Department of Sociology

University of Toronto

725 Spadina Ave. Toronto, Canada M5S 1A1

e-mail: wellman@chass.utoronto.ca www.chass.utoronto.ca/~wellman

How To Write -- and Edit -- a Paper

(c) Barry Wellman December 14, 2001

I. PURPOSE

A. Enhance creativity and clarity of expression.

B. Liberate creativity by systematizing work, focusing work, using tricks to do "maximum output with minimum effort"

C. Enables you to see ‑‑ and highlight ‑‑ connections between your ideas.

D. Communication, not masturbation (writing for one's self)

1. "A vision is just a vision if it's only in your head. If no one get's to hear it, it's as good as dead!" (Stephen Sondheim, "Putting It Together," from Sunday in the Park with George [Broadway musical, 1984?]

2. ALanguage, including language in academic books, is a communication device. After having investigated social and technological change throughout the world for 15 years, when the time came to communicate my findings, I decided to do so in terms that could connect with a wide, educated readership. It is not a matter of using plain language. It is a matter of writing in ways in which the academic style ... could be adapted for freer expression, taking some literary liberties, as long as I could maintain a clear distinction between what the data say and my elaborations on these observations. I did not use the langauge to hide but to open up, to stimulate, to provoke, to engage in metaphoric thinking once the foundations of the analysis were set.” Manuel Castells, “A Rejoinder to Abell and Reyniers,” British Journal of Sociology 52, (Sept. 2001).

3. You want to tell a story, not just get down the facts.

A. Order the data and the analysis so that they will be meaningful and comprehensible. Turn data into information (so that the reader can gain knowledge) by showing the pattern behind the data.

B. This is done by the organization of your writing as much as by your analysis.

Need to work on linkage, dominance and subordination; consistent themes.

E. You are writing prose, not poetry: writing to do a job, not to call attention to itself.

1. See Isak Dinessen, Out of Africa. Clear writing, beautifully precise descriptions, but simple language.

F. Mark Twain‑‑"Genius is 10% inspiration and 90% perspiration".

But this was before systems of writing techniques were developed. Our approach is to see if we can raise inspiration proportion to 20% by cutting down on the perspiration.

G. Nevertheless, writing is hard, frustrating, lonely work.

Student Q: “Prof. Wellman, how do you write so easily?”

A: “I work very hard at making my writing look easy.” (Barry Wellman, September, 01)

“I do it because I have to. If you don’t have to do art, then do something else. [Kevin Cunningham, Executive Artistic Director, 3-Legged Dog {performance art group} at Bellagio Centre for Studies and Conferences, 16Nov99.

“I’m difficult when I don’t write, and I’m difficult when I do write. But at least when I write there is a reason.” [Science-fiction writer/editor Judith Merril, personal communication, about 1995].

H. The solution is to NOT wait for inspirational genius to descend but to keep writing anyway. Usually you will write servicable prose, and often you will discover your genius as you write or as you edit.

"But I keep hitting these typewriter keys. What a magician is the subconscious. If only it would work regular hours." ‑‑ Raymond Chandler. The Long Goodbye. [as quoted in an email to me by Washington Post reporter Joel Garreau, Dec 18 2000.

Your job is “to make the invisible visible through reality.” Max Beckman, painter, as quoted by Herschel Chipp, Theories of Modern Art. Berkeley: University of California Press, 1968: 188-89.

As Martha Graham once told fellow choreographer Agnes DeMille who was having a crisis of confidence: “Your job is not to question if it’s any good. Your job is to keep the channel open!”

III. KNOW YOUR AUDIENCE

A. Differences between a graduate student paper and a professional paper

1. Types of professional papers

A. Report of substantive research

B. Review essay

C. Theory piece

D. Research report for a consultant

E. Grant proposal

F. "How‑to" methods

2. Anonymous refereeing: how it works for papers and proposals.

3. Most papers don't fail because of boring results, but because:

A. Poorly focused: Don't have a clear idea of what they are writing about or who their audience is.

B. Poorly organized and repetitious: Parts out of order; circular writing.

C. Have inconsistencies between their grand theoretical claims and their more limited data.

(1) Sometimes use inappropriate (often too grandiose here, too) labels (claims) for their variables ‑‑ e.g., Siddique and Turk's use of frequency of contact as a label for density, which then was taken as a proxy for evaluating all of network analysis (CRSA, 1983).

D. Poor scholarship: not aware of latest writings/thought in the field.

B. All scientific papers are "a fraud" [Sir Peter Medawar, as quoted by John Duran, New York Review of Books, 28 April 1988]

1. They present a formal and highly idealized account of research, written according to a set of standard conventions. They don't tell about all (or any) of the false trails, bad ideas, missed‑up analyses. This is as it should be: Your job is not to provide a blow‑by‑blow account of how you actually did the research, it is to summarize what you actually accomplished. The result is a sanitized account, but not a lie.

C. AGaughin ... taught that the impression of nature must be combined with an aesthetic sense that selects orders, simplifies and synthesizes.@ Jan Verkade, “Gaughin and the School of Pont-Avens” exhibit, Montreal Museum of Fine Arts, April, 1995.

C. Know your journals

1. General vs. substantive.

2. Major to minor.

A. Acceptance rates: AJS about 15%; others may be greater than 50%. But partially, too, a core-periphery problem. Periphery may lack collegial feedback.

3. Style differences.

D. Oral presentations

1. Cut down on everything but the intellectual question, findings and conclusion.

A. Cut back literature review drastically and merge with intellectual question ‑‑ audience wants to hear YOUR findings.

2. Balance the proportions of your paper beforehand to the time you'll be aloud.

A. Many speakers are surprised when they're just getting started and the chairperson tells them they have 3 minutes to finish.

B. Outline, practice and use a stopwatch.

3. Speak from an outline, not from a written text.

A. Almost impossible not to drone on if you read from a written text, except if you're Ronald Reagan (who has to).

4. Use visual aids to break the monotony, give concrete detail

A. At the minimum, photocopied handouts.

(1) These also give the audience something to take away and think about.

B. Overhead transparencies (easy to make with a photocopier) or slides

(1) Use large print

(2) Avoid clutter.

E. Remind yourself throughout and prepare

1. Paste paragraph in front of you when you write giving title, intellectual question, and audience.

2. Find yourself paper and human role models: analyze already‑published journal articles for form (not content), get advice from veteran‑‑heavily‑published‑‑faculty members.

3. Communicate!: write to be clear for others, to sell your ideas to them.

4. Always work single-spaced -- allows you to see more lines for editing; you can always change spacing at the end for publication.

5. Use 10 to 12 point serif font (such as “Times Roman”). Better readability than sans serif.

6. Consider using a sans serif font (bold, but in same point size) for Heads and Sub-Heads. This makes the heads stand out. Capitalize only important words.

A. First level head (main ideas): Centered, Bold

B. Second level head Left Flush and Bold B on its own line.

C. Third level head: Italic, Bold, not on its own line, starts a paragraph

F.. Hermits rarely write good papers. Develop your artistic community. Talk with appropriate others about your ideas.

1.. Pablo Picasso A[I] never avoided the influence of others.@ (As quoted in Chip Sullivan, Drawing the Landscape, 2d. ed., 1997: 52).

2.. They will help you spot holes

3. Give you new ideas and fresh leads to new sources

4. Give you infectious enthusiasm.

A.. “It is important for artists to be surrounded by others who are driven and motivated.” (Sullivan, 1997: 53).

1. “Artists are nourished by each other more than by fame or by the public.” (Mike and Nancy Samuels, Seeing with the Mind=s Eye, 1975: 169).

2. “[The artist] must have the ... technical knowledge and more broadly situated knowledge of human experience in order to know what technical matters and elements of experience are most widely held to be valid and true.” Alex Rothenberg, The Emerging Goddess: The Creative Process in Art, Science and Other Fields. Chicago: University of Chicago Press, 1979: 132.

III. PARTS OF A PAPER Total length should be about 30 pp., including References (but not Tables and Figures).

"Art isn't easy.”

Every minor detail is a major decision.

Have to keep things in scale.

Have to hold to your vision."

(Stephen Sondheim, "Putting It Together," from Sunday in the Park with George [Broadway musical, 1984?]

A. Title

1. Write it early to help you focus on contents of your paper, but change it often.

2. Use it to sell your paper.

3. One good solution is to use two parts

1. The first short and snappy‑‑with a colon (:) at the end.

2. The second part should be more fully descriptive.

B. Abstract

1. Write at the end, but use it to convey your findings.

2. This way your readers will know the highlights before they start to read your paper‑‑don't make your paper a mystery novel.

3. Do a first draft by stealing key phrases from your text, especially your introduction and summary.

4. Avoid beginning with "This paper reports that sex leads to happiness." Get right to the point: "Sex leads to happiness. The more, the better. The 1,000 men and women we interviewed overwhelmingly agree."

C. Intellectual Question ‑‑ Why This is an Important Thing to Write About 1 p.

1. Where the scene is set; where the grand considerations are dealt with

2. "The very 1st sentence should contain in essence the atmosphere, the emotional content of the story and its final effect." Joseph Skvorecky, the Engineer of Human Souls, p. 68.

4. Literature Review ‑‑ Who's Done What on the Subject. 5‑10 pp.

1. One of main differences between high school and more professional paper.

2. You are not trying to show that you know everything about a subject or that you are smart (it is assumed ‑‑ and you write as if it is assumed ‑‑ that you are a competent professional).

1. On the other hand, for a student paper, you need to convince your prof. that you do know the subject comprehensively‑‑however, this still doesn't mean throwing in the kitchen sink.

3. Rather, you are trying to set the terms of the intellectual debate to which your research (reported on below) is contributing.

4. Hence work carefully to sharpen the terms of the debate but don't get hung up on details. Many people might work most profitably by only sketching out this section initially, and then filling it in after they have written up their research findings.

5. Finding the literature.

1. Use Sociofile, Social Science Citation Index, to find out who's hot in the field; what recent work is.

(1) Computerized version online best to use B checks all fields and all authors

1. If using printed version, work backwards.

2. Don't just stick with 'old masters'—although OK to search SSCI on them.

2. Use Dissertation Abstracts, Social Science and Humanities Index, Canadian Periodical Index.

6. Taking notes

1. Use 4x6 index cards, word processor, or a notetaking program

1. Upper left corner: main topic and subtopic to which it tentatively fits

2. Upper right corner: Author, date.

3. Take notes only on 1 side, so you can see everything at a glance.

4. "Be regular and orderly in your life so that you may be violent and original in your work." [Flaubert, as quoted by Mary McGarry Morris, NY Times].

5. Better to paraphrase than to quote: Shorter, tighter, makes others' ideas work for you.

1. If you do quote, must give page numbers, use exact quotation, although ellipsis .... are OK as are [explanatory] ideas.

2. One 3x5 bibliography card for each reference using standard format.

1. Author, date in upper left corner ‑‑ to link with your research card.

2. Use Endnote, Procite

(1) Boolean searches on keywords, titles, authors

(2) Do a reference only once for life

(3) Easily change reference formats for different publications

(4) Download from Sociofile, etc.

(5) Links to citations in Word, Word Perfect

(6) Can even use as a note-taker.

3. Make piles and sub‑piles of these cards.

(A) Spread out 1 sub‑pile at a time on your work‑ table.

7. Organize your review by intellectual idea, not by author.

8. 4 phases [via David Kaufer, Carnegie Mellon, The Architecture of Argument]‑‑NY Times 7 April 87

1. Summarize other authors

ii. Synthesize their ideas to find common principles

3. Analyze the merit of their positions

4. Contribute new views to the discussion

9. Avoid "Marx said...", etc

i. Better to have debates, comparisons of competing ideas which your research will address.

ii. Martha Graham:" We all steal. It's who we steal from and what we do with it that's important." (Note that I've just stolen this from her.)

10. Evaluate what they have said: Don't just let it sit there.

11. Synthesize ideas to find common principles, analyze the merit of their positions, and contribute your views to the discussion [from D Kaufer, C Giesler and C Neuwirth, The Architecture of Argument, Harcourt, Brace Jovanovich, 1987].

(1) "To me, the only good reason for writing is to try to organize my scattered thoughts of living into a whole, to relate everything to everything else." [English poet W.H. Auden, quoted in "Learning to Love One's Neighbor," Christopher Lehmann-Haupt, NY Times, February 15, 1996.]

2. Consider using an outline program (such as the one I originally wrote this on: Maxthink) to take notes and then interrelate them. Word processor will do.

(1) Fosters a good mixture of structured and intuitive thinking.

(2) Let's you either sketch out main ideas and then add details later

(3) Or fill in with detail each thing

(4) Or a mixture of the two.

3. Easy to move ideas around‑‑not the same thing as sentences.

4. Easy to flesh out topic sentences into paragraphs.

E. Another alternative is a textbase program: searches, interrelationships, moving things around.

11. Technical, consulting reports often scant, or even eliminate, this section‑‑clients want results, not scholarly debates.

E. Hypotheses or Focused Questions 2 pp.

1. In contrast to your broad intellectual question in the beginning, this is what you're 'really' going to look at.

2. Should be set up by your literature review; i.e., your literature should lead up to it in a seamless web so that it appears to the reader only logical and 'natural' that you're doing the actual research in the way you are doing it.

1. In fact, you should rewrite your literature review after you've written your hypotheses so that this is the case.

3. In many cases, you may not need/want formally specified hypotheses {avoid misplaced 'scientism'} ‑‑ you may be evaluating competing arguments (e.g., is community 'lost,' 'saved' or 'liberated'.

1. Don't just list hypotheses ‑‑ spend at least a paragraph per each backing them up, and showing how they link with other hypotheses.

2. This part very important for grant proposals (may even be a bit longer), because you don't have the results to justify your hypotheses and methods ‑‑ you have to back up even more thoroughly the soundness/reasonableness of what you want to do.

4. For example, here are some assessment criteria for getting National Health Research grants:

2. Are the objectives of the proposal clear?

3. Will answers to the questions posed, or attainment of the stated objectives, contribute to new knowledge or understanding of the subject? How?

4. Has this work been done before? If so, does it need to be done again? Why?

5. Has the proposal left out key studies ‑‑ past or present ‑‑ which may have some bearing on the importance of this proposal?

6. Are the methods adequate to meet the objectives of this proposal?

F. Research Design/Methods ‑‑ How you accomplished your research. 2‑5 pp.

1. Avoid spending much time on the obvious, i.e., what a fellow professional would be routinely expected to know ‑‑ e.g., random samples. SPSS, significance tests.

1.. These should be mentioned briefly, e.g., "We interviewed a random sample of 845 adult residents of the inner‑city Toronto borough of East York"

2. Material such as significance tests might be in a very terse note to a table or in a parenthetical statement the first time dealt with in text (in report of research findings section)

3. For a student paper, you may well want to spend more time documenting you know what you're doing. Before you are "certified," your professor (or TA) can't take for granted routine professional competence.

4. After you're a professional, you can often get away with citing your thesis ("see Wellman, 1969 for more details") or writing a technical report (published through your research centre) which goes the tedious but necessary details.

5. On the other, other hand, for grant proposals, you have to spell out much of this (for same reasons as with your hypotheses)‑‑the referees want to evaluate how sound you are, your proposal is, and they also don't have the findings to help evaluate it.

6. May vary by audience. For example, only specialized audiences have heard of multidimensional scaling or network analytic techniques.

2. Provide background context on the setting‑‑what kind of place, situation you studied.

1. Like the scene setting for a novel.

2. Varies in length depending on the exoticness of the setting ‑‑weird places may have to be described more (and defended more as being good places from which to generalize) as opposed to places about which you can assume your readers to know about, e.g., Eaton's Centre, East York.

3. Varies in length depending on kind of research. It may be just 1 paragraph for survey research; may be several pages (broken out into a separate section) for field work.

4. Varies in length depending on audience. For example, you have to tell Americans a lot more about Canada.

3. Tell readers enough so that they can understand what you did:

5. What kind of sample, how large, whether or not random, where and when collected, what kind of data collection method, any special analysis methods used.

4. Don't try to sneak by any quirks/deficiencies. Instead, discuss in common sense terms the extent to which you can generalize from your sample to the 'outer world'. Don't be embarrassed or defensive‑‑all research has deficiencies. It's nice if you can provide statistics (e.g., census data) or other material (briefly) to back up your claim to representativeness.

5. Only mediocrities, the ill‑educated and the close-minded avoid on principle either quantitative or qualitative research

1. Use the appropriate tool for the appropriate question: generalizations, subtle dynamics.

2. Consider mixing the two types‑‑sum may be greater than the parts.

6. Nice if you have a word processor when you are doing this repeatedly. Much of this is standard stuff ‑‑ 'boiler plate' ‑‑ and can be pulled in from a stored file and customized.

1. I also do New Year's letters and grant proposals this way.

7. Research proposal ends here.

G. Report of Research 10‑15 pp. +/‑

1. Basic organizational rules

1. Don't recapitulate your voyage of discovery ‑‑ readers want to learn what you found out, not how you got there, what mistakes you made, what you didn't find out.

A. For example, don't be coy ‑‑ showing something in the 1st subsection, and then in the next subsection saying things are more complicated than they originally seemed.

B. Instead, say right away in one‑half a sentence, something like: "While at first glance there seems to be a close association between the prevalence of storks and the prevalence of babies (Table 1), three‑way analysis (Table 2) shows that the true relationship is between rurality and babies."

2. Organize either in terms of independent variables or dependent variables (decide where the more interesting comparisons are). Examples:

1. Impact of domestic and paid work on different types of support: {Paid workers; Domestic workers; Double loaders}

2. Impact on types of support {Emotional, Services, Companionship, $$s} of different types of work.

3. Work from simple to complex; e.g., bivariate to multivariate, interactive relationships.

4. Use lots of subheads; perhaps sub‑subheads.

2. Text should tell the story without the tables.

1. Refer to table in text first time its used (Table 1) and every time its used thereafter when you have already switched to another table (see Table 1 above)

2. Not necessary to tell everything about a table in the text.

3. Summarize the highlights, making pertinent comparisons.

4. Assume a competent, but somewhat lazy, reader.

5. Rarely necessary to repeat numbers from tables in the text unless you are using the precise numbers themselves to make a point.

6. Be careful of loaded terms: "slightly, moderately, many, most, as much as, as little as"

1. Remember Zen question: Is the glass half‑full or half‑empty?

2. Be consistent in your adjectives.

3. Put ideas into research description. The most boring things in the world are mere summaries of tables or field notes.

1. It is OK to have carnal intercourse in this section between your findings, other peoples' findings, and theory.

1. "Unlike Fischer (1982), I find that...."

2. Do this to a limited extent for specific points; you don't want to interrupt the flow (and you'll have more time to generalize in the Conclusions section)

2. Note that it's OK (actually preferable) to use "I" if you're a single author. Save "we" for multiple authorship. It's pretentious, not modest, to use "we" if you're one person.

3. "Numbers are not the point. Numbers are used to help make a point." [Beverly Wellman, 1988]

4. Tables should be interpretable in their own right without reading the text.

1. Clear headings and subheadings, usually expressing relationships: The Effect of Paid and Domestic Work on Various Types of Social Support.

2. Use English‑language names for your variables, not SPSS or SAS eight‑letter abbreviations that you've gotten overly used to while running data.

3. Work on packing information from many tables into one table:

1. For example, the above mentioned support table may have been compiled from 4 separate crosstabs, each done for a separate type of social support.

2. Yes/no tables (or any dichotomous table) can be reduced in size to just "Percent/number saying yes", etc.

3. Correlation matrices need only be triangular, not rectangular, with perhaps the partial correlation coefficient in the other triangle.

4. Eliminate clutter, e.g.

1. Leading zeroes in correlations, regressions.

2. How many decimal points do you really need? Rarely more than one in crosstabs (I usually don't use any unless less than 10%) or two in correlations, et al.

3. Usually just need %s in one direction (comparing across categories of independent variable); omit row %s, cell %s, cell #s (these latter can be reconstructed from total N and marginal %s).

4. Must give readers enough information though that they can evaluate the accuracy of the tale you're telling in the text.

1. Use footnotes to tables sparingly, labelled "a", etc.

5. Consider using graphs instead of tables.

1. Show comparisons more carefully

2. Emphasize highlights, not petty details.

C. Don't cheat: e.g., cutting off bar graphs.

4. See Tufte, The Visual Display of Quantitative Information

E. Use software such as Excel, SPSS.

(a) Just because you can do it with software, that doesn't mean you should do it.

H. Discussion, 3‑5 pp. Really two parts

1. Summary (1‑2 pp.)

1. Recap the main findings for the reader (and for yourself).

2. Don't necessarily keep to the order in which you originally presented the findings, weave them into a coherent whole, highlighting the main points.

3. Stick fairly closely to your data here. Link back explicitly to your hypotheses, comparisons.

2. Conclusions

1. Where summary links to hypotheses, this should link to your broad intellectual questions and your literature review.

2. What are the implications of your findings for major scholarly debates.

1. And perhaps for major policy debates.

3. Implications for future research.

4. Here's your chance to play a little, too.

I. References

A. List of references, not a comprehensive bibliography.

2. Use standard form

1. See American Sociological Review, American Journal of Sociology

3. Nice to have all your references on a computer program, such as Endnote Then all you have to do is select the ones you need, reformat them to fit the style of the place you're submitting to

HOW TO WRITE -- AND EDIT -- REALLY CLEARLY

1. Learn to think in terms of multiple drafts.

1. At least three: rough, organizational shaping, final polishing.

B. "Telling a story is easy for me. Writing is easy, it's reading what you wrote that's hard." (Irving [Speed] Vogel, co‑author with Joseph Heller of No Laughing Matter, as quoted by Samuel Freedman, NY Times, 27 March 1986).

1. Get in the habit of writing a little daily, rather than trying to do a big burst at the end.

"I find writing as easy as having a shit. I do it daily." [Strangler guitarist/songwriter Hugh Cornwell in Now, Toronto, 9 April 1987].

"It seems to me, as it must have seemed to {D.H. Lawrence,} reasonable to sit down every morning and fulfill a minimal quota of 1,000 words." [Anthony Burgess, {author of A Clockwork Orange}, in Flame into Being: The Life and Work of D.H. Lawrence. Arbor House, 1985].

"Unless I write something, anything, good, indifferent or trashy, every day, I feel ill." [English poet W.H. Auden, quoted in "Learning to Love One's Neighbor," Christopher Lehmann-Haupt, NY Times, February 15, 1996.]

S.J. Perelman wrote 21 books. His "writing flows so effortlessly that it comes as a surprise to learn it was a painful, laborious process for him, and he agonized over every word. A friend recalls telephoning him once and Perelman said, `I'm in the middle of a sentence, I'll call you back when I finish.' He returned the call next day, and said, `I've just finished the sentence.' He maintained that he customarily wrote 37 drafts of each article." [Dorothy Herrmann, Toronto Globe and Mail, 2 Aug 86].

3. Use "gradual writing." Try 1st to outline your paper, but not in traditional grammar‑school way. Sketch out your heads and subheads, and write a few key phrases, sentences or paragraphs for each. You may throw most out, but it gives you a running head start to write that section, and better still, by being quick and dirty, moves you to write the lead‑ins for each part in terms of the whole (you're less likely to wander off into side issues).

1. Start at the end. Know from the beginning where you're taking your readers and give them some sign‑posts. This is neither a mystery story or a tale of someone lost in the woods.

B. This helps you handle the forest/trees problem where you are so immersed in detail you can't keep track of the overall shape and flow of your paper.

3. No real need to write draft linearly from beginning to end

Put interesting ideas that belong elsewhere [into brackets] so you won=t lose them:

A. “When I complete the [first] draft, I review it and index my bracket notes since they may contain the summaries of several additional novels that occurred to me along the way. [A word processor=s search, copy/paste functions are great for this.] A good novel is far too precious to waste; it must be caught the moment it flashes into mental view, or it will escape to the brain of some other writer who really doesn’t deserve it....My creative notions don=t have to wait their turn; they are always welcome.@ [Science-fiction writer, Piers Anthony, Pp. 311-12 in “Author’s Note to On a Pale Horse.” New York: Del Rey/Ballantine, 1983.

B. Write whichever sections you're comfortable with ‑‑ in whole or part ‑‑ knowing you'll come back and expand or polish later.

D. Software

1. Outline program good for starting. It allows you to keep control of structure, move ideas around.

2. Word Perfect best for writing and especially editing.

1. Easy to deal with formats

2. Easy to delete a word, or to end of line.

3. Converts to Word.

3. Turn off your spelling and your grammar checker until the end of the final Polishing phase.

I.F. Stone: "The computer propagates verbosity, but when it comes to making revisions, there's nothing like it." [Toronto Globe and Mail, 25 March 1988].

2. II. ROUGHING:

1. First and most important rule is not to have any style, organizational rules when you write. Get it down any way you can, and then post‑edit.

2. You can't have good ideas unless you have a lot of ideas." Linus Pauling (on Phil Donahue Show, 23 Dec 86)

1. C. Have an audience in mind, either real or precisely imagined. Don't write for yourself (masturbation) or for vaguely‑defined others. You want to have some initial sense of what tone to take, how much detail to put in.

2. Allen Ginsburg (as quoted by Linda Bamber, NY Times Book Review, 13 Dec 86, p. 40): "[Jack] Kerouac is my listening angel. Still. Even though he's dead. [But he's not the only one.] Whenever I write something witty, I think of [William] Burrough's dry laconic intelligence. Whenever I write something romantic I think of Orlovsky's great heart. When I turn a funny phrase I think of Gregory Corso."

4. See Howard Becker, Writing for Social Scientists.

3. SHAPING:

1. Edit for organization (no point in worrying about stylistic sentences, nice grammar or spelling at this stage).

2. Best if you can set the paper aside for a week ‑‑ or even a day ‑‑ so you can minimize pride of authorship, look at it with a fresh eye.

1. C. Outline the paper, writing key words in left margin of each paragraph.

2. Be more critical of your own work than anyone else could be. But don't force yourself to be unduly critical while you're actually in the act of writing. Wait until you've cooled off and gone into edit mode. Train yourself to be able to read what you've written as if someone else had written it. [hints from David McFadden]

3. This lets you find inconsistencies, redundancies; allows you to check if flow of paper makes sense; if you've given the right proportionate emphases (in terms of page length).

1. "Write with your heart, revise with your head." [London Times, 27 April 1988].

4. Try to find a trusted assessor: someone who knows subject, and a good enough friend to tell you when you have problems. [Only wimps, enemies or prospective lovers tell you it's wonderful on a 1st draft.] But remember, only you know what you want to do, and it is your name that goes on the final product.

1. Better for one person to tell you if there are problems, than to flunk, have lots of people think you're stupid. This is where authors in the periphery get into the most trouble.

2. "Writing is such a lonely business that it almost doesn't matter whether the response is positive or negative. You need to be with other people who have shared that solitude." (Phillip Lopate, as quoted by Linda Bamber, NY Times Book Review, 13 Dec 86, p. 40).

3. Perhaps the ultimate were the Brönte sisters, who had the dual advantage of brilliance and living together: "Once or twice a week, each read to the others what she had written, and heard what they had to say about it....The readings were of great and stirring interest to all, taking them out of the gnawing pressure of daily recurring cares, and setting them in a free place. It was on one of these occasions that Charlotte determined to make her heroine [in Wuthering Heights] plain, small, and unattractive, in defiance of the accepted canon." [Elizabeth Gaskell, as quoted by Linda Bamber, New York Times Book Review, 13 Dec 86, p. 40].

3. POLISHING:

1. Edit for style:

1. Carolyn Mullins, The Complete Manuscript Preparation Style Guide; A Guide to Writing and Publishing in the Social and Behavioral Sciences.

2. Here's some sample, genuine, badly‑written sentences (from a professor!). Let's see how to fix them:

A. That is, if a social structure and/or idea structure essentially mimics the pattern that is the most likely outcome of the pursuit of affective maximization, that is, if it has a form of hierarchical differentiation, then individuals pursuing affective maximization re predisposed to shape their worlds by marking them in part with such structures.

1. In addition, given our instinctual poverty, and complex cognitive tools, both of which give us a far greater ability to shape our worlds than any other species has ever had, is it not possible that certain social structures could be created that actually increase in one way or another the likelihood of such affective arousal actually occurring?

B. Basic Rules

1. Minimize words greater than 2 syllables.

1. Avoid "creative obfuscation": big, impressive sounding words to replace small ones with clear meanings. [Note that "creative obfuscation" is an example of itself: "b.s" is just as accurate.]

2. Calculate smog index. 10 sentences from front, middle and back of paper. Count all words greater than 2 syllables. Take square root and add 3. Yields approximate grade level of what you have written. Aim at grade 12 (high school grad.): Makes it more readable even for more advanced people.

3. Use thesaurus routine on computer.

4. Use style checkers on Tools menus of word processors (Grammatik, etc.)

5. Think Anglo‑Saxon, not French.

2. Make all sentences, without exception, three typed lines or less.

1. Break big ones up (compound, complex clauses) into short ones.

3. No passive voice.

1. Hunt for ‑ed phrases ("I found" vs. "It was found"). Actually, it's better to skip both and get right to your findings.

2. Examples of how passive voice avoids issues:

1. "His fall should have been broken by an elasticated bungee rope, but it became detached from the box and fell with him." [Manchester Guardian, 1987, on the death of a volunteer stuntman on a BBC TV show.]

The passive voice avoids the key question of who tied the rope badly.

2. "They have not been told of this diagnosis, for it is felt that as long as the man feels well, is happy at home and at work and his physical condition remains good, nothing should be done."

[Dr Kenneth Smith, MD, an employee of the Johns‑Manville Corp. telling the company why he and the company decided not to tell workers they had fatal asbestosis or to do anything about helping them. From Paul Brodeur, Outrageous Misconduct, Random House.]

4. Put subject and object close to each other, without long qualifying clauses intervening.

3. Some additional hints from David McFadden, Toronto poet and essayist:

1. Do not use a metaphor unless it is necessary, interesting, beautiful or amazingly funny.

2. Rid yourself of phrases such as: very, totally, absolutely, completely. [Tighter is punchier.]

4. Doing these also opens up your text: It's no longer sacred, untouchable script. You'll soon start think of other ways to tighten it up and make it more exciting to read.

1. “Editors ... are, before anything else, taker-outers, lighteners of the over-packed sentence....” [Adam Gopnick, “The Voice of Small-Town America.” {about New Yorker editors}, New York Times, Sunday Book Review, Dec 3, 2000: 44.

5. Payoff is that if you get these mechanics down, not only will your work be easier, but your thinking will be clearly, you'll see more connections, and your writing will be better to read.

6. Here’s a sample paragraph (an abstract) to edit from a doctoral student’s term paper (2001):

1. Before:

“Traditionally research on organizational learning has focused on learning from success, however, failure also plays an important role in learning. But, in order to learn from failure the influence of internal features in an organization such as its size, tasks that it performs and its age on the probability of occurrence or failures must firs be analyzed and understood. This paper examines the interactions between organization size and structure, size and the occurrence of routine task related errors and between size and age and their influence on the probability of occurrence of accidents. The implications for organization learning as a result of these influence are then discussed.”

2. Marked-Up:

Traditionally research on organizational learning has focused on learning from success, however, failure also plays an important role in learning. [“Traditionally” out of tight order. “However” should start a new sentence]

But, in order to learn from failure the influence of internal features in an organization such as its size, tasks that it performs and its age on the probability of occurrence or failures must firs be analyzed and understood. [Confusing word order. Separation of subject and object.]

This paper examines the interactions between organization size and structure, size and the occurrence of routine task related errors and between size and age and their influence on the probability of occurrence of accidents. [Confusing. Readers will get lost in the list. Doesn’t tell findings.]

The implications for organization learning as a result of these influence are then discussed.” [Delete: Empty sentence doesn’t tell reader anything new. Everyone discusses implications. At least, say what they are: “These findings suggest ....”]

3. Edited:

Although most research focuses on learning from organizational success, we can also learn from failure. For example, internal features of organizations affect the amount and kind of accidents in them. I examine the impact on accidents of such organizational features as their age, size, structure and tasks. [Note: As the paper doesn’t actually have findings, we can’t do the more useful job in the abstract of saying what age, etc. does.]

4. EPILOGUE: FROM THE PREFACE TO E.W. HOWE, THE STORY OF A COUNTRY TOWN (1884).

AI do not think a line of it was written while the sun was shining, but in almost every chapter there are recollections of the midnight bell. No one can possibly find more fault with it than I have found myself. A hundred times I have been on the point of burning the manuscript [BW: a primitive form of the DELETE AND WIPE FILE commands], and never attempting it again; for I was always tired while working at it, and always dissatisfied after concluding an evening's work. I offer this as a general apology for its many defects, and can only hope it will meet with the charity it deserves.

". . . I have changed it so often, and worried about it so much, that at its conclusion I have no idea whether it is very bad, or only indifferent. . . . I am so tired now that I am incapable of exercising my judgment with reference to it. If it prove a success or a failure I shall not be surprised, for I have no opinion of my own on the subject."

- 30 -