Scrutiny of the introduction

Scrutiny of the introduction

© 1995 Jon Claerbout
Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.


Throughout the years I have participated in reading committees of more than a hundred doctoral dissertations. Additionally, reports of the Stanford Exploration Project (SEP) contain about sixty papers a year; and I am nominally in charge of making them presentable. In all this activity I have seen many poor abstracts; and in each case I have spared myself and the author much struggle by referring to the short paper, Scrutiny of the abstract, by Landes [1966], which was made available to SEG's aspiring authors. I rarely rewrite authors' abstracts any more---it's usually enough to have them read Landes' paper and rewrite it themselves. Landes' own abstract is worth quoting:

The abstract is of the utmost importance, for it is read by 10 to 500 times more people than hear or read the entire article. It should not be a mere recital of subjects covered. Expressions such as 'is discussed' and 'is described' should never be included! The abstract should be a condensation and concentration of the essential information in the paper.

Introductions are not easy to write either. I am pleased to report that in recent years I have developed a formula for the introduction. With this paper expounding my formula, I am hoping to reduce the need for one-on-one tutoring. You might be able to produce a good introduction without following my formula, but if you are having trouble producing an introduction that pleases other people and you would like to finish the writing and get on with your life, then I suggest you follow my formula.

First I'll describe the three essential parts of an introduction and then I'll give you some tips on overall organization. You will see why introductions are so difficult to write once you understand how introductions depend on that most embarrassing of all words, "I".


My formula for an introduction is a sequence of three parts. They are (1) the review, (2) the claim, and (3) the agenda.

The review

Pick out about 3-10 papers providing a background to your research and say something about each of them. You could paraphrase a sentence or two from each abstract. The review is not intended to be a historical review going back to Newton or Descartes. Try to find a few papers by your office mates, your advisor, your predecessors, or other associates. That way you might find somebody to give you helpful criticism!

Anyone can follow these instructions and write a review that looks presentable. Where intelligence and skill are required is in organizing the review so that it leads up to something, namely, to your claim.

The claim

The most important part of the introduction is buried in the middle. It is the claim. The claim is where you claim your work is a worthwhile extension of the review you just wrote. If someone says your writing is "unmotivated," they are not insulting your humanity; it just means they can't find your claim.

In your claim you should use the personal pronoun "I" (or "we" if you aren't the sole author). The word "I" tells people where common knowledge runs out and your ideas begin. If you are writing a doctoral dissertation or an article for a refereed journal, then you should be making a new contribution to existing knowledge. Your paper is not acceptable without an identifiable claim.

The agenda

An agenda is found at the end of many introductions. It summarizes what you will show the reader as your paper progresses. Your agenda will be dull if it is merely a recital of the topics you will cover. Your agenda should tell how your paper works to fulfill your claim. In this way your agenda should clarify your claim.

The agenda is not as important as the review and the claim. Keep it short.

Occasionally you will be fortunate enough to be writing about something in which some of your conclusions can be made in simple statements. If so, state them early, right after your agenda. You aren't trying to write a mystery! Many more people will begin reading your paper than will finish reading it. Motivate them to finish! Unfortunately, many technical papers do not lend themselves to early conclusions.


Of course you want people to read beyond your introduction too. So think carefully about the order of your material and how you say things. (Notice this tiny paragraph is a small abstract of what follows).

You may use personal pronouns elsewhere in your paper, too. Generally, you should use a personal pronoun whenever you are expressing an opinion or exercising judgement. Another time to use "I" is whenever there is a simple matter of choice. For example, "To test the theory I selected some data," or "To examine the theory I programmed the equations," or "To evaluate the theory I made some synthetic seismograms."

Whether your ideas are solid as bed rock or speculative as clouds, you need first-person pronouns. Where your ideas are speculative, the pronouns signal a disclaimer. Where your ideas are solid, the pronouns signal that you may be credited for them. When your friends see your personal pronouns they will know just where they should offer their questions and suggestions.

Good scientific papers contain wide ranging statements from ancient axioms and common knowledge to speculations and outright guesses. It is the writer's fault if a casual reader cannot distinguish these types of statements. Personal pronouns are good words to help keep the distinctions clear. Other good words for this purpose are "should, could, would, might, may, can, is, does,...". Use them all and pick the best for each purpose.

Some editors of scientific papers mechanically introduce the personal pronoun "I" to avoid the passive voice. I don't agree with them. For example, such editors will change your words "Substitution of equation (1) into equation (2) gives..." into the words "Substituting equation (1) into equation (2) I find...". The first wording states a simple fact but the second wording hints that someone else might get a different result.

Order of material

My papers often describe processes. I find it best to first describe inputs, next I describe outputs, and last I describe the process itself. It is illogical to describe the output before the process which creates it, but that nonlogical approach helps people see where I am going and why I do what I do.

You could write your paper so that each part builds on earlier parts. Like the axiomatic approach to geometry, you could refuse to refer to things not yet proven. But, rather than write your paper that way, it is wiser to maximize your readership. More people will begin your paper than will reach the end. Try to state results before you prove them. Push off complicated derivations and digressions until the end. Complicated mathematical derivations, especially if marginal to your main thesis, should be relegated to appendices.

Wordy constructions rob the prose of force. -Allen Matusow.

What is central and what is peripheral?

In your paper you might want to include digressions, possible applications, etc. That's nice. But be sure to include language that labels them as peripheral. If you don't, some people will miss your main point and quibble with these peripheral matters.

Writing English as a second language

What is a preface?

A technical document rarely has a preface because a preface describes the document itself without describing the material content. For example, if I made a preface to this article it would say that this article previously appeared in a report of the Stanford Exploration Project.

The introduction to a speech

The introduction to a speech is quite different from that of a technical paper.

Paper Lecture
Review Attention getter
Claim Thesis
Agenda Preview


This short article is not a typical technical paper, but you might like to look back at my introduction to see if I follow my own advice.



I appreciate all translations, but I've been advised that Google Translate might do better on European languages.

Azerbaijanian language by Amir Abbasov

Chinese language by Alexander Tse.

Czech language by by Lera Domartina.

Slovak language by Barbora Lebedova

Spanish language by Chema Bescos.

Ukrainian language by Fixgerald