How to Write a Journal Article (in 6 Steps): Step 5 — The Introduction & Argument

It might seem a little perverse to wait until the penultimate step to discuss introductions and arguments, but like I mentioned in step 2, it’s often helpful to work through most of the other material in an article manuscript first, so that when you get to the introduction and argument you have a clear sense of what your argument actually is. So, if you’ve worked through steps 1-4, getting down to the brass tacks of your argument should be a lot easier than if you try and start with the argument — which can be daunting and freeze many writers in their tracks…

intro(In case you want to know what my chalkboard writing looks like, here’s the diagram from my alternative spring break.)

Often, in cultural anthropology at least, articles begin with a little ethnographic vignette — some kind of hook to get the reader’s interest piqued. As I mentioned in step 4, these vignettes are usually linked to the cases that make up the evidence in the rest of the article, and tend to be about a page or so long.

A successful introduction — ethnographic vignette or not — should give your reader a sense of why she or he should keep reading within a page or two, and a good rationale to keep reading is a question or quandary that’s of broad appeal, but which you can provide the answer for. So, it might be an ethnographic vignette, or some other compelling set of data posed in a way that begs questions rather than provides answers, or it might be a question about the existing scholarly literature, or a case from popular media or current events (although I tend to think that current events don’t stay current for long…). Regardless of what your introduction is comprised of, it needs to be returned to later in the article — and where you return to it should make sense based on what the introduction is (i.e. if it’s an ethnographic vignette, continue the vignette in one of your cases, if it’s a problem in the literature, return to it in the literature review, etc.).

After you have your reader hooked, what will actually keep the reader reading is a well articulated argument. One of the most critical things is to not bury the lead — which is to say, make sure your argument appears by the end of page two of your article manuscript (which means it will be on page one of your published article). If your argument comes too late, you risk losing your reader, or, at the least, having your reader begin to wonder why they’re reading what they’re reading. A good argument, posed early on, will do a lot of work for you and ensure that your reader keeps reading — so don’t get carried away with your introductory hook.

What makes a good argumentative thesis? That’s probably the hardest part of this whole project, and it really takes time and experience to develop a solid thesis that you can substantiate with your evidence. I can’t tell you specifically what will make a good thesis for you, but I can give you a few guidelines, some of which are going to seem like no-brainers:

First, a good thesis is motivated by your evidence. This might sound totally crazy, but one of the biggest mistakes I see in articles I peer review is that the thesis makes evidentiary claims that aren’t supported by the evidential cases in the article. My sense is that this is due to people writing articles linearly, starting with the introduction and thesis and then moving onward, and failing to bring the cases in line with the thesis. But if you tend to your literature review and cases before turning to your thesis, you should have a clear sense of what your contribution to the literature in your field is and how your evidence relates to it. One way you might start here is by writing a sentence like ‘Based on [Case 1] and [Case 2], we see that X is actually X1,’ where X is a particular theoretical concept or assumption about a region and X1 is your claim about the same. Once you have that clunky sentence in place, you can work on revising it into something a little more eloquent…

Secondly, a good thesis poses causality. I’m using ‘causality’ here in the broadest sense of the term, mostly because I need some kind of shorthand for all of the kinds of interpretive work that you might do and need to embed in a thesis. In fact, you might be talking about causality (‘X is now X1, because of Y’), but it might also be a little more subtle than that (‘Attending to Y shows that X is actually X1.). Try a sentence like one of those, where Y is what you’re focusing on in your cases. If your two cases are different from one another, then you probably have two sentences here to explain the nuances that each of the cases adds to your claims.

Third, a good thesis engages with questions in the existing literature. Since you’ll be tackling the literature that’s relevant in your literature review — which is coming up very quickly — you don’t need to get into fine details here, and you can often get away with theoretical shorthand, i.e. you can just use keywords, as long as you return to them in your literature review. The right keywords will often motivate a reader long enough to get them through the introduction — as long as you sincerely engage with them as theoretical concepts and make it apparent to your reader how the terms relate to your argument.

It’s often difficult to capture all of these qualities in one sentence (and unwise). It’s best to break them up into sentences in their own right, or at least to start that way and work towards integration. In the end, you should have a paragraph — and it might be a short one — that brings all of these concerns together and gives your reader a sense of the stakes of what you’re focusing on and a clear sense of why she or he should give you the next 30 minutes of her or his time. The weaker or less articulated the thesis, the less likely your reader is going to stay motivated for the whole article…

After your thesis paragraph, you should take the time to detail your methodology in a paragraph. This paragraph can be difficult to write — mostly because it’s difficult to be excited about it — but once you have a good methods paragraph, you can copy and paste it with minor variations for every article you write thereafter. A good methods paragraph lays out the duration of the research, where the research was conducted and what those contexts were like, and what the sample size was (which, for cultural anthropologists, is the number of people interviewed, events attended, etc. — the stuff that makes up your cases). This paragraph might also legitimate methodological choices in reference to key theoretical-methods literature, particularly if the methods are unconventional or experimental for the audience of the journal.

After your methods paragraph, you should turn to your literature review. By now, you should have a pretty clear sense of why you’re discussing the literature that’s in your literature review, so working through it again to revise it is worth the time. Given your argument, tighten up your literature review, both cutting some literature that’s no longer relevant to your argument, and also making sure that you’re highlighting what you should be about the literature that you’re keeping.

The last paragraph in your introduction should be a transitional paragraph that lays out the content of the article in relation to your thesis. So, often what you’re doing is moving from the fairly broad claims in your literature review to the specific content of your cases and surveying them in a sentence for each case. But this is all preceded by a recapitulation of your thesis in a shortened form. ‘As I will demonstrate in the following, X is due to Y resulting in X1. This can be seen in Case 1, where X is A. Further, in Case 2, X is B.’ That feels a little symbolic-logicky, but hopefully you get the idea. You should also explain what will occur in your conclusion, so there are no surprises, i.e. ‘In the conclusion, I discuss the implications of X as X1 for [subfield or regional interest].’ Basically, you’re once again motivating your reader to carry on reading the article, and also ensuring that there are no substantial surprises — your reader needs to be able to anticipate everything that’s coming up (without all of the nitty gritty details). Anticipation is motivation.

All told, your introduction should be about 5 pages long (maybe up to 7, but rarely any longer than that). Five pages might seem really short, but a good introduction shouldn’t be too long (it’s an introduction, after all) — anything longer than 7 pages is really going to tax your reader, and she or he will be wondering when they get to the good stuff (outside of your opening couple of paragraphs, your introduction isn’t so much ‘good’ stuff as ‘necessary’ stuff). Strive for brevity, knowing that it will help your reader stay motivated. And, when the peer reviews come back, know that you’ll have a little wiggle room in your introduction to address the concerns of your reviews.

Once you have it all weaved together, adding your new introduction to your cases and conclusion, that’s your article manuscript. Easy, right? Now it’s time for some fine tuning

How to Write a Journal Article (in 6 Steps): Step 4 — The Evidence

So, if you’re like me, you often start working on an article by beginning with the evidence — or, rather, you have a few pieces of evidence that seem like they should be the backbone of an article and you assemble them into some rough order. Evidence in need of an argument, like an army without a general, is often pretty aimless.

No-Dumping

Starting with your evidence can be a bad way to start, largely because it can be hard to gauge what any evidence is evidence of. We’re often drawn to evidence because it’s compelling to us as lone researchers, but it’s compelling primarily because it stands out amid a host of other evidence that we prefer not to think about (because it’s not so compelling). But whether or not the evidence at hand is interesting for others to think with is another story, and depends on establishing why it’s worth a reader’s time — which, hopefully, you’ve begun to flesh out in writing your conclusion and literature review. So, now that you have a sense of the journal you’re targeting and the audience that journal appeals to, as well as the broader debates you’re ready to speak to, it’s time to get into the evidence and make it work for you.

Generally, my rule is that for every 2,000 words in an article, you can use one case study. (I’ve discussed my case study approach to dissertation writing here, which might be worth taking the time to read if you haven’t already.) So, if you’re writing a 6,000 word article manuscript, you have roughly 2,000 words for the introduction, literature review, conclusion and bibliography. Then, another 2,000 words for the first case and another 2,000 words for the second case. For each additional 2,000 words in a word limit, you can add another case. But round down — it’s better to spend more time with a case rather than less (i.e. no 1,000 word cases).

If you’re writing an article based on dissertation content, I suggest writing the cases anew. Don’t copy and paste, but instead just write from memory (and maybe copy and paste transcription or description — the stuff that won’t change). There are a couple of reasons for this: first, there are often significant shifts in tone between dissertation chapters and articles, and it’s very difficult to iron out tone once it’s in the text. Starting fresh ensures that you’re carrying forward the same tone that you’ve established in your conclusion and literature review. Second, it can often be difficult to gauge what you absolutely need to cut from something you’ve already written (and people often have a hard time letting go); it’s easier to ascertain what you need to add. So, if you’re writing for an audience that’s in your same subfield or region, there are a lot of assumptions that you can make of your audience — if it’s a subfield journal, you don’t need to spend time explaining generic institutional kinds (e.g. you don’t need to explain what a hospital is to a medical anthropologist), and, likewise, you don’t need to spend time on well known context for a regional journal (e.g. while the history of Civil Rights is important to understand racial discrimination in the U.S., you don’t need to explain it to most Americanists). So, start fresh and write the cases up again — they should be looser than their dissertation versions, and easier to mold into your argumentative needs.

Because you’ve already worked through your literature review, you should have a pretty clear sense of what your work is contributing to in the field. But, if not, writing up your cases can be a way of clarifying what you have to add. Often, whether we know it or not, we find evidence compelling because it either runs counter to what we’ve been taught to think about a subject (or at least troubles assumptions) or it’s significantly different than the rest of our collected evidence. If it’s the former, it’s pretty easy to see how to build an argument with it: you’re challenging dominant ways of understanding a theoretical concept through a collection of contrary evidence, and you need to present that evidence in such a way that it logically unsettles disciplinary assumptions. If, however, it’s the latterer and it’s mostly compelling because it’s different than most of your other evidence, you have a different battle to fight. First, you need to provide your reader with a baseline (i.e. the boring stuff) and then turn to the exceptional stuff. So, with that structure, case #1 might be the boring stuff, and case #2 the exciting stuff.

Often, cases begin with some kind of hook — some juicy interview segment, or rich description of an event, person or place. From that hook, cases zoom out to explain the hook’s context, often situating it in its contemporary moment, and then providing some deeper context (history of a place, institution, event, individual) to understand why it’s worth thinking through. After that context is provided, further evidence of the same type (or slight variations of the same kind of evidence) are provided to flesh it out, often with the implicit attempt being to naturalize it. That is, what makes a hook a hook is that it’s not what your audience should expect. But by working through your evidence, they should come to see that the hook is actually suitably normative in your research context. Or, if it really is exceptional, then you work to establish how it’s exceptional by showing how all the other evidence you’ve collected is different from it. This might mean including other evidence without as much context as your first hook depends on, but that can often be okay because you’re providing this other evidence as further context to understand the hook and its merits.

For cultural anthropologists (and, really, lots of people in different humanistic and social scientific disciplines), it’s typical to begin an article with an anecdote, and then return to that anecdote in your first case. If this is the approach you’re adopting (and it’s largely determined by the journal you’re writing for), then you might take the first couple of pages of your first case and cleave them off. Then, rewrite the beginning of the case knowing that the first two pages are going to appear at the start of the paper. This may mean providing a brief summary of the hook that appeared in the introduction, or some other flag that reminds your reader what they’ve read; from there, you need to move into further evidence to flesh the hook out.

One of the persistent problems I see in articles that I’m asked to review for journals is that authors assume that their evidence is self-explanatory. Often, sections of articles dedicated to evidence read like infodumps: pages of unanalyzed and undermotivated evidence, which leave me with my eyes glazed over. Why, I ask myself, am I reading all of this evidence? And what in the world is it evidence of? My general rule is that for every line of evidence, there should be two lines of analysis. So, if you have a quote from someone or a text that runs four lines, you should spend at least eight lines discussing its relevance to your argument. This might include some historical or situational context, but it should primarily be based on your interpretive scheme and further your argument. This might seem like a no-brainer, but one of the problems a lot of people have is that they spend so much time with their evidence (in their dissertation, in conversation with others, in conference presentations) that they just assume everyone understands its implications and relevance to your argument. Let me assure you: they do not. Be sure to spell it all out, and make sure that every paragraph begins and ends with some reference to your larger argument, if only a glancing mention of keywords.

Which provides me with a segue to step five, developing your argument and the introduction of the article.