Scientific Writing Course Syllabus 2023
Scientific Writing Course Syllabus 2023
English B Writing
Course Materials
Spring 2023
1
Scientific Writing
The Research Paper, Spring 2023
Syllabus
I. Background Information
Instructor:
• Raymond Porter
email: rseraphim@[Link]
Course Description
This course is for graduate students who are ready to start writing a manuscript, or a
thesis chapter in the form of a manuscript. Student will learn skills needed to write,
review, and edit manuscripts in the sciences. We will cover aspects of writing a
paper for a peer-reviewed journal, including, but not limited to: organization,
structure, clarity, scope, authorship, submission and reviewing manuscripts.
Course Goals
Course Requirements
Course Material
There is no assigned textbook. Course Material will be posted on the class website.
It contains excerpts from research articles that will be discussed in class, key
vocabulary, and writing exercises to help you develop as a writer.
The Course Materials contain readings from numerous books on writing research
articles that you may wish to consult. The following are particularly relevant to
scientific writing:
2
BOOKS
General Writing
Style: Toward Clarity and Grace, Joseph Williams (University of Chicago Press)
A Manual for Writers of Research Papers, Theses, and Dissertations, Kate L. Turabian
(University of Chicago Press)
Communicating Science
Writing Science: How to write papers that get cited and proposals that get funded, Joshua
Schimel (Oxford University Press).
Essential Writing Skills for Graduate Students. Swales and Feak, 2001
Your English B grade is determined by your scores on (1) the school-wide English B
exam (50%), (2) Writing class (30%), and (3) Speaking class (20%).
Attendance: 5 points
Participation: 5 points
Total: 30 points
3
Class Schedule
Our aims
Our Assignments
Write
Precise
Simple
Necessary
Organization
Continuity (Flow)
Emphasis
Why write?
Where do we write?
What story are we telling?
4
5 October 18 Results I: Function and Structure
7 November 1 Discussion I
A Model Discussion
8 November 8 Discussion II
The Answer
Limitations
Implications
Structure of Introduction
Grammar Focus: Verb Form
5
Read: Class 11 Course Materials
Structure of Aim
Establishing Significance
Introducing Key Terms
Identifying the Big Issue
6
Ask the Experts: Advice on Writing Scientific English
Zhiyong Meng, Research Professor, Peking University
First, read well-written journal articles by native English speakers, make notes of useful
expressions, and review them from time to time. I have a document of useful expression
in different categories such as transitions, figure descriptions, numbers, equations,
comparisons, and conclusions. Writing new text becomes much easier by having a word
bank to look up useful expressions.
Second, use that word bank frequently through small writing projects such as summarizing
after reading an article, writing weekly work notes, and communicating with peers in
formal English. After drafting a piece of work, careful and multiple revisions can improve
the quality by examining if a statement can be expressed in a better way or with fewer
words by checking the word bank or previous literature.
Third, learn from mistakes. Ask someone else, especially a native English speaker, to
review your writing. Remember your weaknesses, and avoid them in the future. The
biggest difficulties I had in writing were using articles (a, an, and the) and overusing the
same words and phrases. A native English speaker can express an idea using only one
short sentence whereas I have to pile on several sentences or even a paragraph. I
learned a lot from the tracked changes or handwritten comments made to my documents
by my advisor and my colleagues when they reviewed my writing. Insist upon these from
your coauthors.
All in all, no easy way exists to improve scientific writing. Besides the above three tips,
persistence is important. It is not realistic to become a good English writer overnight. In
my experience, the best way to improve scientific writing in English is to read the
literature, practice your writing, and learn from mistakes. In time, you may find writing in
English is not that hard any more.
7
Class 1
The Goal: Clear Writing
8
The Goal: Clear Writing
The goal of this course is to show you how to use the details of a research paper to form a
comprehensible story that has a clear message. To achieve this goal, this course presents
numerous specific principles of clear writing and illustrates each principle with examples of
opaque writing followed by revisions showing how the ideas can be written more clearly.
Another feature is the exercises included for each week’s class. The exercises provide
opportunities both to recognize appropriate and inappropriate application of the writing
principles (reading exercises) and to put the principles into practice (writing exercises).
The reason for doing exercises is that application of the principles of writing requires
judgement. There are few if any “rights” and “wrongs” in writing. Rather there are better and
worse choices. To help develop your judgement in making these choices, you can compare
your critiques of the reading exercises and your writing exercising with those in the answer
keys.
9
What we will do this semester
I. The Building Blocks of Writing
Writing uses words. There are two things you can do with words---choose them and arrange
them. The second class of this course deals with choosing words. Most of the rest of the
course deals with arranging words. The arranging is in increasingly larger units of thought---
sentences, paragraphs, sections of a research paper, and the research paper as a whole.
Words, sentences, and paragraphs are the building blocks of writing. In later classes of this
course, the principles of word choice, sentence structure, and paragraph structure will be
expanded to apply to the sections of a research paper and to the research paper as a whole.
In the first section of this course, The Building Blocks of Writing, we examine how to choose
words and hot to arrange words in sentences and paragraphs. In Section II, we move to the
next larger unit of thought---the sections of a research paper.
Principles of writing each section of a research paper are based on principles of paragraph
structure. In addition, some specific principles of word choice and sentence structure that
are particularly relevant to various sections of the research paper are included in the
appropriate classes.
Before examining the principles for writing individual sections of the research paper, we
need to understand the story line that runs through the paper. The story line reflects the
scientific method. Since most research papers test hypotheses, we will focus on hypothesis-
testing papers. The basic story line in hypothesis-testing papers has four parts:
The story line is presented in a paper that has four main sections: Introduction,
Materials and Methods, Results, and Discussion. In addition, the story lines are
supplemented by the other parts of a research paper: figures and tables, title,
abstract, and references. Where in the paper each step in the story line is presented
is shown in the table below:
10
Type of Paper Step in the Story Line Part of the Paper
Hypothesis-testing A Question Introduction
Experiments Methods
Results Results
Answer Discussion
In this course, we will analyze the Introduction, Results, and Discussion sections of a
research paper. As we do, we will see how to write these story lines so that they are
clear from the first word of the Introduction to the last word of the Discussion.
11
Class 2
Word Choice
12
Read Before Class
Word Choice
Effective scientific prose is precise, clear, economical, fluent, and graceful. These qualities
depend on myriad details: the choice of words, the length and flow of individual sentences,
how sentences relate to each other, and how paragraphs are linked.
The choice of words to use in research papers is governed by a few basic principles. Words
in scientific research papers should be:
Precise.
Simple.
Necessary.
Use few if any abbreviations.
13
Words in scientific research papers should be PRECISE
Note that precise, definite, specific, concrete words evoke a mental image. For example,
“dog” evokes more or a mental image than “animal” does. Similarly, “pattern of discharge”
evokes much more of a mental image than “response characteristics” does. Words that
evoke mental images help make writing easy to read. Abstractions (such as “animal” and
“characteristics”) make reading difficult.
Example 1:
Renal blood flow was drastically compromised when the aorta
was obstructed.
Revise:
Renal blood flow was decreased by 80% when the aorta was
obstructed.
Example 2:
Animals were studied 4-9 weeks later
Revise:
POINT: Keep the name of the animal (object of study) in the reader’s mind.
14
Words in scientific research papers should be SIMPLE
Use Simple Words
One of my teachers said, “do not use a $10 word in place of a 10-cent word.” For example,
authors often use the word “utilize” because it sounds more scientific, but “use” has the
same meaning and is shorter. The same thing holds for the following pairs of complex-
simple words: perform-do, initiate-start, facilitate-cause, and propagation-move. Authors
might choose more complex words because they want to make their arguments more
complex (and more impenetrable) or because they want to impress others with their large
vocabularies. Whatever the reason, using complex words when simple ones would suffice
generally makes the writing less clear.
The point is not that big, fancy words are bad and that little, simple words are good. The
point is that you must use technical words, and these tend to be big, fancy, and heavy.
Therefore, to keep your writing from being too heavy, choose simple words for the rest of the
sentence. Simple words usually have few syllables. They are words you would say to a
child. For example, most people would not say “utilize” to a child; they would say “use.”
Example 1:
We performed a study of six-months duration on the mortality
rate of rats following exposure to elevated levels of lead. [20
words / 119 characters]
Revise:
POINT: These sentences say the same thing, yet the second one is easier to read and 20
percent shorter as well.
Example 2:
As an initial step toward understanding the relationship
between multiple trans-acting factors and GC-rich sequences,
we have isolated a cDNA clone for a factor that binds to a GC-
rich sequence.
Revise:
POINT: The technical terms (trans-acting factor, GC-rich sequences, cDNA clone) are heavy
and make reading slow. To keep the sentence light and readable, make the other words as
simple as possible. (Note that “initial” is not a “bad word”; it is just unnecessarily fancy in
this sentence.
15
Words in scientific research papers should be
NECESSARY
Omit Useless Words
Use the fewest words possible. The more noise, the less message. However, remember
that brevity is not the first principle of word choice; it is the third principle. The point is to be
as brief as possible consistent with clarity. If it takes more words to be clear, use more
words.
Example 1:
At 10 months of age, the animals received repeated daily
injections of a percentage of 3.4% cycloheximide.
Revise:
Making writing concise is as much about reducing unnecessary words as it is reducing the
complexity of words. Minimize your use of phrases that have become intimately linked to
one another so as to be cliché (e.g. meaningful dialog, time and time again, first and
foremost, last but not least).
Other phrases are simply redundant such as “smaller in size” or “model simulation.” More
examples are given in Table 2.1. Save some words, and be more creative.
Table 2.1 Redundant word combinations; words that could be eliminated are in
parentheses
Similarly, Table 2.2 lists words and expressions to avoid. One thing to notice about this
table is the large number of phrases that begin with “it”: “it has been noted that,” “it is know
that,” and “it is clear that.” These phrases are bad for two reasons. First, they add
unnecessary length. Try removing these phrases from your sentences---the meaning of the
16
sentence often will be unaffected. Second, the “it” is undetermined. “It has been
hypothesized that enhanced deposition leads to more latent heat release.” What does the “it”
refer to? Who hypothesized this? If “it” is known, reword the sentence to incorporate the
references or the first-person pronoun.
Table 2.2 Words and expression to avoid and their shorter alternatives
17
Class 3
Sentences
18
Focus on the First Seven Words of a Sentence
The key to clearly written sentences is in its first few words. When readers grasp those first
seven or eight words easily, they read what follows faster, understand it better, and
remember it longer. It is the difference between these two sentences:
Example 1:
The Federalsts’ argument in regard to the destabilization of
government by popular democracy arose from their belief in
the tendency of factions to further their self-interest at the
expense of the common good.
Example 2:
To write a sentence like Example 2, or to revise one like Example 1 into Example 2, follow
these principles:
• Avoid introducing more than a few sentences with long phrases and clauses;
get to the subject of your sentence quickly
• Make subjects short and concrete, ideally naming the character that performs
the action expressed by the verb that follows
• Avoid interrupting the subject and verb with more than a word or two
19
Avoid Long Introductory Phrase and Clauses
Compare these two sentences (introductory phrases are boldfaced, whole subjects
italicized):
Example 1:
In view of claims by researchers on higher education
indicating at least one change by most undergraduate
students of their major field of study, first-year students
seem not well informed about choosing a major field of study.
Example 2:
Most readers find the Example 1 more difficult to read because it makes them work through
a twenty-four-word phrase before they reach its subject (first-year students). In the two
sentences of Example 2, readers immediately start with a subject, Researchers, or reach it
after a very short clause, If that is so.
The principle is this: start most of your sentences directly with their subjects. Begin only a
few sentences with introductory phrases or clauses longer than ten or so words. You can
usually revise long introductory phrases and subordinate clauses into separate independent
sentences as in Example 2.
20
Make Subjects Short and Concrete
Readers must grasp the subject of a sentence easily, but they can’t when the subject is long,
complex, and abstract. Compare these two sentences (the whole subjects in each are
italicized; the one-word simple subject is boldfaced):
Example 1:
A school system’s successful adoption of a new reading
curriculum for its elementary schools depends on the
demonstration in each school of the commitment of its
principal and the cooperation of teachers in setting reasonable
goals.
Example 2:
In Example 1, the whole subject is fourteen words long, and its simple subject is an
abstraction---adoption. In Example 2, the clearer version, the whole subject of every verb is
short, and each simple subject is relatively concrete: school system, each principal, she,
teachers. Moreover, each of those subjects performs the action in its verb: system will
adopt, principal demonstrates, she is committed, teachers cooperate.
The principle is this: readers tend to judge a sentence to be readable when the subject of
its verb names the main characters in a few concrete words, ideally a character that is also
the “doer” of the action expressed by the verb that follows.
21
Avoid Interrupting Subjects and Verbs with More than a
Word or Two
Once past a short subject, readers want to get to a verb quickly, so avoid splitting a verb
from its subject with long phrases and clauses:
Example 1:
Some economists, because they write in a style that is
impersonal and objective, do not communicate with laypeople
easily.
In Example 1, the because clause separates the subject some economists from the verb do
not communicate, forcing us to suspend our mental breath. To revise, move the interrupting
clause to the beginning or end of its sentence, depending on whether it connects more
closely to the sentence before or the one after. When in doubt, put it at the end.
Example 2:
Example 3:
Example 4:
22
Put Key Actions in Verbs, Not in Nouns
Readers want to get to a verb quickly, but they also want that verb to express a key action.
So, avoid using an empty verb such as have, do, make, or be to introduce an action buried
in an abstract noun. Make the noun a verb.
Compare these two sentences (nouns naming actions are boldfaced; verbs naming actions
are capitalized; verbs expressing little action are italicized):
Example 1:
During the early years of the Civil War, the South’s attempt at
enlisting Great Britain on its side was met with failure.
Example 2:
In Example 1, three important actions aren’t verbs but nouns: attempt, enlisting, failure.
Example 2 seems more direct because it expresses those actions in verbs: attempted,
enlist, failed.
23
Put Information Familiar to Readers at the Beginning of a
Sentence, New Information at the End
Readers understand a sentence most readily when they grasp its subject easily, and the
easiest subject to grasp is not just short and concrete but also familiar. Compare how the
second sentence in each of the following passages does or doesn’t contribute to a sense of
“flow””
Example 1:
New questions about the nature of the universe have been
raised by scientists studying black holes in space. The
collapse of a dead star into a point perhaps no larger than a
marble creates a black hole. So much matter squeezed into so
little volume changes the fabric of space around it in odd
ways.
Example 2:
Most readers think Example 2 flows better than Example 1, partly because the subject of the
second sentence, A black hole, is shorter and more concrete than the longer subject of
Example 1: The collapse of a dead star into a point perhaps no larger than a marble. But
Example 2 also flows better because the order of its ideas is different.
In Example 1, the first words of the second sentence express new information:
… black holes in space. The collapse of a dead star into a point perhaps no larger than a
marble creates…
Those words about collapsing stars seem to come out of nowhere. But in Example 2, the
first words echo the end of the previous sentence:
Moreover, once we make that change, the end of the second sentence introduces the third
more cohesively:
… the collapse of a dead star into a point perhaps no larger than a marble creates. So much
matter squeezed into so little volume changes…
Contrast Example 1; the end of its second sentence doesn’t flow into the beginning of the
third as smoothly:
24
The collapse of a dead star into a point perhaps no larger than a marble creates a black
hole. So much matter squeezed into so little volume changes the fabric of space around it in
odd ways.
That is why readers think that the passage in Example 1 feels choppier than that of Example
2: the end of one sentence does not flow smoothly into the beginning of the next.
The corollary of the old-information-first principle is to put new information last, especially
new technical terms. So, when you introduce one, put it at the end of its sentence. Compare
these:
Example 3:
Example 3:
In Example 3, the new technical terms are calcium blockers, sarcomeres, filaments, the
proteins actin and myosin, but they first appear early in their sentences. In contrast, in
Example 4, those new terms first appear toward the ends of their sentences. After that,
they’re old information and so can appear at the beginning of the next sentences.
No principle of writing is more important than this: old before new, familiar information
introduces unfamiliar information.
25
Class 4
Paragraphs
26
Paragraph Structure
Even if a paper has perfect word choice and perfect sentence structure, it can be difficult to
understand if the paragraphs are not clearly constructed. Each paragraph must be
constructed to tell a story. Readers should be able to recognize the message and follow the
story of each paragraph whether or not they understand the science.
• The continuity, that is, the relationship between ideas, must be clear.
27
Organization: Topic Sentences and Supporting Sentences
A paragraph is a number of sentences on a single topic. The goals of a paragraph are to get
a message across and to make the story behind the message clear. These goals can be
accomplished in various ways, but the general approach that is clearest for most readers
most of the time is to provide an overview first and then to give details. The strategy behind
this approach is to create an expectation and then fulfill it. In this class, we will call this a
point-first paragraph.
The reverse strategy is used in scientific papers: here are some details; here is what they
mean. We will call this approach a point-last paragraph (this will be discussed later in
the semester).
The classic way to give an overview first in a paragraph is to write a topic sentence. A topic
sentence is a sentence that states the topic or the message of the paragraph. The topic is
what the paragraph is about. The message is the point the paragraph is making. To identify
the topic, use a key term. To state a message, use a verb, along with a subject.
It is clearest to have only one message per paragraph. Including more than one message in
a paragraph makes the paragraph complex and more difficult to understand.
Details that support the topic sentence are written in the remaining sentences of the
paragraph---the supporting sentences. The supporting sentences should be organized in a
logical way that explains the message of the paragraph.
The examples below all begin with a topic sentence. Each topic sentence names a topic and
states a message that make one (and only one) clear point. In addition, the supporting
sentences are organized logically to explain the message stated in the topic sentence, thus
fulfilling the expectation that the topic sentence creates.
Example 1:
1 There are three different theories put forward for the very slow
relaxation of catch muscles of mollusks. 2 One theory holds that catch
is due to some unusual property of myosin in these muscles that
produces a slow rate of detachment (12). 3 In this theory, paramyosin
would have no special role beyond that of providing the long
scaffolding on which the myosin is positioned as well as the
mechanical strength for the large tensions developed. 4 The second
theory holds that tension is developed by actin-myosin interaction but
is maintained by paramyosin interactions (13, 14). 5 Because the thick
filaments are of limited length, interaction would have to occur through
fusion of thick filaments (15). 6 A third theory, to which I subscribe,
pictures a structural change in the paramyosin core affecting the rate
of breaking of myosin-actin links at the filament surface (5, 16).
Example 1, which is from an Introduction to a journal article, has a topic sentence and
logically organized supporting sentences. The topic of the paragraph is the three theories.
The message stated in the topic sentence is that three theories exist.
28
To support this topic sentence, the supporting sentences are organized in a simple pattern---
a list. In this list, one or two supporting sentences briefly describe each theory. This list of
brief descriptions fulfills the expectation created by the topic “three different theories” in the
topic sentences.
The order for this list is not random, but proceeds from least to most important. That is, the
pattern of organization is listing details from least to most important. Thus, the author first
describes the two theories she rejects and then describes the theory she accepts, which for
the purposes of her paper, is most important. The reasons for her rejection are implied in the
extra sentence about each of the first two theories (sentences 4 and 5). The rest of the
paper goes on to explain the third theory at greater length.
Example 2:
In Example 2, which is from the Results section of a journal article, the topic of the
paragraph is pulmonary nerve endings (the subject of the topic sentence). The message
stated in the topic sentence). The message stated in the topic sentence is that pulmonary
nerve endings were relatively insensitive to a drug (sated in the verb and object).
If the word “relatively” were omitted, no supporting sentences would be needed. The reader
could look at the table and figure and see the data showing that the nerve endings were
insensitive. Thus, the expectation created by the topic sentence is to hear what it means for
the pulmonary nerve endings to be relatively insensitive. The first two supporting sentences
fulfill this expectation by stating that <50% of the nerve endings were stimulated at all, and
most of these responded weakly.
Why, then, does the paragraph continue? The reason is that some data do not support the
message “relatively insensitive,” and these data cannot be ignored. Thus, the supporting
sentences are organized according to the type of evidence: first, evidence that supports
insensitively; second, contradictory evidence, which does not support insensitivity.
The contradictory evidence is not merely stated, leaving the reader to decide whether the
supporting or the contradictory evidence is stronger and thus whether the pulmonary nerve
endings are indeed insensitive. Instead, the author argues against the contradictory
evidence by pointing out that the dog was atypical. This argument against the contradictory
29
evidence makes it clear that the supporting evidence outweighs the contradictory evidence
and thus that the pulmonary nerve endings are indeed relatively insensitive.
Related patterns of organization are paragraphs that are all pro or all con. Which pattern of
organization you use---pro-con, all pro, or all con---depends on the point you are making and
the quality of the evidence you have.
Other commons patterns of organization are funnel (used in the Introduction), chronological
order (used in methods and results), problem-solution, and solution-problem (used in
methods papers)
30
Continuity (or Flow)
Even if a paragraph is well organized---that is, has a topic sentence and logically organized
supporting sentences and is not missing any steps in the logic, the story of the paragraph
can be difficult to follow if the paragraph does not have continuity. Continuity is the smooth
flow of ideas from sentence to sentence (and from paragraph to paragraph).
The essence of continuity is a clear relationship between every sentence and the
sentence before it. To create continuity, special techniques can be used. Six
important techniques for creating continuity are
The first two of these techniques are the most frequently used and thus are the most
important. We will limit our attention to these two techniques in this class, but will
revisit the remaining techniques throughout the semester.
Key terms are terms that name important ideas in a paper. Key terms can be technical
terms, such as “G-protein” or “mitogenesis,” or nontechnical terms such as “increase” or
“function.” Repeating key terms from sentence to sentence (and from paragraph to
paragraph) is the strongest technique for providing continuity.
For Clarity. The best advice for clear continuity is to repeat key terms exactly. If a key term
is not repeated exactly and instead another term is used, a mental manipulation is needed to
see the relationship between the two terms. Readers familiar with the field can usually see
the relationship; those less familiar may not.
Example 1:
Digitalis increases the contractility of the mammalian heart.
This change in inotropic state is a result of changes in calcium
flux through the muscle cell membrane.
What is “inotropic state”? How does it relate to the previous sentence? The answer is that
“contractility” and inotropic state” mean the same thing. If no difference in meaning is
intended, why use two different terms and risk confusing some readers?
31
Revision:
In this revision, the continuity is clear to all readers because the key term “contractility” is
repeated exactly. In addition, the key term “increases” is repeated, rather than being
generalized to “change.” Repetition of two key terms makes the continuity even stronger.
Repeated Key Term as the Subject. Continuity is clearest if the key term is repeated
early in the sentence. If the key term is delayed until the end of the sentence, the continuity
is broken and the reader is kept in suspense temporarily. Compare the following sentences:
Example 2:
Digitalis increases the contractility of the mammalian heart.
Changes in calcium flux through the muscle cell membrane
cause this increased contractility.
Revision:
In the revision, the subject (“changes…”) and the object (“this increased contractility”) are
switched, thus moving the repeated key term to the beginning of the second sentence as the
subject of the sentence.
Repeated Key Term as an Aspect of the Subject. So far, to repeat the key term
early, we have repeated the key term at the beginning of the sentence as the grammatical
subject of the sentence. Another possibility is to make the subject of the sentence an aspect
of the key term, as In Example 3. The key term is repeated immediately after the subject.
Example 3:
Signals that confer localization to the endoplasmic reticulum
(ER) have been characterized in the cytoplasmic domain of
many mammalian type I transmembrane proteins that reside
in the ER and in the ER-Golgi intermediate compartment. 2
One common feature of these signals is the presence of two
lysine residues at positions -3 and -4 from the C-terminal end
of the cytoplasmic domains (refs).
In example 3, the repeated key term, “signal,” is not the subject of the second sentence.
Instead, an aspect of “signal” ---“one common feature” ---is the subject, and the key term
(“signal”) is repeated after the subject. The story in this paragraph moves from signals to a
32
common feature of the signals. Because the key term from sentence A is repeated early in
sentence B, the reader can follow the story easily.
Whether the key term is repeated early as the subject of the sentence or immediately after
the subject, the crucial point is that the relationship between the new sentence and the
previous sentence is clear at the beginning of the new sentence.
For a paragraph to have continuity, the reader must understand not merely what each
sentence says, but also why that author is writing each sentence, and why at this point in the
paragraph; how does this sentence relate to the story? In some cases, the relationship is
obvious, as in Example 4:
Example 4:
1 Neuritic plaques and neurofibrillary tangles in brain tissue
are major features of the pathology of Alzheimer’s disease. 2
Neuritic plaques are rich in an amyloid that consists largely of
the 39- to 43-residue amyloid beta-peptide (A4), a proteolysis
product of the beta-amyloid precursor protein (betaAPP) (ref).
In Example 4 the first sentence talks about neuritic plaques and neurofibrillary tangles. The
second sentence gives more information about neuritic plaques. Continuity is provided by a
repeated key term (“neuritic plaques”).
If the reason for putting one sentence after another is not obvious, you need to indicate how
the sentences are related. The technique for indicating how sentences (or parts of
sentences) are related is transitions.
Transition Words
Transition words are standard terms that indicate standard logical relationships between
ideas. Examples include “therefore” and “thus” (conclusions), “for example” (example), “first”
(sequence), “in addition” (addition), “in contrast” (contrast), and “however” (difference). Thus,
transition “words” can be phrases, such as “for example,” “in contrast,” and even “on the
other hand.”
To see how important transition words are in guiding your understanding of sentences and
paragraphs, read each of the following examples both with and without the underlined
words:
The lymphocytes that infiltrate the alveolar walls in this rejection phase are
likely to be conveyed by the blood, because they infiltrate all alveolar walls
synchronously all over the lungs. (reason)
33
Although individual residues in the repeated-sequence blocks in the core have
diverged, the patterns of amino acids are identical. (concession)
1 By widening our focus to the entire trachea, we were able to see that most
ganglion cell bodies (72%) are located in the neural plexuses associated with
the trachealis muscle and submucosal glands, and only a small proportion
(28%) are located along the longitudinal nerve trunks. 2 Furthermore, we were
able to see that most of the ganglia in the superficial muscle and gland
plexuses contain only 1-4 ganglion cell bodies (average, 2.8 ganglion cell
bodies). 3 Thus, previously reported ganglia along the longitudinal nerve trunk
that contain 10-20 ganglion cell bodies are not typical of most tracheal ganglia.
In Example 5, if any of the underlined words were omitted, the logical relationship would be
difficult to see, and the story of the paragraph would be difficult to follow. For example, if
“because” were omitted from the first sentence in Example 5, the reader might not be able to
figure out that “they infiltrate…” is the reason that lymphocytes are likely to be conveyed by
the blood, or might not. But the point is that the reader should not have to construct the story
of the paragraph. It is the writer’s job to make the story clear.
In the next sentence in Example 5, if “therefore” were omitted, the logic would be destroyed.
A reader cannot be expected to invent a cause-effect relationship where only “and” is
written. So, if undergoing proteolysis is a consequence of being synthesized as a protein,
the transition word “therefore” must be included.
The story of a paragraph (or a paper) is not simply what the sentences are saying. It is also
what the sentences are doing, that is, what their function is---giving a reason, adding a
detail, concluding, or whatever. The reader must understand both what each sentence says
and what its function is in order to understand the story. That is why transition words are so
important.
34
COMMON TRANSITIONAL “WORDS” FOR SCIENTIC WRITING
Sequence Time
again, and, besides, then, further, while, since, simultaneously, presently,
furthermore, next, moreover, in addition, meanwhile, thereafter, thereupon,
first, second, third, etc.; (a), (b), (c), etc.; 1), afterward, at the same time, next,
2), 3), etc.; following this, subsequently, to sometimes, in the meantime, eventually,
enumerate, also, another, last, plus following this, later, usually, occasionally,
concurrently, preceding this, as, presently,
Comparison and contrast at the time of this writing, often, rarely,
at the same time, on the contrary, in throughout, by, at, during, continuing
contrast, nevertheless, notwithstanding,
nonetheless, conversely, like, unlike, even Cause and effect
so, in the same way, as, unless, whether, therefore, thus, consequently, as a
though, even though, regardless, consequence, for this reason, hence,
irrespective, otherwise, in comparison to, accordingly, because, due to, in spite of,
even when, to the contrary, but, or nor, yet, despite
inasmuch, contrary to, comparing,
alternatively, rather, despite, ironically Emphasis
surprisingly, indeed, interestingly, curiously,
Examples in fact, of course, naturally, evidently,
for example, for instance, in the case of, in certainly, clearly, obviously, apparently,
general, especially, if, specifically, in fortunately, especially, significantly,
particular, generally, on this occasion, in perhaps, from my perspective, if possible, if
this situation, to illustrate, to demonstrate, so, basically, in reality, essentially
as an illustration, as a demonstration,
unless, such as, provided that, once again, Concluding
another example, a further example, a finally, therefore, in summary, to conclude,
further complication, in such cases, in this in conclusion, to summarize, as I have
way, in some of these cases, for these shown, hence, thus in other words, as said
reasons, one way, another way, as earlier, in any case, as a result, at least, as
discussed, using, particularly, that is, more mentioned above, as said previously,
specifically, except thereby, in the present article, simply put
35
Emphasis
So far, we have been concentrating on organizing paragraphs and providing continuity so
that the reader can find the message of the paragraph and follow the story. In addition, the
reader needs to know what is important. Not all the information in a paragraph, or in a paper,
is equally important. To help the reader find the important information, emphasize the
important information and de-emphasize less important information.
We will touch on each of these techniques throughout the class, but in this class, we will
focus on condensing / omitting information and power positions.
Condensing or Omitting
Possibly the most important technique for emphasizing important information is to de-
emphasize less important information by condensing or omitting it. In many published
papers, the unimportant information outweighs the important information, so we cannot see
the forest (message) for the trees (details). The solution is to find a balance between the two
extremes: telling everything you know and getting a message across. Keep in mind, “The
more noise, the less message.”
Power Position
The middle position is the burial ground. Thus, an important point expressed in the middle of
a sentence or an important result placed in the middle of several sentences of results is
almost invisible.
36
Power Positions
Weak Positions
Middle of a sentence, paragraph, section
Words in parentheses
Adjective or noun in the middle of a sentence
Example 6 presents a list of results in numbing detail. It is difficult to tell which results are
most important:
Example 6:
1 Mean pulmonary artery pressure and cardiac output did not change after
instillation of serum alone or serum with epinephrine or terbutaline (Table 1). 2
Left atrial pressure fell slightly below the baseline after all three treatments, but
the decrease was statistically significant only after epinephrine (Table 1). 3
Peak airway pressure increased slightly after all three treatments, but the
increase was statistically significant only for epinephrine and terbutaline (Table
1). 4 There was a significant increase in lung lymph flow and a significant
decrease in the lymph-to-plasma protein concentration ratio after all three
treatments. 5 Both the rise in lymph flow and the decrease in the lymph-to-
plasma protein concentration ratio were greater after terbutaline and
epinephrine than after serum alone (Table 2). 6 Arterial oxygen tension
decreased after all three treatments, although it was always greater than 85
mmHg.
The most important results---sentence 4---is buried in the middle. To make the important
result easier to find, it should be first in the paragraph.
Revision:
1 After serum was instilled either alone or with one of the two beta-adrenergic
agonists, lung lymph flow increased and the lymph-to-plasma protein
concentration ratio decreased. 2 Both of these changes were greater after
terbutaline and epinephrine than after serum alone (Table 2). 3 Arterial oxygen
tension decreased, although it was always greater than 85 mmHg. 4 There
were no important changes in hemodynamics or peak airway pressure (Table
1).
In the revision, the most important result is first. The three sentences of least important
results are condensed to one sentence and placed last (sentence 4). In addition, two other
techniques of condensing are used: the treatments are named at the beginning of the
paragraph, so they do not need to be repeated in every sentence (as they are in the original
paragraph). A category term (“changes”) is used in sentence 2 to avoid repeating the details
of what increased and what decreased.
37
Now that the important result is in a power position (first) and the noise-to-message ratio is
decreased, the results in this paragraph are clear.
38
Class 5
Telling a Science Story
39
An Overview of Research Paper Structure
Because the problem with most research papers is that they lose the forest for the trees
(“the main idea is lost in the details”), the solution is to build a structure into the paper so that
the forest is clear. Each of the four parts of the text of a research paper has its own
structure. In this course, we will analyze and write three of these sections.
Introduction
The Introduction follows a standard structure: the funnel. A funnel starts broadly and then
narrows. Thus, in a hypothesis-testing paper, the Introduction funnels from something
known, to something unknown, to the question the paper is asking. The Introduction may
end with the question or may go on to state the experimental approach to answering the
question. An example of a brief Introduction that has a funnel shape is given below.
Example 1 Introduction
An important detail to notice in this Introduction is that the key terms in the question
(italicized) repeat words in the statements of what is known and unknown. Repeating key
terms is important because the repetition makes it obvious that the question follows
inevitably from what is known and unknown. It is important that the question follows
inevitably and is stated clearly because the rest of the paper depends on the question.
Specifically, Methods tells what experiments you did to answer the question, Results tells
what results you found that answer the question, and the Discussion states and explains the
answer to the question. To avoid losing the forest for the trees in a research paper, the trick
is to use the question as the touchstone for selecting and organizing ideas in each section of
the paper.
Results
In the Results section, the overall structure is normally chronological. In the Results section
of a hypothesis-testing study that design all experiments in advance, the structure may also
be in the order of most to least important to the question (for example, when the experiments
were done simultaneously).
In addition, within each paragraph of the Results section, the ideas can be organized from
most to least important. Thus, an important result is stated in the first sentence, and less
important results and supporting details are stated in later sentences, as in Example 2.
Example 2 Results
1 Incubation rings of fetal lamb ductus arteriosus in arachidonic acid increased production of
prostaglandin E2 to 3.5 times the baseline value (Fig. 1). 2 This increase was blocked when
40
the rings were incubated in arachidonic acid in the presence of indomethacin. 3 In the
control series of experiments, prostaglandin E2 production measured at the same 90-min
intervals did not change.
Discussion
For the Discussion section, there is no prescribed structure. However, there are some
general guidelines. The first and most important guideline is to state the answer to the
question at the beginning of the Discussion. The reason for stating the answer at the
beginning is that the answer is the most important statement in the paper, so it should
appear in the most prominent position: first. Immediately after stating the answer, give
supporting evidence. An example of the first paragraph of a Discussion in which the first
sentence states the answer to the question and the rest of the paragraph gives supporting
evidence is given in Example 3.
Paragraph 1: 1 In this study, we have shown that a 42-day course of dexamethasone leads
to sustained improvement in pulmonary function and improves neurodevelopmental outcome
in very low birth weight infants who are at high risk of developing bronchopulmonary
dysplasia. 2 Evidence of improved pulmonary function is that after a 42-day course of
dexamethasone given to our preterm infants who were ventilator and oxygen dependent at 2
weeks of age, the durations of positive pressure ventilation, of supplemental oxygen, an of
hospitalization were less than those in control infants, who received saline placebo. 3
Evidence of improved neurodevelopmental outcome is that the infants who received the 42-
day course of dexamethasone had a lower incidence of neurologic handicap and
significantly higher scores on the Bayley Scales of Infant Development than did infants in the
control group.
After stating and supporting the answer, organize the remaining topics either according to
the logic of the science or else in the order of most to least important. Indicate the
organization by using topic sentences to state the point of each paragraph. The reader
should be able to read the first sentence of every paragraph in the Discussion section and
follow the story of the Discussion, as in Example 4, which continues the discussion begun in
Example 3.
Paragraph 3 However, some infants may have had adrenocortical suppression, since mean
serum cortisol levels were significantly lower in infants who received the 42-day course of
dexamethasone than in control infants. (etc.)
Paragraph 4 We have also found that the duration of dexamethasone therapy is important.
(etc.)
41
Paragraph 5 Two points regarding the clinical courses of infants in our study are worth
noting. First, the only two infants who developed pneumothoraces during the study period
were receiving dexamethasone. (etc.)
Paragraph 6 Second, retinopathy was found in a very high number of infants in all three
groups. (etc.)
The middle of this Discussion moves from most to least important topics. After the answer
to the question is stated at the beginning of the Discussion (that the therapy is beneficial,
see Example 3), the middle of the Discussion goes on first to comment on serious
complications, which, if present, would undermine the therapy, then to explain why a long
treatment (42 days) is needed, and finally to explain points not asked in the question that are
interesting but less important (other complications). This story line is clear from reading the
topic sentences.
The Discussion cannot just stop. It must clearly come to an end. Two standard ways of
ending are to restate the answers and to indicate the importance of the answers, or you can
do both. For the Discussion in Example 3 and 4, the author restated the answers and also
the point about complications, thus puling the message of the paper together (Example 5).
Paragraph 7: 1 In summary, we have shown that dexamethasone therapy for 42 days leads
to sustained improvement in pulmonary function and improves neurodevelopment outcome
in very low birth weight infants who are ventilator and oxygen dependent at 2 weeks of age
and therefore are at high risk of developing bronchopulmonary dysplasia. 2 Although
dexamethasone use may be associated with adrenocortical suppression, it is not associated
with an increased incidence of major complications, including infection, hypertension, and
growth failure.
This ending reinforces the message of the paper and feels conclusive.
Note that the answer in the final paragraph in the Discussion is virtually identical to the
answer in the first paragraph. This exact repetition of answers is important. If the answers
were different, we would not know which one to believe.
Finally, it is important that the answer answers the question asked. The question focuses
the entire paper for the reader. If the answer does not answer the question asked, the reader
will be confused as in Example 6.
In this example, the answer actually does answer the question, but that is not obvious
because the key terms are different. To make clear that the answer answers the question
42
asked, and thus to make the message of the paper clear, use the same key terms in the
question and answer, as in the revision of Example 6.
Revision
The rest of this course explains in greater detail how to follow these guidelines so that you
will be able to write research papers in which both the trees (details) and the forest (main
message) are clear.
43
Results
44
Class 6
Results: Function and Organization
45
The Results
Functions
The function of the Results section is to state the results of the experiments described in the
Materials and Methods section. In addition, the Results section directs the reader to figures
or tables that present supporting data.
Story Line
The story line of a paper consists of the Question, Experiments done, Results found, and
then the Answer to the research question.
The Results section continues the story line in different ways in the two types of hypothesis-
testing study. For hypothesis-testing studies in which all the experiments are designed in
advance, the Results section describes the third step in the story line: the results. For
hypothesis-testing studies in which the results of one experiment determine what the
next experiment will be (Sequential Results section), the Results section describes both
the second and the third steps of the story line: the experiments done and the results found.
Step in the Story Line All Experiments Designed in One Experiment Determines
Advance the Next
Content
The primary information in the Results section is results. However, not every result that you
obtained from your experiments or observations needs to be reported in the Results section.
The Results section should report only results pertinent to the question posed in the
Introduction. Results should be included whether or not they support your hypothesis. Both
experimental and control results should be included.
In addition to presenting results, the Results section can include data. However, most data,
and in particular the most important data, should be presented in figures or tables, where the
data are highly visible and easy to read.
Normally, the Results section does not include statements that need to be referenced, such
as comparisons with others’ results. However, if a brief comparison (one or two sentences)
would not fit smoothly into the Discussion, it can be included in the Results section.
46
The results section of one type of study---hypothesis-testing studies in which the results of
one experiment determine what the next experiment will be --- also includes questions, study
design (overview of experiments done), answers, and, as needed, background, purposes,
and reasons in the Results section.
Results are different from data. Data are facts, often numbers, obtained from experiments
and observations. Data can be raw (for example, all phospholipid concentrations measured
during an experiment), summarized (for example, mean and SD), or transformed (for
example, percent of control). Results are general statements that interpret data (for
example, “Propranolol given during normal ventilation decreased phospholipid
concentrations”).
Data can rarely stand alone. The result (= the meaning of the data) must be stated.
For example, what is the reader supposed to think after reading the two sentences of data in
Example 1?
Are the data similar? Different? What is the point? The purpose of the Results section is to
make the point clear. To make the point clear, state the result first and then present the
data, as in Revision 1A below, or (better) cite a figure or a table.
The mean resting blood pressure was higher in the 30 tennis players
than in the 20 control subjects [94 ± 5 (SD) mmHg vs. 85 ± 5 mmHg, P
< 0.02].
The sentence now states the result. The data are given in parentheses after the result. (A P
value for statistical significance is added to provide evidence that the difference was not
likely to have occurred by chance.) However, in most cases the data should be presented in
a figure or a table rather than in the text.
In addition to simply saying “was less than,” “was greater than,” “decreased,” or “increased,”
you can, when appropriate give a general idea of the magnitude of a difference or a change
by using a percentage, as in Revision 1B.
The mean resting blood pressure was 10% higher in the 30 tennis
players than in the 20 control subjects [94 ± 5 (SD) mmHg vs. 85 ± 5
mmHg, P < 0.02].
47
This statement of the result (“was 10% higher”) gives a simpler and therefore clearer idea of
the magnitude of the difference than do the data alone (94 ± 5 (SD) mmHg vs. 85 ± 5
mmHg).
Organization
For studies in which all experiments are designed in advance, and which therefore include a
Study Design subsection in Methods, the Results section simply describes the results, one
topic per paragraph. The results can be organized either chronologically or from the most
to least important. For organization from most to least important, put results that answer the
question at the beginning of the Results section or at the beginning of paragraph 1 and
successive paragraphs.
In Results Excerpt 1, the variables were measured simultaneously, so the Results section
is organized from most to least important.
Study Design: To answer this question, the authors performed a complete Doppler
echocardiographic examination in each of 28 patients and 20 control
subjects and compared the results with those from a similar study done
5 years earlier. The variable used as an indicator of pulmonary
hypertension was pulmonary artery pressure. In addition, pulmonary
vascular resistance was measured as the possible cause of the increase
in pulmonary artery pressure.
Results
6 Of the 5 patients who had pulmonary hypertension at the first study, 2 died at the time of
the second study, 1 had persistent pulmonary hypertension, and 2 had normalized systolic
pulmonary artery pressure. 7 Of the 12 patients who had pulmonary hypertension at the
second study, 11 had had normal pulmonary artery pressure at the first study. 8 The mean
increase in pulmonary artery systolic pressure for these 11 patients was 9.4 mmHg.
48
9 Pulmonary Vascular Resistance. 10 Pulmonary vascular resistance also increased in the
lupus patients, from a mean of 5.1 mmHg/L/min in the first study to a mean of 7.1
mmHg/L/min in the second study (Fig. 2). 11 In the second study, mean pulmonary vascular
resistance was higher in the lupus patients than in the controls (7.1 vs. 5.6, p< 0.005). 12
Additionally, in the 12 patients with pulmonary hypertension at the second study, pulmonary
vascular resistance was higher than in the 15 patients who had normal pulmonary artery
pressures (Table 3). 13 Of the two patients whose pulmonary artery pressure normalized
from the first to the second study, one had no change in cardiac output and a decrease in
total pulmonary vascular resistance (from 6.0 to 3.9 mmHG/L/min), and the other had a
decrease of 2.3 L/min in cardiac output.
This Results section, for which the entire experiment was designed in advance, is organized
from most to least important: first the indicator for the dependent variable in the question
(paragraphs 1 and 2); then another dependent variable (paragraph 3).
Similarly, within each subsection, the results are organized from most to least important.
First are the results for the variables (sentences 2, 3, and 9). Then come supporting details
(4,5; 11,12) and details for individual patients (6-8; 13).
49
Studies in which one experiment determines the next experiment
For studies in which the results of one experiment determine what the next experiment will
be, the Results section is organized in a repeating four-part pattern. Ideally, each repeat of
the pattern is in a separate paragraph. If necessary, background indicating why a question
was asked and purposes or reasons explaining why an experiment was done are also
included. The four-part pattern is:
Question
Overview of experiments
Results
Answer to the question
The question in the first paragraph of the Results section is the question of the paper. The
questions of the succeeding paragraphs are questions asked as steps toward answering the
question of the paper.
The sequence of overview of the experiments running from the first to the last paragraph of
the Results section constitutes the Study Design.
The answer stated after each result leads to the next question, thus triggering the next four-
part pattern (that is, the next paragraph) of the Results section. An answer at or near the
end of the Results is the answer to the question asked in the paper.
Results Excerpt 2: Studies in which one experiment determines the next experiment
Question: Whether the nematode gene ceh-22 and the vertebrate gene nkx2.5
perform similar functions
Experimental Examination of the ability of zebrafish nkx2.5 gene to substitute for the
Approach: nematode ceh-22 gene in transgenic Caenorhabditis elegans.
9 We next asked whether Nkx2.5 directly interacts with the same sequences recognized by
CEH-22. 10 To answer this question, we examined expression of a reporter gene under the
control of multimerized CEH-22 binding sites. 11 CEH-22 binds a region within the myo-2
enhancer termed the B subelement (ref). 12 In wild-type animals, a lacZ reporter under
control of a synthetic enhancer consisting of four copies of a 28-bp B sub-element
oligonucleotide is expressed specifically in pharyngeal muscle; only occasional expression is
observed outside the pharynx (Table 1; ref). 13 In a transgenic strain bearing the unc-
54::nkx2.5 expression construct, we found a significant increase in the number of animals
expressing ß-galactosidase in body wall muscle (from 2.5 to 16.5%)(Table 1; Fig. 1C). 14 To
rule out the possibility that Nkx2.5 was indirectly increasing expression of myo-2 or the B
50
sub-element reporter by activating ectopic expression of the ceh-22 gene, we examined
expression of a ceh-22::lacZ fusion in animals bearing the unc-54:nkx2.5 transgene. 15
Expression of ß-galactosidase was limited to pharyngeal muscle (Table 1), a pattern
identical to that observed in wild-type animals (ref). 16 Thus, Nkx2.5, like CEH-22, activates
transcription by interacting directly with the B sub-element of the myo-2 enhancer.
This segment from a Results section in which the result of one experiment
determined the next experiment follows the four-part pattern: question, experiment,
results, answer. Each repeat the pattern (that is, each paragraph) moves the story
line forward by adding more evidence that Zebrafish nkx2.5 can function like
vertebrate ceh-22.
In addition, within each paragraph, other information is included to make the story
line clear: in paragraph 1 and 2, reasons for the design of the experiment; in
paragraph 2 the purpose of the control experiment; in paragraph 3, background
leading to the next question.
51
Paragraph Structure: Use Topic Sentences to State Results
Putting important results first and less important details later can be a way of making
a result into a topic sentence, as in the following example.
The idea is to state the message so that the reader cannot miss it. In other words, if you
want the reader to know something, state it; do not make the reader guess.
In summary, the main ways to emphasize important results are by omitting data from the text
of the Results section and condensing the results. In addition, putting results at the
beginning of a paragraph (strongest power position) and using a result as a topic sentence,
accompanied by subordination of less important information (figure legends, table titles,
control results, methods statements---see next class) help emphasize the important results.
52
Class 7
Results: Emphasis
53
Emphasis
In the Results section, results should be emphasized. Very different techniques of emphasis
are used in the Results sections of studies in which one experiment determines what the
next experiment will be and Results sections of studies in which all experiments are
designed in advance.
In Results sections in which the result of one experiment determines what the next
experiment will be, the results are always in the middle of a paragraphs, which is the least
visible position. To emphasize the results, you can use a signal at the beginning of the
sentence: “We found,” or a variation such as “We observed” or “We detected.”
In the Results section of studies in which all experiments are designed in advance, several
techniques can be used to emphasize the important results. These techniques include
omitting, condensing, or subordinating less important information, and putting important
information in a power position. In addition, topic sentences can be used to give overview
before you give details.
Most data belong in figures and tables. If a lot of data are presented in the text, they can
overwhelm the results. Therefore, data should be kept to an absolute minimum in the
Results section. Data that are presented in a figure or a table should be omitted from the
text. However, one or two especially important values can be repeated in the text for
emphasis. In addition, brief secondary data that do not warrant display in a figure or a table
can be presented in the text by being placed within parentheses after the result.
The results should be condensed to avoid unnecessary repetition. For example, if the result
for several variables is the same, the result should not be stated over and over, variable by
variable. Instead, the result for all variables should be stated once.
54
results were the same regardless of whether alkalosis was
induced by decreasing PICO2 or by infusing NaHCO3 (Figure 3).
Thus, although both metabolic and respiratory alkalosis
decreased baseline pulmonary resistance, they did not
decrease constriction of the pulmonary artery in response to
subsequent alveolar hypoxia.
In this example, only the last sentence of each paragraph report results. The earlier
sentences report data that are shown in figures and can therefore be omitted, except for one
sentence of baseline results and data. These data should be in parentheses, as shown in
the following Revision.
Revision 1A:
In the revision, results are prominent. However, the overview would be clearer if repetition
were avoided and similar results were reported together, as in Revision 1B. (For this revision
to work, metabolic and respiratory alkalosis would have to be identified in the figures.)
Revision 1B:
55
Subordinate Figure Legends and Table Titles
Do not use a figure legend or a table title as a topic sentence. Cite figures and tables (in
parentheses) after statements that give results, preferably after the first result relevant to the
figure or table.
The first sentence is essentially a figure legend: Fig. 2. Renal function data. For a more
powerful topic sentence, omit the figure legend and state the results. Cite the figure in
parentheses at the end of the sentence that states the results.
Revision 2A:
The reason that a result is more powerful topic sentence than a figure legend or a
table title is that a result states a message. In Example 2, the first sentence
indicates only the topic of the paragraph---renal function data. The second sentence
states a message---that renal variables decreased, which is what the reader wants to
know. Therefore, this sentence should be placed first, in a power position, where the
reader will find it readily. It should not be buried in the middle of the paragraph.
Furthermore, to use an entire sentence to direct the reader to the figure is wasteful.
All that is necessary is to cite the figure in parentheses at the end of the sentence
that states the result, as in the revision.
56
Subordinate Control Results
Control results sometimes need to be described first, for example, if the stability of the
baseline needs to be established. Otherwise, control results, both for baseline and for
control series, should be described along with or after experimental results whenever
possible. For example, baseline data can sometimes be incorporated into a sentence
describing the experimental results.
In this example, 2.44 ± 0.32 is the baseline value. To make the baseline value more
noticeable, write “increased from a baseline value of ….”
Similarly, for a control series of experiments, the results can sometimes be
incorporated into a comparison between experimental and control results.
57
Subordinate Methods
In the Results section of studies in which all experiments are designed in advance, do not
use a methods statement as a topic sentence. The topic sentences in these Results
sections should state results.
In this example, since all of the methods details (except the number of cats) appear in the
second sentence, the first sentence is unnecessary. A stronger way to begin this paragraph
is to omit the first sentence and incorporate the number of cats into the second sentence.
Thus, the paragraph includes methods but highlight results.
Revision 5A:
At least two techniques of sentence structure are available for including methods in
sentences that state results. One technique is to make the method the subject of the
sentence, as in the revision. Other techniques are to state the method in a transition phrase
or clause and the result in the main clause.
58
C. Methods subordinated in a transition phrase;
result in subject + verb
The point is that in the Results section, the main verbs (bold in the examples above) should
describe the results, not methods, so avoid sentences that state methods only. Having the
main verb describe results is particularly important in the topic sentence.
59
Discussion
60
Class 8
Discussion: Function and Contents
61
The Discussion
Functions
The main function of the Discussion is to answer the question(s) posed in the Introduction.
Other important functions are to explain how the results support the answers and how the
answers fit in with existing knowledge on the topic.
Story Line
The last step in the story line of your paper comes at the beginning of the Discussion. In a
hypothesis-testing paper, the last step is the answer to the question (whether the hypothesis
is true or false).
Content
In a hypothesis-testing paper, the Discussion includes the answers to the questions posed in
the Introduction and any accompanying support, explanation, and defense of the answers. In
addition, the Discussion includes explanations of any results that do not support the
answers; indications of the newness of the work; explanations of discrepancies with others’
results; explanations of unexpected finding; explanations of limitations of the methods, of
weaknesses in the study design, or the validity of assumptions; and indications of the
importance of the work.
62
Answering the Questions
Question-Answer Match
The answer should answer the question exactly as it was asked, using the same key terms,
the same verb (when appropriate), and the same point of view. The verb should be in
present tense, because the answer should be true for your experimental design. For
example, if the question was “Does sympathetic stimulation increase norepinephrine
synthesis in rat superior cervical ganglia in vivo?” the answer would be either “This study
shows that sympathetic stimulation increases norepinephrine synthesis in rat superior
cervical ganglia in vivo” or “This study shows that sympathetic stimulation does not increase
norepinephrine synthesis in rat superior cervical ganglia in vivo.” Altering the key terms or
the verb or changing the point of view would make the answer more difficult to recognize.
Before the answer is stated, it should be signaled, so that the reader knows it is the answer.
Some signals of the answer are listed below. The verb tense used in the signal of the
answer depends on the subject of the sentences. If the subject is “study” or “results,” the
verb in the signal of the answer is in the present tense. If the subject is “we,” the verb in the
signal is in present, present perfect, or past tense, depending on the verb.
63
Supporting the Answer
Sometimes the answer to the question is short and simple---merely a statement of the chief
result. For example, for the question “to determine surface tension within alveoli at total lung
capacity,” the answer was “In this study, we found that, at 37 ºC, alveolar surface tension at
total lung capacity is 29.7 ± 5.6 (SD) mN/m.” In this case, no supporting results are needed.
However, the authors did defend their answer by comparing it with other published values
(see Example 5 below).
Usually the answer is not the same as the results. Rather the answer is a generalization
based on the results, either directly or indirectly. Therefore, to convince the reader that the
answer is valid, state the relevant results after stating the answer. Also cite a figure or a
table if seeing specific data would help the reader. Do not assume that the reader has
memorized the results or will search through the Results section, figures, and tables to find
the results that support the answer: The purpose of the Discussion is to weave that story
together for the reader.
Finding a transition to link the results to the answer is challenging. Other than “because,”
which would create a very long sentence, there are no obvious transition words you can use.
So you must create a transition phrase or transition clause, or use a topic sentence. Some
examples are listed below:
Transition Type
In our experiments Transition phrase
The evidence is that… Transition clause
Evidence that (answer) is that Transition clause
We found that* Transition clause
(Answer) has been demonstrated in two ways Topic sentence
Examples 1 and 2 show two ways to link results to the answer they support. Example 1 uses
transition clauses, Example 2 uses a topic sentences and transition words.
64
alpha (Fig. 4) and that in nuclear “run on” assays, cloned human
tumor necrosis factor alpha decreased the rate of MYC gene
expression.
In this example, two answers are stated in the topic sentence (sentence 1), and the results
supporting them are introduced by transition clauses that repeat key terms from the topic
sentence.
In this example, a topic sentence and transition words link the results to the answer.
If others’ results help support your answer, mention these results, and cite the appropriate
references. Neither overplay nor underplay your own or others’ contributions. If others’ work
clinches your point, say so. Conversely, if your work is the missing link that pulls together a
lot of loose threads, say that too. Be neither too modest nor too boastful. Example 3
illustrates one way of giving appropriate credit to yourself and others.
In this example, the first sentence of the Discussion states the answer. The second
sentence gives credit to others and also makes clear what the author’s contribution is.
65
Explaining the Answer
In addition to stating the answer and supporting it with results, you may need to explain the
answer. For example, why is the answer reasonable, or how does it fit in with previously
published ideas on the topic? An example of an explanation of how the answer fits in with
previously published ideas is given in the second and third paragraphs of Example 4.
In this example, the first paragraph states the answer to the question and supports the
answer with the results. The next two paragraphs explain how the answer (clustering of
collagen genes) fits in with previously published ideas (clustering of globin genes and of
histone genes).
66
Defending the Answer
If other possible answers have been proposed for the question you asked or if other answers
are easy to imagine, you should explain why your answer is more satisfactory than those
other answers. When defending your answer, you need to explain both why your answer is
satisfactory and why others are not. That is, you must argue both for your answer and
against the others. Only by giving both the pro and con arguments can you make your
answer convincing. An example of a pro-con defense of an answer is given in Example 5.
In this example, the first sentence states the authors’ answer (a surface tension value
obtained by a direct technique) and the second sentence defends I by comparing it with the
known value at equilibrium. This is the “pro” part of the pro-con argument. The rest of the
paragraph presents the “con” argument. It argues that different answers are not likely to be
accurate because the method used may not be measuring the same quantity. If we accept
both the pro and con arguments, we are willing to accept the authors’ answer.
67
Explaining Conflicting Results
In addition to stating, supporting, explaining, and defending your answer as necessary, also
mention any results you got that do not support your answer, and explain them the best you
can. An example of such an explanation of conflicting results is given in the last three
sentences of Example 6 below (“However,…”).
In this example, the third sentence states a finding that conflicts with the answer to the
question. The fourth sentence rules out two possible explanations, and the fifth sentence
proposes another possible explanation.
68
Establishing Newness
The newness of your work should be established in the Introduction by the statement of
what is unknown. If you want to remind the reader of the newness of your work, one way to
indicate newness is to contrast the point you are making with what was known before, as in
Example 7, which is the first two sentences of the Discussion.
1 Partial cDNA clones have been reported for mouse (38-40), rat (41,
42), and human (24) beta-glucuronidase. 2 In this study, we report the
complete sequence of the full-length cDNA for human beta-
glucuronidase.
In this example, newness is indicated by the contrast between “partial” in the first sentence
and “full-length” in the second sentence.
Avoid claiming priority (“This is the first report of…”). It is possible that the same or very
similar work has been reported in the literature of another country either in English or in
another language. If you feel strongly that you must claim priority, word your claim
cautiously. For example, you might say, “To our knowledge, this is the first report of…”
69
Explanation of Discrepancies
At the opposite extreme from others’ results that agree with yours and support the answer to
the question are others’ results that should agree but do not. Such discrepancies need to be
explained as best you can. An example of an explanation of a discrepancy is given in
Example 8.
70
Explanation of Unexpected Findings
Unexpected findings can range from minor to exciting. In a few cases they are more exciting
than the original question and take over the paper. When describing an unexpected finding,
state at the beginning of the paragraph that the finding was unexpected (or surprising) and
then explain it as best you can, as in Example 9.
71
Limitations of the Methods, Weaknesses in Study Design, Validity
of Assumptions
If your methods have limitations or if your study design is weak or is based on any
assumptions, you should state what the limitations, weaknesses, or assumptions are and
explain why the limitations, weaknesses, or assumptions are acceptable. If the explanation
is brief (one or two sentences), it can be stated in the Methods section. If the explanation is
longer (one or two paragraphs), or if the limitation, weakness, or assumption is likely to
affect your results seriously, the explanation should be included in the Discussion. If
possible, explanations of limitations of methods, weakness of study design, or the validity of
assumptions should be woven into the story. An example of an explanation of the validity of
an assumption is given in Example 10.
In this example, the author first states what the assumption is and then gives three reasons
that the assumption is reasonable.
72
Establishing Importance
The importance of the work is sometimes obvious from its newness---for example, if the
cause of a disease was unknown and you are reporting the cause. Similarly, basic research
newness is usually sufficient evidence of importance. But if importance needs to be
established, that can be done either in the Introduction or in the Discussion. In the
Discussion, importance can be established by describing applications or implications of the
answer or by stating recommendations or speculations based on the answer. Applications,
recommendations, implications, and speculations are often used as an ending of a
Discussion.
73
Class 9
Discussion: How to Begin
74
The Discussion: Organization
To ensure that your Discussion is organized rather than rambling, think of the Discussion as
telling a story, and focus the story on the question you asked in the Introduction. To tell a
story, give your Discussion the three standard parts of a story---a beginning, a middle, and
an end.
The beginning of the Discussion (that is, the first paragraph or more) should state the
answer to the question and support the answer with results. Beginning with the answer is the
best way to focus your story on the question.
The middle of the Discussion should organize topics either in an order dictated by the
science or in the order of most to least important to the answer.
The end of the Discussion (that is the last paragraph), which we’ll call the Resolution, should
conclude by restating the answer to the question or by indicating the importance of the work.
The guidelines below explain how to tell a story in the Discussion. The most important of
these guidelines are to state the answer to the question at the beginning of the Discussion,
to use a topic sentence at the beginning of every paragraph to state the topic or the
message, and to link each paragraph to the previous paragraph.
75
How to Begin the Discussion: Answer the Question
Begin the Discussion with the answer to the question, followed immediately by support for
the answer. The reason for beginning with the answer to the question is that the beginning of
the Discussion is a power position and therefore should be used for the most important idea.
In the Discussion, the most important idea is the answer to the question. The reason the
answer is the most important idea is that the answer fulfills the expectation created by the
statement of the question in the Introduction and by the statements in Methods and/or
Results of what experiments you did to answer the question and what results you found to
answer the question. Thus, the answer to the question is the culmination of the paper. It
deserves the most prominent position in the Discussion---the beginning.
The answer can be stated in the first sentence of the Discussion (the strongest position) or
after one sentence that either restates the question or give brief context.
If the study had two questions, both answers can be given in the first sentence, or the
answers can be in separate sentences at the beginning of the first and second paragraphs.
When both answers are given in one sentence, signal the second answer by using “and
that”. When the answers are in separate sentences, use a different signal for each answer,
for example, “This study shows that (answer 1)…” “We also found that (answer 2)…”
Having more than two questions is unlikely. If your paper has three or more questions, you
may be fractionating the question or writing a detailed experimental approach.
76
Examples of Ways to Begin the Discussion
State the Answers to the Questions. The most straightforward way to being the
Discussion is to state the answers to the questions.
If your study dealt with one question, begin the Discussion by answering that question. Put a
signal before the answer. Be sure the answer answers the question you asked.
In this example, the sentence begins with a signal of the answer (italicized) and ends with
the answer (underlined). The answer matches the question stated in the Introduction: “The
purpose of this study was to further define the effects of continuous positive airway pressure
on renal function in newborns.” The key terms are the same (continuous positive airway
pressure, renal function, newborns) and the point of view is the same (cause and effect).
If your study dealt with two questions, you can begin the Discussion by answering both the
questions, as in Example 2, or you can answer only one question at the beginning of the first
paragraph and the other question at the beginning of a later paragraph, preferably the
second paragraph. The only information that should be given between the two answers is
results supporting the first answer. Put a signal before each separate answer. Answer the
questions in the same order as they appear in the Introduction.
This sentence answers both of the questions posed in the Introduction. It begins with one
signal for both answers (“Our experiments show that”) and then uses “and that” to signal the
second answer. The questions as stated in the Introduction were “to determine the severity
of cigarette-smoke-induced bronchoconstriction…to determine the mechanism of cigarette-
smoke-induced bronchoconstriction.” Because the answers use the same key terms
(underlined), it is clear that the answers answer the questions asked.
77
Example 3 Question and Answer
In this example, the question and answer use the same key terms and the same point of
view (cause and effect). In addition, the question restated in the Discussion matches the
question stated in the Introduction: “The aim of the present study was to determine whether
hypercapnia and hypoxia affect the magnitude of the abdominal expiratory neural activity in
the absence of any changes in procioceptive afferent activity from the lungs and chest and
abdominal walls.” The results that support the answer are given immediately after the
answer is stated and are signaled by transition phrases that repeat key terms: “For
hypercapnia,for hypoxia…”
If you restate the question at the beginning of the Discussion, go on immediately to state the
answer. Do not, for example, launch into a long review of your experimental approach. If you
state the experimental approach, keep it brief. The long the answer is delayed, the more
difficult it is for the reader to find the answer. Since the answer is the most important
information in the Discussion, it should be presented in the most prominent place---the very
beginning of the Discussion. In example 4, the answer is delayed until the third and fourth
sentences, which is about as long a delay as is tolerable.
In this example, the first sentence restates the question and the second sentence states the
experimental approach to answering the question. The third and fourth sentences state the
answer. The third sentence begins with a single of the answer. It then states a rejected
answer, which is parallel to the question and uses the same key terms and the same verb.
78
The third sentence ends with the reason for rejecting that answer. The fourth sentence
states the accepted answer. Signaling the rejected answer and using parallel form are
especially important because of the sentence of experimental approach that intervenes
between the question and the rejected answer. Although the fourth sentence changes the
point of view and the verb, it is clear that this is the answer because “rather” signals a
contrast with the third sentence, which stated the rejected answer. The last two sentences of
this paragraph present results that support the answer; the results are signaled by “In our
experiments.”
Provide a Brief Context and Then State the Answer. Providing a brief
context is another way of avoiding an abrupt beginning of the Discussion. The context must
accurately restate a point made in the Introduction and must be brief. One or at most two
sentences are enough, as in Example 5.
In Example 5, the first sentence is context. The second sentence states the answers to the
questions. The first answer closely parallels the first part of the context, making it easy to
see the difference between what was previously suggested and the message of this study.
Although restating the question and providing context are more sophisticated ways to begin
the Discussion than simply stating the answer to the question, the best way to begin the
Discussion is usually the simplest way---by stating the answer. Answers stated at the
beginning of the Discussion are easiest for the reader to find and have the most impact.
79
Examples of Ways Not to Begin the Discussion
The Discussion is not eh place to rerun the Introduction. At the beginning of the Discussion,
you can restate your question or give one sentence of background, but that is all. When we
get to the Discussion, we are ready for the answer. Give it to us. And then go forward,
explaining how those references you want to include fit in with your answer.
80
Class 10
Discussion: Relating your findings
to existing research
81
Class 11
Discussion: Dealing with Limitations
and How to End the Discussion
82
Dealing with Limitations
No research is perfect. Every study has limitations, and every data set has blemishes. You
have to address these even while focusing on meaningful results and that understanding
that grows from them. The question is: how do you address the negatives without
undermining the positives?
Many writers address limitations using what is called a “yes, but” strategy, where they
present the results and conclusions and then discuss their limitations. The most classic form
of this is the “more research is needed” resolution. The problem with “yes, but” is that it puts
the “but” last, making it the resolution and most powerful message.
So, what is the alternative to “yes, but”? Turn it around. Instead of “yes, but,” use “but, yes.”
Deal with the limitations as early as possible, get that discussion out of the way, and then
get on with developing a strong story. Constrain your conclusions to fit within the limitations
but end with a “yes.”
“But, yes” allows you to end powerfully, making for better storytelling. Importantly, though, it
is also easier for the reader to follow and, perhaps, more honest. It achieves these aims by
stating your limitations up front, rather than having it feel like something you hope readers
will miss. Having expressed the limitations, readers can work through your data and
discussion in light of them. They never have to back up and reprocess information---that
takes work and creates confusion. Depending on the issues, you may need to address
limitations in any part of a paper; as a rule, earlier is better.
83
Limitations of the Methods
When limitations relate to experimental details or analytical methods, discuss them as soon
as possible. An excellent example of how to do this is in a paper evaluating overwinter
survival of snowshoe hares in northern Canada. One of the experimental treatments
involved a large fenced exclosure to keep out predators.
Example 1:
1 Our final two treatments increased hare survival by
excluding predators. 2 One 1-km2 area was enclosed by
chicken-wire and electrical fencing (predator-exclosure) from
1987 to 1996 to exclude terrestrial predators, primarily lynx
(Lynx canadensis Kerr) and coyotes (Canis latrans Say).
The limitation here is that the design is unreplicated---they had only one exclosure.
Traditional statistical logic might argue that without replication it is impossible to determine
whether effects were due to the exclosure itself. Some people might read this and think
“unreplicated and unstatistical, therefore unreliable and unscientific.” How to address this?
The authors did so with this paragraph.
Example 2:
1 The predator-exclosure and predator-exclosure + food sites
were not replicated, owing to the high costs of maintaining 8
km of electric fencing on a nearly daily basis for a decade. 2
Ecologists are caught increasingly in the tension between the
manipulations necessary to understand large-scale ecological
processes and the impossibility of replicating large
manipulations. 3 Our control sites sampled a wide range of
natural variation within our study area, increasing the
likelihood that differences on the treatments were due to our
manipulations rather than to underlying natural variation.
This addressed nonreplication in a way that was effective and unapologetic. First, the
authors made it clear why it would be impossible to replicate. Then they grounded this within
the larger discussion of how we study ecology: we often can’t use nice, tidy replicated
experimental designs to study messy nature. Environmental scientists must (and do) accept
pseudo-replicated experiments as long as the investigators are sensitive to and deal
appropriately with their limitations. Inferences of causality have to go outside simple
statistics. Finally and importantly, they discussed how they dealt with the limitations to
convince readers that differences between control and exclosure were likely real treatment
effects. Having done this, they didn’t raise the issue again. The got on with discussing the
real science and the important results.
84
Limitations that affect how you interpret the data
When limitations affect how you interpret the data, you must address them in the Discussion.
You must openly discuss limitations but without highlighting them so strongly that they
become an argument for rejecting the paper entirely. Doing this involves both structure and
language. In terms of structure, you should avoid the power positions of the Discussion’s
opening and resolution. Those are for critical story points, and if limitations are the critical
point, why submit the paper? You should generally find some convenient place early in the
body of the Discussion to discuss the work’s limitations and constraints.
As for language, do it briefly and honestly, then get on with the story.
Example 3:
1 Soil bacterial diversity varied across ecosystem types. 2 Of
all soil and site variables examined, soil pH was, by far, the
best predictor of soil bacterial diversity with the lowest levels
of diversity observed in acid soils. Because soils with pH
levels > 8.5 are rare, it is not clear whether the relationship
between bacterial diversity is truly unimodal or whether
diversity simply plateaus in soils with near-neutral pHs.
Likewise, because our fingerprinting method underestimates
total bacterial diversity, we cannot predict how the absolute
diversity of bacteria changes across the pH gradient. When
we compare paired sampling locations with similar vegetation
and climate but very different soil pHs, we find evidence for
the strong correlation between bacterial diversity and soil pH
at the local scale.
The authors included two constraints on the conclusions, but they put them in the middle of
a point-first paragraph---after the topic sentence but before the rest of the discussion.
A good discussion of methodological limitations comes from a paper that explored the side
effects of inhaled cortical steroids (ICS) as a treatment for asthma. Clinical trials may
underestimate side effects because trials are usually short and have restricted selection
criteria for who participates. To better evaluate side effects of ICS, the researchers surveyed
people who routinely used them and found effects ranging from gummy mouths to mood
swings and sleeping problems, some of which had not been well documented. But survey-
based studies always have limitations that the authors rightfully felt they needed to bring to
the reader’s attention.
Example 4:
1 Limitations of our research include potential selection bias. 2
Although our response rate (56%) is not atypical for a
questionnaire study, we acknowledge that responders may
have had a greater propensity for perceiving ICS side effects
as compared to non-responders, and it is possible that side
effect reporting may have been inflated in all dose groups as a
result. 3 Alternatively, ICS users in our study may have
experienced relatively mild ICS side effects, since they
continue to use ICS regardless of their perception of side
effects.
This pointed out that there is some quantitative uncertainty about the results that can never
be overcome in a survey study---you can’t make people answer, or answer honestly. But
85
there were robust patterns in the results, offering information that physicians should be
aware of, which the authors then elaborated.
Another example of dealing with limitations in the Discussion comes from a paper about how
variation between individuals can affect transcription patterns in different tissues, ultimately
leading to different physiology.
Example 5:
1 Despite the limited number of validated examples,
osteoblast-specific transcript isoforms are present and
indicates that expression of certain alternatively spliced
variants in bone tissue are controlled by different regulatory
SNPs and vary between individuals and/or populations, with
potentially important biological functions.
The authors start with the opening clause “Despite the limited number of validated
examples,” which frames the “but”; they then give the “yes” by saying that osteoblast-specific
transcript isoforms are present and suggest the implications of that. They use the “but, yes”
approach within a single sentence. They remind us of the limitations but don’t dwell on them.
86
Limitations within the Resolution (Conclusions)
You should never make limitations the conclusion, but sometimes you may need to mention
them within the conclusions. In such cases, you need a tightly defined “but, yes” structure to
frame the limitations as quickly as possible and then to push on to the conclusions---what
you learned.
A particularly clever approach to doing this is illustrated in the following example. This is
from a paper that evaluated approaches to estimating the mortality of children under five
years old in nations that don’t keep records of births and deaths. There was no absolute
data set to use as a reference to assess the reliability of the estimates. That left
uncertainties that constrained the conclusions. But if the authors had included that entire
discussion in the concluding paragraph, it would have overwhelmed it. Instead, they
discussed the limitations in the next-to-last paragraph, which was long and involved (21
lines). The final conclusions paragraph was only 11 lines.
Example 6:
1 Despite these limitations, the methods proposed here
represent a major advance on current practice and offer the
prospect of vastly increasing our knowledge about levels,
recent trends, and inequalities in child mortality. If we are to
make rapid progress with the unfinished agenda of reducing
child deaths, policy and practice must be better informed by
more comprehensive, relevant, and timely information.
Systematic application of the methods proposed here will
establish that evidence base, and thereby increase
accountability among countries and the global health
community to accelerate efforts to reduce the global toll of
child deaths.
This reminds us of the constraints, but with a tidy nominalization (“these limitations”) that
refers back to the previous paragraph. This simultaneously highlights and minimizes them!
By separating the negatives and the positives into separate story arcs and using effective
language to link them together, the authors addressed the constraints honestly in a way that
did nothing to undermine the power of their conclusions. This was nicely done.
No research is perfect, and there is nothing to be embarrassed about admitting it. Rather,
the opposite is true---it’s our responsibility to address our work’s limitations. The “but, yes”
approach does this in a way that is open about the limits but highlights the conclusions. It
makes the presentation clear, strong, and credible. By clarifying both the limits and the
power of your work, it will motivate others to pick up and build from it, that is, to cite it.
87
How to End the Discussion: Make a Point
The Discussion should not simply stop. It should come to a definite, clear end.
Two standard ways to end the Discussion are to restate the answers to the questions, and to
indicate the importance of the work by stating applications, recommendations, implications,
or speculations. Or you can do both. These endings are explained and illustrated below.
A statement that further studies are needed is not a particularly strong ending. It is stronger
to end by stating what knowledge you are contributing than by stating what remains to be
studied.
The restatement of the answer should be preceded by two signals: first a signal of the end
and then a signal of the answer. Examples include “In summary, we have shown that…” and
“In conclusion, this study shows that…”
Example 1:
In this example, the beginning of the first sentence signals firs the end (“In summary”) and
then the answer (“we have shown that”). The rest of the paragraph states the answers. Note
that the verbs that state the answers are in present tense (“is brought about,” “allows”)
If you restate the answer as an ending of the Discussion, it is crucial that the answer at the
end of the Discussion matches the answer at the beginning. If the two statements of the
answer are different, how is the reader to know which one you believe?
88
implication is a logical step that follows from an answer. But whereas answers are solid
because they are based on data, implications are tentative because they have no data to
support them nor were any experiments done to obtain data for them. Speculations are
similar to implications, but speculations are even more tentative. Whereas an implication is a
logical next step, a speculation is more of an imaginative leap. Nevertheless, reasonable
speculation can be useful, for example, for suggesting relationships between ideas, and
therefore should be included in a Discussion when warranted.
Signals are useful to identify recommendations, implications, and speculations, but are not
needed for applications. To signal recommendations, use the verb “recommend.” To signal
implications, use “suggest” or “imply.” To signal speculations, use “speculate.”
In addition, use appropriate helping verbs within the statement. Applications should use
helping verbs that imply certainly, such as “can” or “will.” Recommendations should use a
helping verb as “should” or “must.” Implications and speculations should use a helping verb
such as “should” or “must.” Implications and speculations should use cautious verbs---for
example, “may inhibit,” “might prevent”---or should add a cautious adverb---for example,
“probably reflects.”
Example 3 Application:
In this example, the subject of the sentence (“Isolation of the genes…”) is the answer,
“reported here” reminds the reader that it is the answer, and the rest of the sentence gives to
applications that indicate the importance of the answer. The verbs used for the application
indicate certainty: “provides” and “can be used.”
Example 4 Recommendations:
89
negative margins. 4 Incomplete resection should be
undertaken only for symptom relief.
In this example, the first sentence restates the answer. The last three sentences state
recommendations. In each recommendation, the verb is “should…” To signal the
recommendations, the authors could have added “We recommend that” before “Patients
with primary disease…”
Example 5 Implication:
In this example, the findings the implications are based on are alluded to at the beginning of
the first sentence. The implications are signaled by “suggest” and the verbs stating the
implications indicate minimal certainty (“may antagonize,” “may have,” “may permit,” “may
limit,” “may not have”).
Example 6 Speculation:
In this example, the subject of the first sentence (“The chromosomal pattern encountered for
the type IV genes”) is the answer, “encountered” reminds the reader that it is the answer,
and “leads us to speculate that” signals the speculation, which is stated in the rest of the
sentence. The verb used in the speculation (“will be clustered”) indicates certainty, but it is
tempered by the preceding “if.” Two more speculations, based on a related answer reported
by other authors (second sentence), are given in the last sentence. The verbs indicate
moderate and minimal certainly (“probably reflects,” “may influence”).
These standard endings can also be used in combination. Usually a restatement of the
answer(s) is given first and is followed by an application, a recommendation, and implication,
or a speculation.
90
1 In summary, our results indicate that expansion of
plasma volume by 400 ml in untrained men increases stroke
volume during exercise by 11% but that further expansion of
plasma volume has no apparent hemodynamic benefit. 2
These findings imply that in untrained men, the
measurement of stroke volume during upright exercise when
blood volume is normal may not provide an adequate
measure of intrinsic myocardial function. 3 It appears that
about one-half of the difference in stroke volume normally
observed between untrained and highly endurance-trained
men during upright exercise is due to suboptimal blood
volume in the untrained men.
In this example, the first sentence restates the answer. The second sentence states a
clinical implication of the findings. The third sentence quantifies the inadequacy mentioned in
the second sentence. The implication is signaled by “These findings imply that,” and the
verbs used are cautious (“may not provide,” “appears”).
In this example, the first sentence restates the answer, the second sentence states
supporting evidence, and the last sentence, states two speculations. The speculations are
signaled by “we believe that,” and the verbs used are cautious (“may help,” “may result”).
As these examples show, restating the answers or indicating the importance of the work by
stating applications, recommendations, implications, or speculations gives the Discussion a
sense of coming to a definite, clear end.
91
Introduction
92
Class 12
Introduction: Function and
Organization
93
The Introduction
Functions
The Introduction has two functions. One is to awaken the reader’s interest. The other is to
justify your research question and experimental approach.
For a hypothesis-testing paper, what the reader needs to know from the Introduction is what
the question of the study is and where it comes from, that is, why the author is asking this
question. The question, which is the most important statement in the Introduction, is stated
either as a question or as a hypothesis. The story of where the question comes from is
composed of what is known or believed about the topic and what is still unknown or
problematic.
References
The statements about what is known must include references. The references should be
chosen to reflect the key work that led to the question of your paper.
The number of references should be kept to a minimum. If a lot of work has been done on
the topic, select papers describing the first, the most important, the most elegant, and the
most recent studies. Keep in mind that reference lists in papers you cite can lead readers to
other references. You can also cite review articles.
94
Organization For Hypothesis-Testing Papers
Known
The known is the first step in the funnel. The known often includes many sentences, which
narrow by appropriate scientific logic to the unknown.
Unknown
The unknown is usually just one or two sentences. Though brief, the statement of the
unknown is important for two reasons. One reason is that the unknown indicates that the
work is new. The other reason is that the unknown is the step that links the known and the
question, creating a story line.
A crucial feature of the unknown is that it is virtually the same as the question. For example:
Example 1:
1 It is known that several general anesthetics, including
barbituates, depress the bronchomotor response to vagus
nerve stimulation (refs). 2 However, the site of this depression
has not been determined. 3 To determine which site in the
vagal motor pathway to the bronchioles is most sensitive to
depression by barbiturates, we did experiments in isolated
rings of ferret trachea in which we stimulated this pathways at
four different sites before and after exposure to barbiturates.
In Example 1, the site is unknown. The question is, “what is the site?” Thus, once the
unknown is stated, the question is determined.
Question
The question is the specific topic of the paper and the end of the funnel. (The general topic
of the paper is named in the first sentence of what is known.)
Experimental Approach
The Introduction can end with the statement of the question or can go on to state the
experimental approach. If the experimental approach is included, the logical place for it is
after the question. Putting the experimental approach before the question (treating the
95
experimental approach as “background”) does not work. The background (known, unknown)
is what leads to the question. The experimental approach does not lead to the question: it
follows from the question: “Here is our question; here is how we went about answering our
question.”
Example 1 illustrates a simplified Introduction consisting of a brief funnel: one sentence each
for the known and the unknown, and a third sentence for the question followed by the
experimental approach. Specifically, the first sentence names the general topic of the paper
(barbiturates’ depression of the bronchomotor response to vagus nerve stimulation) and tells
what is known about it. The second sentence states the unknown (which site is most
sensitive to depression), which leads immediately to the question. The question names the
specific topic (which site in the vagal motor pathway to the bronchioles is most sensitive to
this depression).
Longer introductions also follow this funnel shape but expand it. One way to expand the
funnel is to extend what is known, as in Example 2. In this Introduction, two paragraphs state
what is known. Paragraph 3 finishes the funnel: sentence 9 implies the unknown; sentence
10 implies another unknown and also implies the importance of the work (that we are finding
an overarching mechanism for a variety of similar functions in disparate species); sentence
11 states the hypothesis; sentence 12 states the experimental approach.
Example 2:
96
Class 13
Introduction: The Aim (Question)
97
The Aim
The aim of the paper is generally stated at the end of the introduction. Here, you describe
the knowledge you hope to gain. What is the question that drove you to do the research? In
some fields, you formulate a hypothesis. Also, by stating specific objectives and describing
the information we will present, you give readers a way to assess the paper as they read.
In general, the aim is usually simplified as “to learn X, we did Y.”
Stating the aim of a paper clearly usually involves two to three elements: (1) the research
question, (2) the research approach, and (3) predictions / hypotheses. Read the last
paragraph of LaDage et al.’s paper on the controls of brain development. Can you identify
these elements? Are they all present?
1 Our goal in this study was twofold. 2 First, we tested whether an animal’s
physical environment would affect hippocampal attributes. 3 Specifically, we
tested whether food-caching mountain chickadees (Poecile gambeli) housed
in captivity differed in hippocampal volume, hippocampal neuron number and
neuronal density as compared with fully developed wild-caught conspecifics.
4 We predicted that captivity, with reduced environmental complexity and
restricted memory-based experiences (compared with memory-based
experiences afforded in natural environment), would reduce hippocampal
volume, neuron number and, potentially, neuron density. 5 Second, we
attempted to determine, within the context of captivity, whether differences in
hippocampal structure, neuron number, and/or neuron density are based on
differential memory-based experiences, while controlling for physical activity.
6 We expected that deprivation of memory-based experiences, mediated
through caching and retrieving food items, would decrease hippocampal
volume, neuron numbers, and neuronal density, as compared with captive
counterparts with memory-based experiences.
In papers for specialist journals, a good aim almost always condenses conceptually to “to
learn X, we did Y.” That is, they present the question and lay out an approach to answering
it, as illustrated in example 1. This paper explored how living in a complex environment may
enhance brain development.
In sentence 1, the authors make a general statement that their study has two questions. The
first question is stated in sentence 2. Sentence 3 provides a general description of the
experimental approach and the data collected (hippocampal volume, hippocampal neuron
number, and neuronal density). In sentence 4, the authors predict results that this
experiment will yield. Sentence asks a second question and hints at the experimental
approach, which seems similar to the first experiment, but controls physical activity. The
authors end this aim by states their predictions for the second experiments.
98
The Question
Precision
If a question is stated as a question, it should be equally precise. That is, for greatest clarity,
the question should name the variables studied and use a precise verb, in present tense
(e.g. in this study, we asked whether respiratory alkalosis during exercise increases blood
lactate concentration). If the question were stated as “The purpose of this study was to
determine the effect of respiratory alkalosis during exercise on blood lactate concentration,”
the reader would have only a hazy idea of what answer to expect---something about the
effect of respiratory alkalosis during exercise on blood lactate concentration. The precise
question is much more helpful. The reader now expects a precise effect: that respiratory
alkalosis during exercise increases blood lactate concentrations more than exercise alone
does or that it does not.
So one reason a precise question is an advantage is that the reader immediately can read
the paper in a directed way rather than blindly. The experiments make more sense. The
results and the answer fit into the expectation. All the guesswork is taken out of the reader’s
job.
Inevitability
The question should follow inevitably from the previous statements of what is known or
believed and what is still unknown or problematic. Thus, the topic of the question should be
the same as the topic in the statement of what is known, and the question should be the
question we would expect after reading the statement of what is unknown or problematic.
Note that the question follows inevitably from the previous statements. The question is,
“Which site in the vagal motor pathway to the bronchioles is most sensitive to depression by
barbiturates?” This question clearly derives from the statement of the unknown (“the site of
99
this depression has not been determined”), which in turn clearly derives from the statement
of the known (“several general anesthetics, including barbiturates, depress the
bronchomotor response to vagus nerve stimulation”).
To check that a question follows inevitably from the statements of what is known and
unknown, look at the key terms in the question. Each key term in the question should have
appeared earlier in the Introduction. In the example above, “site” is in the unknown
(sentence 2). The other key terms (“depression,” “barbiturates,” “vagal,” “motor,” “pathway”)
all appear in some form in the known (sentence 1).
Sometimes, the question is not stated immediately after the unknown or the problem.
Instead, evidence suggesting a possible answer is stated. In this case, the question should
follow inevitably both from the unknown and from suggestive evidence as in Example 3:
In this Introduction, the known is stated in sentences 1 and 2, and the unknown is stated in
sentence 3. Sentence 4 states a possible answer and rejects it. Sentence 5 proposes a
different answer. Sentence 6 give suggestive evidence that supports the proposed answer.
Sentence 7 then states a hypothesis (question) based on the unknown and on the
suggestive evidence. We can tell that they hypothesis is based on the earlier statements
because key terms from the known (“prostaglandin E1 and E2”), from the unknown
(“regulate the responsiveness”), and from the suggestive evidence (“degree of constriction of
the ductus arteriosus”) all appear in the hypothesis.
Sometimes the question is presented in the statement of the unknown. In this case, it is
unnecessary to state the question again. Instead, at the beginning of the next sentence, you
can simply identify the previous statement as the question by using a transition phrase, such
as “to answer this question” or “to test this hypothesis,” as in Example 4:
100
1 The occurrence of a thermal transition in human serum
lipoproteins depends on the triglyceride-cholesteryl ester ration
and the size of the lipoprotein particle (5). 2 The triglyceride-
chloesteryl ester ratio is known to correlate negatively with the
peak temperature of the thermal transition of intact low-density
lipoprotein (3). 3 However, it is not yet known how low a
triglyceride-cholesteryl ester ratio and how small a particle size are
necessary for the occurrence of the thermal transition in
triglyceride-rich lipoproteins from human serum. 4 To answer
these questions, we assessed the triglyceride-cholesteryl ester
ratio and the particle size in two classes of triglyceride-rich
lipoproteins whose ratios of triglyceride to cholesteryl ester and
whose particle size are between those of low-density proteins and
very low-density lipoproteins.
In this example, sentence 3 states two unknowns, which are complete statements of the
questions this paper answers. Sentence 4 begins with a transition phrase identifying the
statement of the unknowns as the questions (“To answer these questions”).
However, if, as in 3, the question is stated only partially in the statement of the unknown (as
it is in sentence 2), the entire question should be stated in the next sentence (as it is in
sentence 3). The reason is that the question is the focal point of the paper. Readers should
not have to compose the question for themselves. The question should be stated for them in
a single sentence.
101
Experimental Approach
Sometimes the Introduction can end after a statement of the question. However, often it is
helpful for the reader to know the experimental approach to answering the question,
especially if the approach is new, unusual, or complicated, if the question alone does not
give a clear idea of what the experiments will be, or if the study needs to be identified as
having been done in vitro, retrospectively, or whatever.
In general, the experimental approach is short---usually one sentence; at most, two or three
sentences. Usually, the experimental approach describes one of the variables (independent
or dependent). In the example above, the animal studied is named (ferret), the independent
variable is described (“we stimulated this pathway…before and after exposure to
barbiturates”), the number of dependent variables is stated (“four different sites”), and the
fact that the experiments were done in vitro is made clear (“in isolated rings of ferret
trachea”).
102
Signals of the Question and the Experimental Approach
Signals are needed to identify the question and the experimental approach. The signals
vary depending on whether the question is stated as a hypothesis or as a question, and on
whether the question and the experimental approach are in the same sentence or in
separate sentences.
For a question stated as a hypothesis, the word “hypothesis” should be used in the signal.
For a question stated as a question, the signal can take a variety of forms. In all of these
forms, the crucial element is a verb followed by a question word. A common example is “to
determine whether.” (Here the verb is an infinitive, “to determine,” and the question word is
“whether.”)
When the question and the experimental approach are in the same sentence, only the
question needs a signal. The signal of the question is written as a purpose (for example, “To
determine whether…”), and the experimental approach follows naturally as a description of
the experiments done. When the question and experimental approach are in separate
sentences, the closest continuity comes from repeating a key term from the signal of the
question in the signal of the experimental approach. If no key term can be repeated, you
can use “For this study” to signal the experimental approach.
Commonly used signals of questions and the accompanying signal of the experimental
approach are listed in the table below. Numerous variations on these signals are possible.
103
Class 14
Introduction: Writing about Previous
Research
104
The Funnel: Using References to Build Justify Your Aim
In the Funnel of an Introduction authors use selected literature from their field to justify their
study to help construct a niche for their own work. The opening of the paper identifies the
large problem, while the aim defines a specific research question. The main body of the
Introduction (“Funnel”) must connect these elements. It narrows the focus and leads
readers from the general to the specific.
In the Funnel, authors write sentences supported by references to the literature they have
selected. In this context, the term literature refers to all the published research articles,
review articles, and books in a given field. The term also includes information published on
website that have been peer-reviewed or belong to organization with appropriate scientific
reputations.
References to other published studies, also known as citations or in-text citations, can be
used in all stages of the Introduction. They appear in the text either as a surname and year
in brackets, e.g., (McNeill, 2000), or as a number, e.g. 7. The details of the presentation
depend on the style stipulated by the journal. Check the Instructions to Contributors of your
target journal for the necessary information on referencing style. These references refer to
the list of references at the end of the paper, where the full publication details are written.
Citations are particularly vital in showing that you know clearly the work that has been
conducted by others in your field, and therefore what has not been done and needs to be
done: the gap that your study will fill. What you are required to do here is, in effect, to
construct an argument which justifies your own study and shows why and how it is
important.
Example 1:
In example 1, sentence 1 is the topic sentence of the paragraph. The topic sentence states
simply that the key term “propagule pressure” influences the success of another key term
“invasion.” Sentences 2-3 provide support for this topic sentence. Sentence 5 states the gap
in our knowledge. Here, that gap is two-fold: a limitation of previous studies and a
phenomenon (interaction of factors) that is not well understood.
105
Using Citations to Develop Your Argument
Below are examples of parts of paragraphs using three different citation methods. These
methods can be called information prominent, where the focus of the sentence is only on the
information being presented; author prominent, where the name of the author of the
information is given prominence in the sentence; and weak author prominent, where the
ideas of the author(s) are given prominence, but author names do not appear in the main
part of the sentence.
Example 1:
Shrinking markets are also evident in other areas. The wool
industry is experiencing difficulties related to falling demand
worldwide since the development of high-quality synthetic
fibres (Smith, 2008).
Author prominent citation provides the option to show the writer’s view of the cited fact. In
the following example, it shows that the writer (you!) agrees with Smith.
Example 2:
Shrinking markets are also evident in other areas. As Smith
(2008) pointed out, the wool industry is experiencing
difficulties related to falling demand worldwide since the
development of high-quality synthetic fibres.
Author prominent citation style 2 also allows the use of verbs such as “argued,” which give
the reader advance notice that a “however” or some other contrast may be coming, and
indicate that what is being cited is not necessarily accepted as correct by you, the writer.
Example 3:
Shrinking markets are also evident in other areas. Smith
(2008) argued that the wool industry is experiencing difficulties
related to falling demand worldwide since the development of
high-quality synthetic fibres. However, Jones et al. (2010)
found that industry difficulties were more related to quality of
supply than to demand issues. It is clear that considerable
disagreement exists about the underlying sources of these
problems.
However, there is a danger attached to the author prominent style. If it is over-used, it can
make the text sound like a list, rather than a logically constructed argument. We recommend
that you use this style rarely, perhaps when you are approaching the specifics of the gap
106
your study will address. It is also useful to pay close attention to the papers you read in your
own field, to check how often, if at all, this style appears.
Weak author prominent citation has a reference to authors in the subject and then more than
one reference in the brackets. This style can be useful as a topic sentence when beginning a
new subtopic or line of argument. Note that this style requires the use of the present perfect
tense (have reported).
Example 4:
Several authors have reported that the wool industry is
experiencing difficulties related to falling demand worldwide
since the development of high-quality synthetic fibres (Smith,
2008; Wilson, 2009; Nguyen, 2011). For example,…
107
Citation and Tense
Tense choice in reviewing previous research is subtle and somewhat flexible. (It is also not
very much like the “rules” you may have been taught in English classes.) The following,
therefore, are only general guidelines for tense usage.
Several studies have shown that at least two-thirds of all citing statements fall into one of
these three major patterns.
Example 6:
The causes of illiteracy were investigated by Jones (2017).
In Example 5 and 6, attention is given to what previous researchers did. The main
verbs used in this type of citation refer to what a previous researcher did
(investigated, studied, analyzed, etc.). Generally, in these cases the past is
obligatory. However, the main verbs can also refer to what the previous researcher
wrote or thought (stated, concluded, claimed, etc.). With these reporting verbs, tense
options are possible.
Example 7:
Jones (2017) concluded that illiteracy can be related to....
Example 8:
Jones (2017) has concluded that…
Example 9:
Jones (2017) concludes that...
The differences among these tenses are subtle. In general, a move from past to
present perfect and then to present indicates that the research reported is
increasingly close to the writer in some way; or close to the current state of
knowledge.
The present tense choice is sometimes called the citational present and is also used
with famous or important sources.
Example 10:
Plato argues that....
Example 11:
Confucius says…
Example 12:
The Constitution states...
108
Comparable options exist in the subordinate clause.
Example 13:
Jones (2017) found that illiteracy was correlated most closely
with poverty.
Example 14:
Jones (2017) found that illiteracy is correlated most closely
with poverty.
Example 13 shows that the writer believes that the finding should be understood within the
context of the single study. In Example 14, the writer implies that a wider generalization is
possible.
Example 16:
Several researchers have studied the causes of illiteracy
(Jones, 2017; Ferrara, 2015; Hyon, 2004).
In Example 15 and 16, attention is also given to what previous researchers did, but
one researcher is not named. Also note that these examples reference several
studies and / or an area of inquiry.
Example 18:
Illiteracy appears to have a complex set of causes (Jones,
2017; Ferrara, 2015; Hyon, 2004).
109
Past: reference to single studies
Present perfect: reference to areas of inquiry
Present: reference to state of current knowledge
110
The Unknown
Example 1:
1 It is known that several general anesthetics, including
barbituates, depress the bronchomotor response to vagus
nerve stimulation (refs). 2 However, the site of this
depresseion has not been determined. 3 To determine which
site in the vagal motor pathway to the bronchioles is most
sensitive to depression by barbiturates, we did experiments in
isolated rings of ferret trachea in which we stimulated this
pathways at four different sites before and after exposure to
barbiturates.
In Example 1, the site is unknown. The question is, “what is the site?” Thus, once the
unknown is stated, the question is determined. The authors write “the site…has not been
determined.” Other ways of stating the unknown outright are “has not been established,” “is
unclear,” “is unknown,” etc.
The unknown may also be implied rather than stated. One way to imply an unknown is to
state a suggestion or a possibility, as in Example 2 (sentence 10):
Example 2:
111
hypothesized that the nematode gene ceh-22 and the vertebrate gene
nkx2.5 perform similar functions. 12 To test this hypothesis, we examined
the ability of the zebrafish nkx2.5 gene (8,9) to substitute for the nematode
ceh-22 gene in transgenic Caenorhabditis elegans.
By saying that functional and molecular similarities suggest that the genes perform similar
functions, Example 2 implies that whether the genes actually do perform similar functions is
not yet known. Similarly, if you say something is possible, that implies that it is not yet
known.
Sometimes an Introduction funnels from the known to a problem with what is known. In this
case, the unknown should still be stated or at least implied, as in Example 3:
Example 3:
1 Metabolic alkalosis during exercise increases blood lactate
concentrations substantially beyond the concentrations
observed during exercise in the absence of metabolic
alkalosis (refs). 2 Conversely, metabolic acidosis decreases
blood lactate concentrations. 3 However, for these
investigations, alkali was ingested or infused, which is an
artificial situation. 4 More important clinically is the effect of
respiratory alkalosis, which occurs during exercise in a variety
of circumstances that involve increased respiratory drive. 5
These circumstances include interstitial lung disease and
congestive heart failure. 6 Therefore, in this study we asked
whether respiratory alkalosis during exercise, like metabolic
alkalosis during exercise, increases blood lactate
concentrations more than exercise alone does. 7 We used a
new biofeedback method by which ventilation, and thus
arterial PCO2 and pH, can be precisely adjusted independently
of metabolic rate.
In Example 3, the problem with the previous studies is that alkalosis was induced artificially.
The next sentence states the solution: using a more clinically relevant way of inducing
alkalosis. The implied unknown is that the clinically relevant situation has not been studied.
This implied unknown leads to the question.
112
Class 15
Introduction: The Opening
113
The Opening
The opening of the Introduction begins with a single sentence but typically encompasses the
first paragraph. The first sentence sets the topic by beginning with broad statements that
would generally be accepted as fact by the members of their reading audience. Authors then
seek to move their readers smoothly from these broad, general statements towards one sub-
area of the field, and then to the authors’ own particular topic. Typically, the last sentence of
the opening paragraph contains the point. Generally, that point is what large issue the
researchers wish to contribute to.
The Opening of a paper must accomplish three main goals: (1) set the topic and establish
the significance of the work, (2) introduce key terms and (3) identify the problem area the
research addresses. Read example 1:
In example 1, sentence 1 sets the topic of the paper (“Biological invasions”) and indicates
that they are important (“global problem” and “substantial economic costs”). Sentence 2
again notes that research on this issue is important. Sentence 3 narrows the focus to the key
goals to this type of research. Sentence 4 introduces key terms (“propagule pressure” and
“disturbance”). Sentence 5 identifies the large issue that the papers would like to understand
(how different processes / factors interact to regulate biological invasion).
As their name implies, General-Specific texts move from broad statements to narrower ones.
However, they often widen out again in the final sentence. In example 1, the paragraph
begins with a generalization (in this case, a statement of significance) and then narrows to
the individual factors that play roles in whether biological invasions are successful. In the last
sentence, the paragraph widens a little to state the major issue that ecologists would like to
understand about biological invasions.
114
Sentence Definitions
Let us start by looking at sentence definitions. A sentence definition is often a useful starting
point for a General-Specific paragraph. In a formal sentence definition, such as the example
that follow, the term being defined is first assigned to a class or group to which it belongs
and then distinguished from other terms in the class.
Read example 2:
To what class does the sole proprietorship belong? How is it different from other members of
the class? How would you define a partnership?
Read example 3:
A star is a celestial body that shines by itself and whose source of energy is
nuclear fusion occurring in its core.
To what class does a star belong? Can you think of other celestial bodies? How is a star
different than these?
Read example 4:
To what class does annealing belong? How is it different than other members of the class,
such as hammering or welding?
Notice that each of these definitions is completed by some form of restrictive clause and has
the structure shown here.
Generalizations
We have focused so far on starting General-Specific texts with definitions. Of course, this is
neither always necessary nor always appropriate. It is also possible to start with a factual
generalization. Suppose, for instance, the topic is “The English Language.” Now, if we were
to write on this topic as philologists, we might still open the text with a definition.
115
Read example 5:
However, in most other circumstances, it is more likely that we would start with a
generalization.
Read example 6:
Read example 7:
Although Chinese has the greatest number of speakers, English is the most
widely distributed language in the world today. This position derives from the
fact that English is widely taught as a second language in schools and widely
used in international communication.
Example
Example
Persistence is an attribute valued by many.
The PRESENT PERFECT is then used to show how the problem has been approached from
the past until the present day.
Example
Because different physical phenomena are at work during the fragmentation of a solid body,
it has mainly been studied from a statistical viewpoint (refs).
Example
Persistence has most often been studied in terms of cultural differences.
116