Good writing in political science

Below is an essay that I wrote for my undergraduate class last semester, providing them with my (doubtless idiosyncratic) ideas about how to write good political science essays. It’s also available under a CC license in “PDF format”:http://www.henryfarrell.net/pol_science_essays.pdf, as well as “MultiMarkdown”:http://www.henryfarrell.net/pol_science_essays.txt (its native format), “LaTeX”:http://www.henryfarrell.net/pol_science_essays.tex and “RTF”:http://www.henryfarrell.net/pol_science_essays.rtf in case someone wants to play around with it (e.g. by adapting it for another discipline). Feel free to suggest improvements, point out grammatical errors or typos etc in comments, or indeed to comment generally on good and lousy writing in undergraduate papers.

UPDATE: some small improvements made and “Jottit version”:http://jottit.com/wbr7p/ added.

[Cross posted to “The Monkey Cage”:http://www.themonkeycage.org ]

Good Writing in Political Science: An Undergraduate Student’s Short Illustrated Primer1

By Henry Farrell

Version 1.01

This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License.

Leo Tolstoy famously observed that “happy families are all alike; every unhappy family is unhappy in its own way.” 2 Tolstoy, happily for all of us, was not a teaching political scientist. Had he been, he might have observed that undergraduate political science papers are subject to a different logic. Really good papers are unique – each has its own particular thesis, style of argumentation, body of empirical evidence and set of conclusions. Really bad papers, in contrast, tend toward a dismal uniformity. They draw on the same evidence (garbled versions of what the professor has presented in class), are organized according to similar principles of incoherence, and all wend their eventual ways towards banal conclusions that strenuously avoid making any claims or positive arguments whatsoever.

This short set of guidelines cannot make you into a really good essayist. For that you need time, practice, and native genius. What it can do, however, is help you avoid some of the most common pitfalls of undergraduate essay writers. You can surely avoid being a very bad essayist, and you can very likely become a better essayist than you are already. What follows is a short set of suggestions, accompanied, where available, with cautionary examples drawn from online essay mills.

Read the Requirements for the Assignment

This suggestion may be taken as insulting because it is so obvious; still, it is commonly ignored in practice. The professor usually drops some very strong hints about what she is looking for when she assigns a piece of writing. It is best to pay attention to those hints. For example, if you are asked to write a term paper on a problem of international cooperation, you should ensure that your term paper explicitly focuses on a topic that (a) involves cooperation, (b) has, at the least, some international aspect, and (c) is potentially problematic.

Sometimes, assignments are ambiguous. Professors too may err. The assignment may be inexactly worded or involve contradictory requirements. In these cases it is obviously best to ask the professor what she is looking for (it is often better to ask via email, to ensure that you have a written explanation of what the professor wants that you can refer to later). Where this is not possible (e.g. if you are trying to write an exam answer), you may want to be quite clear in saying how you are interpreting the question. For example, you might want to start by saying “In this answer, I interpret the phrase ‘international cooperation’ to mean …” If your interpretation is a reasonable one, this places the onus on the professor to either read the essay according to your interpretation or to justify (at least to herself) why not.

Avoid Data Dumps

Poor essays very often ignore the question asked in a quite specific way. The student spots some topic in the assignment that seems familiar, and immediately sets about writing an essay which tells the professor everything they know about that topic, in no particular order. For obvious reasons, such essays rarely receive high grades. Universities encourage (or, at least, they should encourage) students in the social sciences and liberal humanities to criticize, to analyze, and, ideally, to think. Mere demonstration that one possesses a disorganized body of knowledge on a topic, and is prepared to inflict this jumble upon the professor in printed or (worse) handwritten form, suggests that this encouragement has fallen on untilled ground.

Cut to the Chase

Undergraduate essays frequently begin with an extended session of throat-clearing irrelevances and vague generalities. They talk about everything except the question that has been asked. Take this example (drawn from a free term paper).

The onset of computers on the general population has given a boost to the Economy in the world’s market. People who weren’t much aware of the world became drawn to computers, which in turn brought about the Internet, connecting the world all over. The Internet has played a major role in the lives of people all over the world. Now, it is not limited to just important organizations or governments. Everyone who owns a computer is logged on to the Internet; and this has made the world seem smaller. No one has to wait for the postman to deliver the mail, but instead one can just connect to the Internet and right away, you got mail.

This, like many other essays that I have corrected on the political consequences of the Internet (I sometimes teach a course on this topic) begins with a paragraph that has nothing whatsoever to say about the politics of the Internet. Instead, the paper’s author sees the word ‘Internet’ and grabs desperately for banalities that he associates with this word.

Alternatively, students sometimes state and re-state the question in a manner intended to suggest that they understand it, without ever providing anything so provocative as an actual answer.

Should the internet be censored ? The internet is a wonderful place for entertainment and education, but, like all places used by millions of people, it has some pecularities that lead to a lot of talking and arguing over, should the internet be censored? Most of people who use the internet are furious about the censorship on the internet. The issue of whether is it necessary to censor the internet is being argued all over the world.

This essay starts off well. It sets out a short, pithy question that the reader might hope will be answered in the paper. But then it goes horribly, horribly wrong. The second sentence restates the first, garnished with a couple of irrelevant commonplaces. The third sentence suggests that there is controversy surrounding the topic of Internet censorship (a safe guess, given that the writer has been asked to write a paper about this controversy). The fourth sentence repeats the third. And so on. The writer evidently knows little or nothing about the essay topic, and is trying to conceal that fact. Unfortunately, he or she is failing.

These are the beginning sections of very bad essays. Most undergraduate essays are not nearly as bad as that. Still, many essays do begin with weak and meandering introductions that do not address the topic of their paper. This is a shame. It is important that you get the introduction right. This is your best opportunity to grab the reader’s attention and to persuade her that you have something interesting to say. Don’t waste it.

By the time the reader has finished reading the first two sentences, she should know which question the essay addresses. By the time the reader has finished reading the first five or six, she should have a pretty good idea of how the author is going to tackle the question. The following provides one example of a punchy beginning (nb: this is not taken from an essay mill):

Should the Internet be censored? While many Americans would say no, there is in fact a very good case for limited Internet censorship. Pedophiles can use the Internet to find each other and to swap child pornography. Terrorists can use the Internet to propagandize for their beliefs, and to recruit for their causes. Neo-Nazis and others can spread disinformation to the gullible, and persuade them that the Holocaust never occurred. In this essay, I argue that some kinds of Internet speech (child pornography, terrorist recruitment and hate speech) should be banned. I acknowledge that this may hurt legitimate forms of free speech if they become confused with the harmful kinds, but show that the beneficial consequences of banning bad speech outweigh the harmful consequences of accidentally banning (some) good speech.

This is, in my opinion, a good opening paragraph (since I wrote it myself as an illustration, it is perhaps unsurprising that I like it). It immediately states the question that the essay will try to answer. Shortly afterwards, it provides the reader with the proposed answer, and briefly describes the kind of evidence that it will use to support this answer. The introduction also acknowledges that there is a strong opposing case (that banning ‘harmful’ speech will hurt other kinds of speech), and promises that it will try to answer that case. The essay will not necessarily convince its readers (it takes a quite controversial stand), but it does signal to the reader that it has a clear question, a clear answer to that question, and a willingness to address the best arguments against the case it is making. That is all that any professor may reasonably ask for; not that she agree with the writer’s argument and conclusions, but that she recognizes them as well written, well structured, and well supported by the evidence.

Organize, Organize, Organize

Many student papers are badly organized. They wander from point to point. They tack an introduction and conclusion onto a main body that does not have any internal system of order. Or they do not have a distinguishable introduction, body, and conclusion at all.

Some excellent essayists can get away with apparently disorganized writing. It is usually a very bad idea to try to emulate them. Very often, apparently disorganized work is in fact highly organized. The author has merely kicked away the essay’s supports and scaffolding (e.g. an explicit introductory section and so on) as soon as it was strong enough to stand on its own. Sometimes, apparent disorganization is instead the product of a highly subtle mind, or of an elliptical writing style that approaches its topics indirectly rather than directly. Unless you are very confident indeed (and have evidence in the form of past work, print publications etc to justify this confidence) I strongly recommend that you avoid overly clever and non-linear approaches to writing. They require a lot of practice (usually at the more traditional sorts of writing) before they can be carried off well, and when they are carried off badly, they are very bad indeed. Genius may do as it will; mere intelligence and talent should be appropriately modest in their ambitions.

Thus, the need for structure. You should structure your essay at three levels.

Macro-structure

This is the broad structure of the essay itself. Unless you feel very comfortable that you are an excellent writer, it is usually best to stick to the traditional frame of an introductory section, a main body, and a conclusion. The introduction tells the reader what you are going to say. The main body tells the reader what you are saying. The conclusions tell the reader what she has just read (perhaps adding some thoughts as to its broader implications if you are feeling adventurous).

This not only helps the reader understand your argument, but disciplines your thought and prose. It forces you to begin your essay with a clear, concise account of your major claims. When you write the main section of the essay (or re-write it, as needs be) the introduction will provide you with a roadmap of what you need to do. Your conclusions, in contrast, should draw the threads together, showing how the facts and arguments you have laid out in the main body actually speak to the broad themes discussed in the introduction, and drawing the threads of your narrative together into a proper whole. Of course, for this to work it is necessary that the main body of your essay actually speak to the arguments laid out in your introduction, that your conclusions relate to the main body, and so on.

Meso-structure

This is perhaps the most commonly neglected element of structured writing. It concerns the paragraphs into which your prose is organized. Each paragraph should focus on one main point. The point of each paragraph should build on that in the previous paragraph, and create the foundations of the next. Each paragraph should be a necessary part of the overall structure of your essay.

I find that a useful mental exercise is to boil down the arguments of each paragraph, one after the other, into single sentences. Then, put all these sentences together into a consecutive narrative, looking to see whether each sentence can be made to flow naturally from the sentence previous to it, and into the sentence following. This will highlight any major structural problems. If you are not able to boil down each of the paragraphs into a single sentence summary (however simplistic), then the offending paragraphs most likely need to be rewritten more clearly. If there are gaps or non-sequiturs when you put the one sentence summaries together, then the meso-structure of your essay needs to be re-organized, by cutting and pasting paragraphs, or by introducing new paragraphs to fill the gaps, or deleting old paragraphs that detract from the flow of your argument.

Micro-structure

What is true of the paragraph is also true of the sentence. Each individual sentence should flow in a logical and obvious way from the sentence before, and into the sentence after. Consider the following paragraph, taken from a term paper on global warming which is available for free online.

Weather these days has become very unpredictable. The increase in the world’s temperatures, believed to be caused in part by the greenhouse effect which is known as global warming has and will have a serious effect on the future. Global warming creates massive concerns for the entire earth. If the heat continues to increase several species may struggle to survive. There are numerous political, environmental, economic, and social issues when it comes to global warming. Global warming is an inevitable issue and by no stretch of the imagination can be slowed down easily. There is an inconceivable amount of causes that connect to global warming.

This is quite wretched writing. The first sentence is a vague generality that does not mean very much. The second sentence does not flow in any obvious way from the first. What does the greenhouse effect have to do with unpredictable weather? No explanation is provided for the reader. The third sentence merely repeats the argument of the second, with greater rhetorical alarm. The fourth does a little better, but loses force because it is so badly written (the claim that ‘several species’ may struggle to survive suggests that only five or six species are in danger, which sits awkwardly with the previous sentence’s suggestion that global warming causes “massive concerns” for the entire earth). The fifth sentence seems to build a new set of claims, and should be at the beginning of a new paragraph. However, it never goes anywhere. Instead, the sixth sentence warns that global warming is “an inevitable issue” (whatever that means), while the seventh sentence wrings its hands in despair over yet another new claim – that there is an “inconceivable amount” (sic) of causes “that connect” to global warming. These sentences are not only bad in themselves – they are not connected in any logical or orderly way. The result is that they do not add up to a coherent argument.

Exercises in Style

Political science is not a discipline notable for lovely prose. The best historians often write beautifully; the best political scientists rarely do. Good political science writing does not require striking metaphors or clever verbal constructions (while these are not precisely discouraged, they are not commonly regarded as necessary). Instead, it requires simple, direct writing, which communicates its arguments and evidence as clearly and unambiguously as possible.

The implications for prose style are straightforward.

First, use direct language when at all possible. This not only reads better; it communicates clearly who is responsible for what. For example, the sentence

The Iranian government censors newspapers and political websites.

not only reads much better than

Newspapers and political websites are subject to a censorship regime in Iran.

but it conveys more information in fewer words. It tells the reader who is responsible for censoring information (the government). The alternative version provides less information (the reader may guess that the government is responsible for censorship, but she cannot be sure). It also sounds cumbersome and laborious. Students sometimes use indirect constructions or the passive voice rather than direct language and active verbs because they think this will make their writing more sophisticated and ‘academic.’ They are wrong. Even worse, they sometimes prefer indirect language because they believe that it allows them get away with knowing less, by fudging their argument so that it can be interpreted in more than one way. Neither are good reasons. Indirect language often sounds weak, uncertain, and bureaucratic, and experienced readers will recognize when it is being used to bamboozle them. Sometimes, passive and indirect writing is appropriate, but it should be used with caution. It is usually better for students to err

Second, prefer simple words to complex words, and plain language to jargon. Sometimes it will be impossible to avoid jargon or obscure terms. However, it will usually be possible to use simple terms to convey your meaning. When you can do so, do so. Plain language makes life easier for the reader. It also makes it harder for the writer to get away with nonsense. If you use flowery language, you can sometimes persuade yourself that you understand topics and debates which you really do not. If you use plain language you will be forced to confront your areas of weak understanding and to rectify them.

Third, prefer straightforward sentence structures to complex ones. Again, simple sentences usually read better. Some writers (the historian Edward Gibbon is a fine example) can use complex sentence structure to convey irony and secondary meaning. You – unless you have grown up conversant with a prose tradition like Gibbon’s, in which case you have no need whatsoever to read primers like this – probably cannot. You should typically prefer simple sentences with the bare minimum of sub-clauses needed to convey your argument. Formless and incoherent sentences usually suggest formless and incoherent thought, and indeed they may plausibly cause intellectual incoherence. If you reduce your language down to plain, simple sentences with clear structure then you will again be less likely to hide any lack of understanding from the reader and yourself.

Conclusion

Writing good political science essays is not as hard as it seems. It does not require verbal creativity so much as an ordered and disciplined mind. Most obviously and simply, you should read and understand the essay assignment. You should begin by grabbing the attention of the reader with a clear statement of the question that you wish to answer, and how you wish to answer it. You should ensure that your essay is structured and well organized, so that each part does its part, and fits together well with the other parts. Finally, you should ensure that your prose style does not get in the way of clear thinking and clear exposition. If you adhere to these simple rules, you are not guaranteed to write a good essay. No set of mechanical rules can provide such a guarantee. You will, however, avoid the basic mistakes that have plagued 80% of the bad political science essays that I have read over my nine years of teaching.