on a separate line, by a hyphen followed by a link to a document on another university’s website. Click on that and you find that the paragraph you have just read is a direct quotation.

In a culture where plagiarism has become one of the greatest academic sins, this practice sails close to a cold wind: no quotation marks to signify it is not original statement, no explanatory text, only a slender hyphen to carry the force of ‘the above paragraph is quoted from…’. Perhaps I will be told this reveals my tribally-based way of seeing; perhaps other disciplines have a yet more subtle use of the minutiae of punctuation than history does. It is not, in fact, this to which I want to draw attention.

It is, rather, the manner in which Anon. 2014 replicates text complete with citations by author and date but does not give on the page a bibliography to explain those references. Let me repeat: this is only an example. In my recent travels, my little boat surfing the internet (and sometimes crashing), I have found several instances where webpages or uploaded presentations include only one half of an author date system: the citation without the concomitant bibliography. This strips the architecture of the citation system of its essential foundations, leaving the short references unbuttressed with full information.

But, you clamour, consider the medium: these are webpages or PowerPoints, and for those the standards are different — after all, does bonae litterae include footnotes? The answer, of course, is no — this site, unlike some impressive medievalist blogs, consciously does not provide citations. There are two reasons for that. The first is that the web allows an even more elegant solution of embedded links which allows the reader to go straight to the source at the point the reference is made — it is not a fully scholarly solution, as it can suffer the imprecision of some author date systems, failing to point the specific page, but it can serve the purpose of this site. And that is the second reason: while there are nuggets of new information hidden herein for the truffle-hunting scholar to find, this is not a work of research; they are musings, first drafts, not polished learned prose. They pretend to nothing more — and that is the issue with Anon. 2014 et al. There is a pretension to scholarship that, even when it is not borrowed plumage, is inappropriate. It acts on the rituals of scientific scholarship without understanding their full import. Worse than that, it excludes: by providing citations in code but not explaining that code, it seems to be intended not just to impress but to impose upon the reader. It is not just a ritual, it is an academic haka.

To put it another way, the abuse of the author date system creates a sort of élitism. I do not mean this in the general sense that is in common parlance, where any élite is considered élitist and thus suspect, a usage which leaves begging the question of what we call the best in a meritocracy. I use the term, rather, in its narrower sense of the construction of a select group on a basis other than merit or ability. This is surely what happens when knowledge is presented in a manner which is intended to be impenetrable or unquestionable, the antithesis to our understanding of liberal enquiry. The likes of Anon. 2014 are not simply pretending to knowledge but also fraudulently claiming membership of an élite, in the process demeaning that élite, turning it into a clique.

I am well aware that in the anti-pantheon of sinners, élitists sit only a few circles above the well of plagiarists in the new academic inferno. I also appreciate that the practices I have described have not been created by bad men bent on a desire to be exclusive. They reflect a more human failing, into which all of us, whatever our tribe, can fall. To avoid the calamity of our weaknesses, we need our elders to bring us up in practices that keep us from erring. That, and a pious prayer to our tribal gods: Lord, save us.

The battle against endnotes

I was reading a review today which ends:

‘[The author] has also committed the considerable offence, which one would have thought the concentrated venom of reviewers would have killed off long ago, of herding all her notes together at the end of the book.’

The review appeared in History in 1939. Clearly, the tongue of reviewers, however spiteful, spits an ineffective poison.

And let us not be uncharitable: endnotes are not the worst. Even more infuriating are notes gathered at the end of each chapter. Nor are they the most unscholarly: that accolade should surely be reserved for the ‘author / date’ system which not only interrupts the text and often fails to provide enough citation to be usefully specific but also seems to arouse the spirit of Lethe, where the writer, their copy-editor and their publisher all forget to check that the full reference is actually included in the bibliography. Yet both styles seem to be gaining ground on the now old-fashioned footnote (it would be interesting to know what proportion of monographs use each system and how that changes over the years: does anyone have those figures to hand?).

Perhaps, though, we should remember that the footnote itself has not always been viewed with a kindly eye. Hilaire Belloc, in an essay curiously not mentioned in Tony Grafton’s The Footnote: a curious history, roundly rounded on the practice of freighting a page of text with the ballast of small-typed notes. A form of lying, he claimed, with Edward Gibbon accused of being its originator. His venom too was patently proven innocuous.