Structuring the argument of a theoretical paper in the social sciences A guideline for presenting original ideas convincingly to colleagues in humanities and sciences
Richard Parncutt, Uni Graz, revised November 2015
“The essence of the independent mind lies not in what it thinks, but in how it thinks.” Christopher Hitchens, Letters to a Young Contrarian, 1968.
What is the most important thing that a student learns at university? The answer is surely the ability to think and communicate clearly, critically, and independently, and to shed light on difficult issues by constructing a convincing argument based on the best available sources of information. A related skill is to think constructively and critically about the arguments of others, highlighting their good points and identifying and avoiding logical fallacies. This applies in all academic disciplines; the present guideline is tailored to the social sciences, but it can also be applied to any complex qualitative question in any discipline.
Put another way: One of the main reasons why universities exist is to train students to think. University students and teachers must maintain and defend a tradition of independent thought. Otherwise the universities will turn into vocational colleges - which are also important and have their place, but a different function.
Those islands of independent thought known as “universities” have always contributed significantly to human development - otherwise they would not exist. This is especially true in an age of existential threats to humanity, such as continuing global poverty and human rights violations, intercultural conflict, the thinning ozone layer, deforestation, climate change, ocean acidification, genetic manipulation of disease, proliferation of nuclear weapons, religious extremism, terrorism and so on. These are political problems, of course, and solutions will always depend on power relationships. But it also helps if the people involved (i) genuinely want to solve the problems, (ii) are well informed about the opinions of recognized experts and the results of the best studies, and (iii) have a good command of knowledge and argument construction. A good theory of argumentation should be applicable to both politics and academic research (more). That is the practical, general level at which I would like to pitch the theory.
Given the central importance of argumentation, you would think that every student would study the underlying principles. But most are too busy learning the details of their chosen discipline, which is invariably full of busy scholars and researchers trying to publish a constant stream of new (or seemingly new) ideas. As the volume of literature grows (apparently it doubles every 20 years), students are increasingly stressed out trying to get a grasp of at least some of it. They feel increasingly powerless to rise above the detail and see the big picture: where their discipline is going, what it is for, what are the main issues, what kinds of mistakes can be made by even the best researchers, and so on. One aim of this document is to give students that power back.
Seeing the big picture and evaluating the literature - separating the more important things from the less important things, and being able to explain the difference - are essential skills of any student in any discipline. These skills are particularly important when it comes to doing original research. These are not easy skills to acquire; they must be applied repeatedly to different questions, and students must be prepared to learn from their mistakes. In our traditional academic system, the first time a student does research that is internationally recognized as "original" is as a doctoral student. It follows that students should be acquiring general skills of argumentation before they start the doctorate. They should be learning general principles of theory construction, and applying them to central problems in their own field, during their Master's course. The process can begin at the end of a Bachelor's program, when students write a Bachelor's thesis. In many disciplines, the Bachelor's thesis is supposed to cover and interprete a body of literature on a given topic. That is an excellent opportunity to get started on the academically and pedagogically central task of structuring a theoretical argument. The Master's curriculum should then provide several opportunities to improve these skills, before the doctorate begins.
At present, to my knowledge, general principles of argument construction and critical thinking are studied only within the discipline of philosophy. There is remarkably little work of this kind going on in other disciplines. Philosophy students also study the historical development of philosophies of knowledge and knowledge construction. That is important for its own sake, but it is not a prerequisite for understanding the kind of argumentation theory that students need in other disciplines. We need a modern approach to argumentation theory that is geared to the needs of tomorrow's academics in a post-industrial knowledge society. The terminology within the theory should be expressed in everyday language so that no philosophical or mathematical background is necessary. The theory should be grounded in common sense, and it should be written in a way that is easy to understand across contrasting disciplines, including humanities and sciences.
In summary, there is a need for university course units that:
Present general principles of clear, constructive, critical thinking, argument, and persuasion that apply to qualitative arguments in any discipline;
Apply those principles to current research issues in specific disciplines;
On that practical basis, allow students to theoretically appraise and revise the general principles, regardless of the specific question or discipline; and
In that way, put doctoral students in a good position to make a significant and valuable contribution to their specific discipline (or to several disciplines, in an interdisciplinary project) - and possibly to society in general, as well informed, caring citizens.
Theoretical presentations at scientific conferences
Conferences in the natural, social and formal sciences often distinguish between two kinds of paper: empirical and theoretical. An empirical paper is a report on an empirical study and comprises introduction, central sections (usually method and results) and conclusions. A theoretical paper presents and evaluates a claim (or series of claims) and/or a theory, and comprises introduction, central sections (with topics corresponding to aspects of the question or theory) and conclusions.
The present guideline is intended for theoretical papers. There are many different guidelines for writing empirical papers in the literature, so students and young researchers have a lot of materials to choose from. By comparison, there is not much out there about writing a theoretical paper.In many disciplines (including my own disipline, music psychology) there seems to be undue emphasis on empirical work at the expense of theoretical work, and sometimes colleagues seem unsure how best to approach purely theoretical work, or how to evaluate it.
This guideline is intended to fill that gap. The idea is be as systematic and comprehensive in theoretical work as we already are in empirical work. A broader goal is to highlight the equal importance of empirical and theoretical work, and the balance that should exist between them.
More specifically, this guideline should help students prepare theoretical papers for research conferences. In that case, the idea is not to follow the guideline rigidly, but instead to think about whether your presentation contains the key elements listed in the guideline and incorporate them appropriately. The result should be a presentation that experts in your chosen question will find interesting, because it is addresses a question of current interest (perhaps even a "hot topic") and in that way attempts to expand the boundaries of knowledge. You can do that by asking questions that do not yet have clear answers, and developing possible answers (theses) that are nottrivially true. They should, of course, be probably true, and it is your task to convince the experts of that.
A talk based on this guideline is fundamentally different from an introduction to a specific question that is directed toward non-experts. The difference will usually be clear from the title of the paper. An introduction for non-experts may have a short title; examples from my discipline might include "Music and emotion" or "Counting pitch-time patterns in musical databases". A theoretical paper for an expert audience generally has a longer, more specific title that alludes to a specific question or thesis and distinguishes it from all other papers in the literature on that topic, for example "The role of articulation in the expression of basic emotions" or "The changing frequency of occurrence of suspended triads in European vocal polyphony from the 13th to the 19th centuries".
The role of metacognition
Skills of argumentation and critical thinking are an example of metacognition: cognition about cognition, or knowing about knowing. Metacognition includes knowledge about how to solve problems: what strategies are necessary and when they should be applied. Any student who can explain how they are setting about solving a problem is displaying metacognitive skills, and research on teaching and learning has repeatedly shown that students with more of these skills learn more efficiently than students with less of them. Metacognition is what students need to know when searching for answers to central questions and writing about those questions and answers in a theoretical paper.
When students are asked to theoretically appraise general principles of argument construction, they are being asked to think and talk about metacognition, which could be called meta-meta-cognition. That sounds complicated, but in my experience it lies well within the capabilities of typical Bachelor's or Master's students in all disciplines, if the material is presented clearly and appropriately.
Avoiding logical fallacies
Clear thinking includes avoiding logical fallacies. Philosophers have identified many different kinds of logical fallacies and applied a lot of somewhat dry names to them (link). It would be nice to know all of these categories, but it is more important to be able to spot logical fallacies wherever they occur, just by thinking critically about what you are reading or hearing. For a good, accessible, general introduction to logical fallacies from the point of view of natural sciences, watch to Colin Frayn's video on Understanding Science Lecture 8: Logical Fallacies. For a more political approach, study the internet page entitled Thou shalt not commit logical fallacies.
If you identify the main conclusion of an article as a logical fallacy, the best solution is simply not to cite it. Focus instead on the other literature. If you think an article is interesting but may contain a logical fallacy, discuss the problem in a positive way that might eventually solve the problem. These are important aspects of the critical evaluation of research literature, and more generally of critical thinking.
Political background and relevance
Structuring a convincing argument is not a dry, abstract task. It is a central human skill that has always been part of the human condition - especially in politics. In a democracy, people vote for people whom they believe are presenting the right ideas and supporting them with the right arguments. But what, in general, is the "right argument"? Considering the following two quotes:
Die Wahrheit ist dem Menschen zumutbar. (Ingeborg Bachmann, 1959)
The first step towards nonviolence, which is surely an absolute obligation we all bear, is to begin to think critically, and to ask others to do the same. (Judith Butler, 2016)
I think what we mean by reason can be defined by three characteristics. In the first place, it relies upon persuasion rather than force; in the second place, it seeks to pursuade by means of arguments which the man (sic.) who uses them believes to be completely valid; and in the third place, in forming opinion, it uses observation and induction as much as possible and intuition as little as possible. (Bertrand Russell, 1935, p. 56)
Of course it is reasonable to expect the truth, as Austrian poet and author Ingeborg Bachmann famously pointed out. The reason why her statement became famous is evidently that we are often not told the truth. Sometimes that is because people are lying: deliberately withholding, distorting or changing the truth for personal gain. Sometimes the truth is so complicated that even well-meaning, honest, intelligent people have trouble formulating it. That is where the theory of argumentation comes in.
The second quote is by British philosopher Bertrand Russell. It is part of his attempt to understand the rise of fascism and (to a lesser extent) communism in their historical context, and develop democratic-socialist alternatives. Russell regarded the emerging political movements and doctrines of 1930s Europe as "anti-rational" revolts against reason, reasoning and reasonableness. Authoritarian structures had been created to promote utopian, fundamentalist ideologies, and to deny their evident moral and practical flaws by suppressing open discussion. To understand and overcome anti-rationality, Russell reconsidered the nature of reason.
Each of Russell's three points is relevant to my approach to structuring the argument of a theoretical paper in the social sciences.
Focus on the argument itself - not the person presenting it or the scholars being cited. Students should not be convinced by lecturers simply because lecturers have more power or charisma; similarly, lecturers should not be convinced by professors. In the absence of such power games, academic communication relies heavily on skills of persuasion - and these skills can be learned. Students can learn to discuss and argue with their superiors as if they were peers, in order to create a level playing field and a foundation for academic democracy - while at the same time appropriately acknowledging differences in expertise and experience.
Systematically evaluate and compare different possible arguments, theories, claims. Students should believe in their arguments, but I disagree with Russell's positivistic formulation that one should believe one's arguments to be "completely valid". Instead, students should believe and demonstrate that their arguments are superior to other feasible or possible arguments. This idea is implemented systematically in my guideline below by a comparative evaluation of different possible theses, regarded as answers to a main question.
Ground arguments in systematic observation and transparent reasoning. Russell refers to processes of observation and induction. In the social sciences, these correspond to empirical methods and theoretical work respectively. The following guideline is for writing a theoretical paper that refers primarily to the result of empirical studies. Intuition is also important, but in work of this kind it should not be allowed to take a front seat, unless little or no empirical evidence is available. By intuition I mean spontaneous insight - knowledge acquisition without conscious reasoning or inference, or without reflection on the thought processes (aka metacognition) that lead to a given idea, conclusion or hypothesis. Intuition is an important aspect of creativity, but if taken too seriously and not intersubjectively controlled (e.g. by democratic expert processes such as peer review) it can lead to unnecessary conflict.
It can be useful to think of theories, theses and arguments as having independent existence (Karl Popper's "objective knowledge" or World 3). But in reality theories are always created by people, and people have different qualifications, reputation, experience, power, connections, attitudes (e.g. to morality), beliefs (e.g. religious), interests, motivation (driven by emotion), and so on. Arguments can be affected, distorted, or biased in their content or presentation by a long list of personal factors. These factors may reduce our ability to think critically or objectively, or to reflect in an honest or reasonable way on our own subjectivity.
Acknowledgement of the qualification and expertise of an author or discussant can play an important role in clear critical thinking, especially if you do not feel sufficiently qualified to decide whether a given statement is true or not. Experts are necessary to solve difficult problems. To take advantage of their expertise, it must be possible to evaluate their expertise relatively objectively. A classic modern case of failure to recognize expertise is climate denial. Climate deniers fail to recognize the expertise of climate scientists and infrastructures of climate science that are used to guarantee scientific quality, such as peer review procedures, which they may fail to understand because they never participated in one. Climate denial is just one of many forms of scientific denial (tobacco-cancer denial, HIV-AIDS denial, CFC-ozone denial, evolution denial, vacchine denial). The deniers often have something to gain (usually financial, often indirectly) from denying the validity of expert opinion.
Power relations often distort the way people talk about important issues, so the ability to recognize and interprete power effects is another important aspect of clear critical thinking. Journalists can't always write the truth if the media are all owned by Berlusconi or Murdoch. It may be difficult to contradict large powerful corporations (Google, Microsoft...) or religious authorities (Christianity, Islam...), especially if the critic depends in some way on the organisation. In many countries, journalists and artists who openly criticize their government risk serious consequences.
Emotion is another important issue. On the one hand, clear thinking should depend on logic and evidence; it should not be distorted by emotion. On the other hand, it is normal and indeed necessary for people to be emotionally attached to their beliefs, if only because we are more likely to remember something if it has emotional significance for us (LaBar & Cabeza, 2006). A classic example is the question of the existence of God (although I guess the word "classic" is hardly adequate for such an important question). Philosophers have tried to prove the existence of God for millennia, but their proofs were never completely watertight. If you look behind those arguments, you find two considerable sources of distortion: the power of the Catholic church (for most of two millennia it has been dangerous to question the existence of God) and the positive emotional consequences of believing in an omnipresent, all-knowing, all-loving God (which presumably still influences thinking on this subject today). This is an extreme example, but there are many other examples in which emotion clearly plays a role: parents may believe in homeopathy because they are convinced that it cured their child's disease, lonely people may believe in astrology because they are convinced it can help them find a partner, and so on. An interesting source on such issues is Skeptics Society and Magazine.
Further features of a good theoretical paper
Clarity and directness. A good academic essay or theoretical paper is not self-gratifying, important sounding blabla. Instead it is written in a clear and honest fashion, striking a good compromise between clear everyday language and more rigorously defined academic terminology. The text is written with a given group of readers in mind - often the author's peers, researchers in the same area in other places. The writing style should help the target readers understand the material quickly and easily. That gives them the power to criticize it, come up with alternative ideas, and test those ideas in some kind of hermeneutic or empirical process. Good academics are generally open to criticism of this kind (e.g. in peer-review procedures), and many even go out of their way to get it. Poor academics may develop unnecessarily complex or unclear terminology or arguments in an attempt to "blind their audience with science" and in that way to hinder criticism by their peers or students. Another idea of Karl Popper is relevant here: the principle of falsification. If it is not logically possible to falsify or disprove a theory, the theory is unlikely to be true. A theory is more likely to be true if we can conceive of a convincing rebuttal, based on a clear observation, a good source of evidence, or a logical argument. If in spite of this opportunity the theory is not falsified, it is probably ok. It follows that a theory is more likely to be true if it is expressed in clear language that enables and open critical discussion. This is a scientific idea, but it applies equally well in the humanities; no-one in the humanities will take an idea seriously that is obviously wrong, for whatever reason. So if you have the feeling that a speaker or writer is saying something important, but you are not quite sure what that important thing is, because the speaker or writer deliberately failed to explain it clearly, then watch out: "bullshitting" (truth distortion) is very common in academe, and one of the most important things that students should learn at university is how to spot it and avoid it.
Focus. A good piece of academic writing is an attempt to push back the boundaries or knowledge, or expand existing ways of thinking. Changes in the culture of knowledge and thinking usually happen in small steps, and sometimes small steps have large implications. We should therefore refrain from trying to change the world and instead focus our writing on a specific issue, looking at the issue in depth from a number of different angles. Often, this means formulating and testing a thesis. We then focus all parts of the text on the question of whether, or under what circumstances, the thesis may be valid. The materials below show in detail how this can be done effectively.
Interdisciplinarity. Interdisciplinarity is anathema to fundamentalist ideologies, whose proponents are often either unwilling or unable to see things from different viewpoints, or to seriously consider different possible epistemologies. But that may merely be an extreme example of a widespread phenomenon. In my experience, a certain entrenched resistance to appropriate interdisciplinarity can be observed in most academic disciplines. There may be good reasons for that: interdisciplinary scholars run the risk of appearing as amateurs to one or both sides, and the additional effort that is normally required for interdisciplinary research may be insufficiently rewarded within institutional infrastructures. Interdisciplinary research is characterised by difficulties of communication between researchers from different academic traditions, and the difficulty becomes acute when the research attempts to traverse the treacherous gap between the humanities and sciences (the "two cultures" of Snow, 1960). But this gap must surely be bridged if humanities and social sciences are to achieve a high academic standard. Humanities scholars and scientists who are interested in similar questions should not simply ignore each other. Given this background, my materials (especially the "tabular argument" or "Argumentationsgang" below) aim to balance and combine the usual approaches of humanities and sciences - to develop and apply basic ideas and assumptions about "good" research that are common to both sides. This is not a popular approach: when humanities scholars read my materials, they may find them too scientific (a bad thing, apparently), while scientists find my approach too close to the humanities for comfort (ditto). But my goal will have been achieved when I can convince leading scholars on both sides that the contrasting approaches of humanities and sciences are complementary (not contradictory). Beyond that, I hope to convince colleagues that this academic process is related to global political processes of conflict resolution. Interdisciplinary collaboration can be promoted for its own sake, with the goal of improving academic standards; or it can be promoted as a model for constructive collaboration, as part of a more general pacifist orientation.
My approach is inspired by to that of Toulmin (1969), who uses similar concepts but sometimes different terms: claim (which I call thesis), data (evidence), grounds (arguments for the thesis), warrants (how the evidence supports the thesis), qualifiers (limitations), rebuttals or reservations (evidence and arguments against the thesis), backing (subtopics that support different aspects of the thesis).
Another related approach is that of Booth, Colomb, and Williams (2008). An important focus of their book is the art of effective communication with potential readers, which can be broken down into several different elements: identifying target readers, convincing them of the importance and relevance of the research question, presenting material in most logical and comprehensible order from their perspective (old before new, specific before general, simple before complex), and anticipating their reservations and responding to them appropriately. Booth et al describe the structure of an argument in a similar way but with some differences in terminology: they talk about reasons rather than grounds, and subordinate arguments rather than evidence for and against subtheses.
Using these materials
This guideline focuses strongly on the student's thesis (or main claim), which is intended to be original but in practice is similar to theses already expressed in the literature (sometimes embedded within other arguments). The student's task is to formulate a thesis that is interesting (which means it has practical implications and could well be true, but is not obvious). She or he then asks: to what extent might the thesis be true, or not true? And on that basis: Can this thesis expand our knowledge of the topic? A secondary focus of the method is interdisciplinarity: How do people in contrasting disciplines approach these issues? What can we learn from trying to combine their approaches? To answer these questions it is of course necessary to survey the relevant literature. But the presentation should never slip into a series of summaries. Instead, it should be structured around the student's original argument. All references to the literature should refer directly to that structure.
To enable this to happen in practice, the student fills in a table called a "tabular argument" (link below) that is intended to cover the central features of a good academic argument. Typically, the contents of the table are revised many times before the student starts to write the final text, which may for example be a Bachelor's thesis. The table is then included as an appendix. Many colleagues find this approach overly formalistic, but they usually also agree that the points addressed in the table are important ingredients of any argument. The tabular argument is intended as an exercise to be done thoroughly, but only once, or a few times at most. If students apply the ideas in the table to research questions that interest and motivate them, they will internalise the main ideas through a process of learning by doing. In my experience, that is pedagogically much more valuable than an abstract theoretical consideration of the same ideas. After that, students can return to a less formal approach to structuring an academic text, armed with a better understanding of general principles of critical thinking and academic communication.
The basic elements of an argument
Let us assume that you want to convince your target audience that you have found the right answer to a given question. That sounds arrogant, but this is often the main motivation that drives academics. They want to become known for having answered interesting questions. So imagine for a moment that you are one of us arrogant, obsessed academics, and you want to say something like this: "I believe that explanation X is the best answer to question Y, for the following reasons: reason 1, reason 2 and reason 3." Imagine that you then go on to explain why each of these reasons is separately true, in order to convince your target audience - most likely your fellow researchers in different countries - that X is true. Have you ever tried to do something like that? Not with international academics as your audience, but a family member or a friend? If so, you already have some basic skills in the art of structuring an argumen and writing a good theoretical abstract. In the jargon of this approach, incidentally, X is the main thesis, Y is the main question, and 1, 2 and 3 are subtheses.
Now think about the structure of this statement: "I believe that explanation X is the best answer to question Y, for the following reasons: reason 1, reason 2 and reason 3." This is not the only possible structure for an argument, of course. But it is a very common form. It is an argument based on convergent evidence: the three subtheses provide convergent evidence for the main thesis. The thesis is convincing if it is supported several times, and the different sources of support are plausible and independent. In general, we might say that the chance of convincing a critical expert depends on the number of subtheses (the more the merrier), how convincing they are, and how independent (ior non-overlapping) they are.
Have you ever been amazed by someone's failure to understand a simple argument of this kind? There are several possible reasons for this kind of stubbornness. There may be problems in the detail of the arguments you are presenting; you have to be self-critical to notice them, listening to objections and adjusting your argument accordingly. Maybe that person is so emotionally attached to a given outcome (i.e. to a given possible answer to your main question) that no amount of logical explanation will convince them. Maybe that person has never learned the basics of the art of critical thinking, either directly or indirectly, or is not aware that good arguments often have this form.
Writing an abstract
Writing an abstract is one of the most important academic skills, because the abstract is read much more often and more carefully than any other part of an academic article. People often find the abstract in the internet and either are unable to find the main text or have no time to read it. Clearly, it is important to attract the attention of your target readership with your abstract. To do that, you should get to the point quickly, focus on the main issues, say specific things about those issues, and do all of this clearly and concisely.
In modern academic research, a theoretical paper generally begins with an abstract whose structure reflects the structure of the paper. In the present approach, the abstract usually comprises the following elements:
(first subtopic:) first subthesis
(second subtopic:) second subthesis
(third subtopic:) third subthesis
main counterargument and rebuttal (i.e. why it is weak or invalid)
implications if the main thesis is correct
That's pretty straightforward, and here are some examples of how it works. Let's start with an important current issue at the interface between research and politics. Imagine you are writing an essay on "The causes of global warming". That is the title on your title page; in the jargon of this approach it is called the "main topic". Your main question is "What is the main cause of global warming?". Here is a possible abstract that includes all the above elements, one after the other:
Global warming may be the most serious problem ever faced by humans. What is its main cause? Convergent evidence suggests that the main cause is greenhouse gases produced by human activity. (Temperature data:) Apart from fluctuations with periods of one or a few years, global mean temperature has risen steadily since humans started burning fossil fuels. (Influences on global mean temperature:) In recent decades, natural influences on the long-term development of global mean temperature have been smaller than human influences. (Modeling results:) Models of the history global mean temperature data work best when they combine several factors, of which anthropogenic emissions is the biggest. (Magnitude of the temperature increase:) A model that includes the thermal capacities of oceans, crust and atmosphere can explain the relationship between total emissions since industrialisation and the magnitude of the increase in atmospheric temperature. (The role of expertise:) Almost all qualified climate scientists agree that global warming is caused by humans; there is no reason to expect bias, and non-experts are not in a position to disagree. Many people are skeptical about global warming because they cannot perceive its effects, but research shows that the effects are already serious. Urgent and radical political action is required to prevent global warming exceeding 2°C later this century.
Can you identify the above elements (starting with the introductory sentence) in this abstract draft? If so, you have understood the main features of this method. Incidentally, the words in brackets should be deleted to make the abstact flow; I have included them to illustrate the difference between a subtopic and a subthesis.
Here is another interesting and controversial political issue. The essay might be called "The advantages and disadvantages of university fees":
Universities are funded from a mixture of sources, and in many cases they include student fees. Should university students have to pay fees? A range of available arguments and evidence suggests that they should not. (Social justice:) Fees can prevent young people from studying whose parents have less money. (Social well-being:) Societies with lower wealth gaps are happier - more peaceful, healthier - than societies with higher wealth gaps. (Universities as long-term social investment:) The general public generally gets a good return on long-term investment in universities. (Taxation:) Universities can be entirely publicly funded if existing taxation schemes are applied fairly - e.g. by closing tax havens and enforcing progressive tax scales - and plans for wealth, environment and transaction taxes are implemented. Students may be more motivated to study if they have to pay for the privilege, but students can also be motivated by adjusting challenges to skills, to increase the sense of achievement. ((I have omitted implications here, because the implications have already been stated as arguments for the thesis.))
Here is a musical example for my students in systematic musicology. The essay might be called "Role models and the motivation to play a musical instrument":
Children vary considerably in their motivation to play a musical instrument, which in turn seems to determine whether they will become musicians and, if so, what standard they will achieve. What can account for this variability? The main source of motivation may not lie in the music itself or in the child's personality or physiology; instead, it may come from role models. Parents can be important role models if they regularly sing and play music themselves, and the child experiences their enjoyment of this activity. Much the same goes for siblings ((subtopic 2)) and peers ((subtopic 3)): friends and acquaintances both at school and outside of school. There is also evidence of motivation from the emotional reward of the musical sounds themselves; that certain personality types are more likely to want to play music; and that the brains of children who play music are different from the start. However, the evidence for these alternatives is weaker. It follows that, if parents or teachers want their children or pupils to engage in music, they must do so themselves, and genuinely enjoy doing so.
Here is another musical example. The essay might be called "The psychological foundations of consonance and dissonance in Western music":
The structure of mainstream Western music depends crucially on consonance and dissonance: dissonances resolve onto consonances with specific voice-leading patterns. What are the psychological foundations of consonance and dissonance? Musical consonance/dissonance is not a unitary construct, but a combination of several elements. (Roughness:) First, musical sonorities are more consonant if they sound less rough - if there is less peripheral interaction between partials falling in the same critical band. (Harmonicity:) Second, musical sonorities are more consonant if their spectra are more similar to a harmonic series. (Familiarity:) Third, musical sonorities are more consonant if they are more familiar to the listener. (Temporal relationships:) Fourth, the consonance of a musical passage also depends on relationships between successive sonorities. One of these factors may be dominant in given situations (for example, familiarity may be dominant for modern listeners), but when the historical dimension is included, there is no clear evidence for one main factor. On this basis, traditional Pythagorean ratio-based concepts of consonance should be abandoned, with interesting consequences for the foundations of music theory.
To write an abstract in this style, try first to concisely formulate your main question, main thesis, subtopics and subtheses. Then identify what is probably the main counterargument (the main objection that experts would have to your thesis) and try to refute it. If you can't refute it convincingly, think about changing your thesis! After that, add an introductory sentence and think about the implications if your thesis is correct. Bingo: you have a first draft of your abstract, and the most important points in your tabular argument.
The deep structure of your argument
A good presentation moves from the specific to the general and back again. The introduction leads the reader or listener from a specific example to a more general issue or question. The main text considers specific aspects of that general question. The final section develops a general answer to your general question and considers specific implications. The process may be conceived to involve a kind of cycle, as illustrated in Figure 1.
Figure 1. Specific versus general aspects of an academic presentation
Your text should begin with a (specific) example, which attracts the reader's or listener's attention and vividly illustrates your main issue or question. If your topic is the role of personal relationships in a child's musical development, you may present a brief case study of a 7-year-old girl called Sarah who is learning the violin. The story includes specific events involving her sister Mary, her brother Jim, her schoolfriends Tatevik and Mohammed, her parents Jill and John, her teacher Susanna, and so on (if the names are invented, just explain in a footnote; but it may also be possible to find a real story that is already published, to which you can refer). The story suggests that each of these people played a specific role in Sarah's musical development. On this basis you can now formulate your main (general) question (e.g. "What role do personal relationships play in musical development?"), knowing that your reader or listener now has good feel for the question's nature, scope, and relevance. These are some of several elements of a good introduction (on which more later).
The central section of your presentation or article then looks at your question from different perspectives, entertaining a range of available arguments and evidence (for example from empirical research in music psychology). After this detailed analysis comes the final section, in which your (general) thesis is formulated to answer to the initial question, and the (specific) implications of that thesis are considered. What if the thesis were really true? What then? The implications are answers to these questions. Students often have trouble with implications, because to formulate them they have to think creatively, laterally, and "outside the box" - an essential skill that many high schools unfortunately do not seem to train. Your thesis has general character, whereas the implications are more specific. Finally, you can explain your thesis and its implications by reconsidering your original (specific) example, which closes the loop in Figure 1.
To convince an expert reader or audience that your thesis is valid, you should present convergent evidence for it, and also consider the main evidence against it. Convergent evidence means that your subtheses should support your main thesis although they are approaching it from different directions. To make sure this is going to happen, subdivide your main question into subquestions (explain how and why you did this at the end of the introduction). The answers to your subquestions are your subtheses. All of your subtheses should support your main thesis (the start of the conclusion is the place for this). These relationships are illustrated in Figure 2.
Figure 2. Subdividing your main question into subquestions, and your thesis into subtheses
There are many different ways to divide your topic into subtopics. You may consider your question from the point of view of different disciplines. You may address some specific context-independent issues, and then consider the more global contexts for which you question has relevance. You may contrast qualitative and quantitative approaches. You may compare different theories. Every interesting question has a unique set of issues to address. The material in subtopics should not be introductory, so the main aim of a subtopic should not be to explain or define a concept or an empirical method, for an example. Instead, each subtopic should allow you to formulate, support and defend a unique subthesis. This subthesis should then support the main thesis (so of course it should not be the same as the main thesis); taken together, the subtheses provide convergent evidence for the main thesis.
Consider again the example introduced above. What contributes most to the success of a young musician? After studying the empirical literature, you may believe that the most important factor is personal relationships. That is an interesting idea, because it is not entirely clear whether it is true, or how much it is true; but good arguments can nevertheless be found to support it. You might then divide the main part of your presentation into four subtopics addressing the role of parents, siblings, teachers and peers respectively. In the first subtopic, you would then present arguments and evidence that parents make an important contribution to the later success of a young musician; in the second subtopic, siblings; in the third, teachers; and in the fourth, peers. In each case you would address different literature sources. After putting together all this convergent evidence, you would stand a good chance of convincing an expert reader that your thesis is correct, or at least a promising idea.
Another important aspect of the structure of your argument is the relationship between evidence, thesis (or conclusions) and implications. As illustrated in Figure 3, this relationship is simple and one-directional: the thesis depends on the evidence, and the implications depend on the thesis. The evidence does not depend on the thesis; nor does the thesis depend on the implications. In the previous example, the arguments and evidence that you present should be independent of your conclusions; to make this clear, you might systematically compare different possible answers to your main question and subquestions, or consider the results of empirical studies whose goal was quite different from yours, but nevertheless produced evidence in favor of your thesis. Your conclusion that the success of a young musician depends on personal contacts should depend only on the arguments and evidence; it should not depend on the implications. The implications may be that music educators should pay more attention to personal relationships of all kinds surrounding music lessons. They may not want to do that, and for that reason they may doubt your conclusion, or try to show that it is incorrect. But the conclusion depends only on the evidence! It does not depend on the implications. Similarly, politicians who are worried that this conclusion may increase the public cost of music education may also question your thesis. In reply you should remind them that thesis does not depend on the implications.
Figure 3. The one-way relationship between evidence, conclusions and implications
For a contrasting example, consider the question of climate change, and who or what is causing it. Many climate deniers refuse to believe that the planet is warming. Alternatively, they refuse to believe that humans are causing climate change, because that would make us (and especially them) responsible. They have a feeling that climate change cannot be caused by humans, because they don't want to share responsibility for such an enormous crime. In an attempt to escape from this responsibility, they talk as if the conclusions of climate science research depended on its implications. In fact, the conclusions depend only on the evidence. Climate deniers may also accuse climate scientists of rigging their evidence to support a given conclusion. But the scientists know that their evidence is (or should be) independent of their conclusions; they are trained to produce independent evidence, and they have access to diverse, tried-and-tested methods for this purpose, from experimental design to peer review. Climate deniers may have little knowledge or experience of these methods. Climate scientists realise similarly that their conclusions should be independent of the implications. They first develop their conclusions from the available evidence, and then consider the implications in a separate step.
Responding to review
Whether you are submitting an article to a journal or an essay for grading - in both cases you may get constructive expert feedback, and you may then have the opportunity to revise and resubmit your manuscript (possibly to get a better grade). A good teacher-researcher will give you feedback on your paper, just as s/he would when asked to review an article submitted to a journal.
This procedure typically takes at least a month, and sometimes several months. First, the reviewer or grader may need a few weeks to prepare the review, because everyone is busy with other things. Then you may need a few weeks to implement the recommendations. After that there will be a second review. So be patient and allow enough time.
Please regard expert feedback as generally valuable. In fact, it is exactly what you need to improve your work. If an expert takes the trouble to remind you that your argument is problematic or you forgot to read an important article, or if s/he is asking you to make corrections that are intended to improve your paper, your response should be one of gratitude. Don't be surprised or disappointed, and it definitely does not help to be angry or offended.
The best policy is to respond constructively to all comments. If you disagree with a suggestion, first try to implement it. Do this even if you disagree with it, or you cannot see the point of it. In the long run, you might find out that it wasn't such a bad idea after all, and even if it is not the best suggestion, you will probably learn interesting things from trying to implement it.
If you really are sure that a suggestion is poor or the reviewer made an error, don't ignore the suggestion and hope s/he won't notice. Instead, try to convince her/him that you are right. This can only work if your arguments are good, and you don't try to do it repeatedly.
Here's a procedure for responding to comments in a way that reviewers are known to appreciate:
Turn on "track changes" in your word processor so the reviewer can see what you did.
Revise the paper taking into account each comment.
Read through the whole text again, fixing up any other problems that have arisen because of the corrections, e.g. in the flow of the text. You may have to rewrite some passages and do some re-organising.
Go through the reviewer's suggestions again and insert a comment after each one, explaining briefly how you responded to it or - better - copying and pasting the corresponding new piece of text under the comment.
If the changes to your text are minor, it may be enough just to resubmit the text with the changes marked. If the changes are major, it may be easier to accept all the changes in the main text so they are no longer marked (because there are so many of them, or you changed the structure) and instead respond carefully to each comment.
These ideas apply more or less equally to advanced student essays (term papers, seminar papers, Bachelor's theses) and articles submitted to peer-reviewed journals. If you are thinking of a research career, it is a good idea to start practising these procedures when you are a student, at any level.
About the author
The content and structure of these materials is based on many years' experience as a member of editorial and advisory boards of academic journals (Jahrbuch Musikpsychologie; Journal of Interdisciplinary Music Studies; Journal of New Music Research; Music Performance Research; Musica Humana; Musicae Scientiae; Musicology Australia; Music Perception; Music Theory Online; Psychology of Music; Research Studies in Music Education; Royal Musical Association Research Chronicle; Systematische Musikwissenschaft) and as peer reviewer of submissions to these and other journals (Acta Acustica; Behavioral and Brain Sciences; Attention, Perception, & Psychophysics; Contemporary Music Review; Cognitive Science; Connection Science; Cortex; Emotion; Empirical Musicology Review; International Journal of Psychophysiology; Interdisciplinary Science Reviews; Journal of the Association for Research in Otolaryngology; Journal of the Acoustical Society of America; Journal of Experimental Psychology: Human Perception and Performance; Logopedics Phoniatrics Vocology; Perception& Psychophysics; Proceedings of the National Academy of Sciences; Psychomusicology; Quarterly Journal of Experimental Psychology).
At the moment (to my knowledge), these materials are used only by my undergraduate and graduate students in systematic musicology - an epistemologically diverse subdiscipline of musicology. The scientific side of systematic musicology includes psychology, acoustics, neurosciences, computer sciences, and empirical sociology; the humanities side includes philosophy and theoretical sociology, and involves aesthetics, semiotics, and hermeneutics. Representatives of these diverse disciplines have similarly diverse ideas about the content and structure of a convincing argument, but their approaches also have a lot in common. Colleagues in other disciplines and at other universities are welcome to use or adapt these materials without permission. I would be grateful for feedback so that I can improve them.
Finally, a personal word of warning to my students: As I help you develop your argument, I may be inconsistent, making different suggestions each time we talk about your project. I may even contradict myself! When I finally grade your work, I may come up with new suggestions that I did not think of before. Please value this inconsistency. It is both normal and constructive. In fact, it may be just what you need! The reason: We are doing real research now. There are no clear answers to your questions; if there were, you would not be asking them. I am training you (or more precisely: you are training yourself) to make your own independent decisions, based on advice from different sources. You are collecting evidence for and against your theses, and one of those sources is me. Every time you ask me for advice, you present your work in a different state of completion and I become aware of different sources of information; so of course my suggestions change. It is up to you to decide what is best at each stage of your research, to believe in your own decisions, and - if they disagree with my recommendations, which is normal and ok - to try to convince me that you are right. I will be grateful if you challenge me, because that way I learn, too. This is a gradual process of increasing academic independence; some people would say that is what university is all about, in the end. I wrote more about it here.
Acknowledgment. Thanks to Christian Fleck and Malik Sharif for comments on earlier drafts of this page.
Bachmann, Ingeborg (1959). Die Wahrheit ist dem Menschen zumutbar. Dankesrede für die Verleihung des Hörspielpreises der Kriegsblinden, Bundeshaus, Bonn.
Becker, H. (1998). Tricks of the trade: How to think about your research while you're doing it. Chicago: University of Chicago Press.
Becker, H. (1986). Writing for social scientists (2nd Ed. 2007). Chicago: University of Chicago Press.
Booth, W. C., Colomb, G. G., & Williams, J. M. (2008). The craft of research. Chicago: University of Chicago Press.
Kamitz, R. (2007). Logik-Faszination der Klarheit: eine Einführung für Philosophinnen und Philosophen mit zahlreichen Anwendungsbeispielen. LIT Verlag Münster.
LaBar, K. S., & Cabeza, R. (2006). Cognitive neuroscience of emotional memory. Nature Reviews Neuroscience, 7(1), 54-64.
Paul, R., & Elder, L. (2003). Kritisches Denken: Begriffe und Instrumente - ein Leitfaden im Taschenformat. Stiftung für kritisches Denken. www.criticalthinking.org.
Russell, Bertrand (1935). The ancestry of fascism. In B. Russell, In praise of idleness and other essays (pp. 53-71). London: Routledge.
Snow, Charles Percy (1960). The two cultures and the scientific revolution. Cambridge: Cambridge University Press.
Toulmin, S. (1969). The uses of argument. Cambridge: Cambridge University Press.