关于如何写好paper的一点个人感想-转载

转载文章，一切CREDIT属于HIGHLY。

发信人: Highly (Highly), 信区: Faculty
标  题: 关于如何写好paper的一点个人感想
发信站: BBS 未名空间站 (Thu May 15 13:17:12 2014, 美东)

多谢大家鼓励！终于写完了。希望共同探讨。


                   On Scientific Writing



“The ancient historians gave us delightful fiction in the form of fact; the
modern novelist presents us with dull facts under the guise of fiction.” (
Oscar Wilde)


1. Adopt the Right Attitude

Just like everything else, before we start, we have to believe what we do
matters. If we can’t persuade ourselves, how are we going to persuade
others? Writing scientific essays should not be considered as an obligation,
or a pain we go through to create a nice-looking CV. It is true that we
have the duty to publish what we have done. “If you don’t publish, you
didn’t do it,” my PhD advisor used to tell me. But bear in mind that the
development of any discipline would not have been possible if people didn’t
know what one another was doing. Consider the current stage of a research
field as a mansion, and your publication as a brick that may appear anywhere
from the basement to the roof. Insufficient details make it difficult for
others to replicate your work; ambiguous phrasings can cause
misunderstandings. A single badly crafted brick may deteriorate into a leaky
passage.  Too many substandard materials can cause the mansion to collapse.

Let’s not treat essay writing as a burden. Maybe you are lucky to have
worked in a modern lab with cutting-edge equipment, or not that lucky like
me who sits surrounded by antiques and cockroaches. No matter what, after
conducting your experiments for months or years, now it is the time for you
to reach out to other smart people, to show your findings and gain their
respect. Your writing doesn’t have to impress people---although, if you can
, by all means do it! At least it should provide them with a fair chance to
judge your work. You deserve it.

The good part about writing a paper is that, unless you give up, it usually
ends up being published somewhere. Another form of scientific writing is to
prepare research proposals for funding applications. Given the fact that
both governmental and private budgets for non-clinical projects are
shrinking almost universally, it can be quite frustrating to spend months
collecting data and writing a proposal just to have it rejected. But don’t
be so pessimistic. There are several benefits of writing grant proposals. 
First, it helps you clear your mind and nail down what you really plan to do
. “I write entirely to find out what I'm thinking, what I'm looking at,
what I see and what it means.” (Joan Didion) You may have a vague idea
about a brilliant project. By writing it down you are forced to work through
your logic and examine its feasibility.

Second, it garners valuable opinions from your peers. A pair of fresh eyes
can help catch flaws in your design; somebody might have done similar work
that you are not aware of; you might have overestimated your ability to step
into a field without having acquired relevant expertise … It can be
painful to face criticisms, but you always learn something. Even if all the
reviewers misunderstood you, you know you need to tell a better story (to
the lesser men, if that makes you feel better).

Third, it makes things happen, not just being dreamed of. To collect
preliminary results, you can’t wait for the best time to come. You gotta
start it right now! Nothing inspires us more than an approaching deadline.
And remember, in the worst case that your application is dismissed, you may
have a manuscript ready to be submitted.

In short, attitude is important. "Love the writing, love the writing ... the
rest will follow." (Jane Yolen)


2. Hone your craft

Here I urge everybody to learn the basic techniques of creative writing. It
’s true that we need to write professionally, but a boring article may
discourage the readers from digging out the treasure buried in your
monotones and sloppy grammar. Right off the top of my head I could think of
several tips.

First, use active voice. I know, scientific writing is one place where
passive voice is indulged, if not encouraged. There have been constant
debates about whether active or passive voice better serves scientific
writing (like this one: https://cgi.duke.edu/web/sciwriting/index.php?action
=passive_voice). I don’t want to go in depth to discuss their pros and cons
. Here is an example from one of my old papers: “Activity evoked by low-
level background noise is reported to be more suppressed by inhibitory
inputs than is tone-evoked activity”. If I were to rewrite this sentence
now, it would be: “Inhibitory inputs generate stronger suppression on
activity evoked by low-level background noise than by tones.” The original
sentence has 19 words, whereas the revised has only 15 words. In addition,
it is clearer which is compared with which in the revised sentence. In my
experience, changing a passive voice into active usually introduces a strong
verb (e.g., generate), which is better than “is found, is observed, or is
reported to be”.

Second, expand your vocabulary. In scientific writing, we tend to use the
same words again and again. Next time when you want to write the verb “show
”, consider its synonyms such as “demonstrate”, “display”, “exhibit”,
“indicate”, or “suggest”. I’m not saying you should randomly choose a
synonym just to avoid repetitions. These words differ from one another and
depending on the context, one of them will be the most accurate. For example
, “display” and “exhibit” give a more passive feeling, whereas “
demonstrate” implies some kind of proof. “Suggest” sounds more active
than “indicate”. By doing so, your writing will slow down. But, here is
one of my favorite quotes: “Write quickly and you will never write well;
write well, and you will soon write quickly.” (Marcus Fabius Quintilianus)

Third, pay attention to the melody and vary the sentence length. It’s very
tedious to read an article that has one long sentence followed by another.
Short sentences are punchier, but too many in a row gives the impression of
lacking sophistication. The best is to have long sentences interleaved with
short ones, but by all means, avoid complex structures with layers of
meanings. It’s hard enough for others to understand your basic science. Don
’t play word games here. In fact, for more skilled writers they would even
vary the structures of adjacent sentences. Another commonly overlooked
aspect is what the word sound like. Many novelists believe that the sound of
a word is, at least, as important as its meaning. This might be a high
standard for scientific writers. I would say, the least you can do is to
stay away from mouthful phrases.

Fourth, avoid repetitions. Here is a perfect example. At the end of the last
paragraph, I initially wrote “the least you can do is to avoid mouthful
phrases.” Then I realize I have another “avoid” just below it. So I
changed one of them. Another idea I want to get across here is that you don
’t have to repeat in the text what is already there in the figure captions.
In the captions, we tend to write, “The x-axis is … The y-axis is … The
closed symbols are …” I have seen many authors who constantly repeat those
phrases in the Results. Not necessary. Another bad habit is to repeat the
entire results in the Discussion. The Discussion should really discuss. It
should expand. If readers forget about a result, they can go back to look up
for it.

Fifth, refrain from using adverbs or modifiers. Replacing a weak verb
followed by an adverb with a strong and accurate verb is the number one
golden rule of creative writing. For example, “He was walking leisurely in
the garden,” should be rewritten as “He was rambling in the garden.” We
also tend to say, “The change is very large.” Some people believe that, in
scientific writing, any change should be either statistically significant
or not. This rule is kind of hard to follow (note: delete “kind of” for
better writing), but comply with it whenever you can.

Last, use informative subtitles, especially in the Discussion. Like what I’
m doing here: “1, Adopt the Right Attitude”, rather than “1, The Attitude
”. Instead of saying “The Negative Level Effect”, make it clear that “
The Negative Level Effect Is Only Found for Localization in Elevation.”
Under each subsection, instead of plunging into the results, first use a few
sentences to tell the readers what to expect, such as in what order will
the data be presented. Some people even suggest adding a summary sentence to
the beginning or the end of every paragraph in the Results. Your choice. I
would always have a summary sentence at the beginning of each figure caption.

In short, never underestimate the power of language. If you aren’t careful
with what you say, people would wonder if you have been careful with what
you do. Unless you are in a hurry, treat every email as an opportunity to
improve your skills.


3. Tell a story

For research articles, different people write in different orders. I always
start with the straightforward part, the Methods, which merely needs to be
organized and presented in the clearest way. Then I describe the results.
Although I tend to think I understand the results by the time the experiment
is finished, there are always surprises waiting for me in the finalized
figures. Then based on the results, I select what I want to include in the
Introduction. Here is a tip: like writing a novel, instead of revealing all
the findings, adding a little bit of suspense to the Intro usually keeps the
readers more engaged, although they do know what to expect in general,
assuming they have read the Abstract.

Now it comes to the critical part, the Discussion. Many people tend to
organize the Discussion based on how they present the Results. I can’t say
this never works, but why not give it another thought? Ask yourself several
questions before you start. For example, “Why did we do this?” In fact, a
better question would be, “Why did we have to do this?” Using my PhD
advisor’s word, every study we do is a must-do; if it’s optional, we
should’ve found a better way to spend the taxpayers’ money. You can argue
that your study helps discover something new, or fill a gap between things
we’ve already known. A common mistake people often make here is the
assumption that anything unknown is worth studying, e.g., “We are the first
to …” Well, it’s true only if you are paying for the experiment out of
pocket. Otherwise, it needs to be a critical piece of information that
solves a long-lived myth, settles ongoing arguments, or links segregated
knowledge into a whole picture. Some prestigious journals may even want you
to elaborate how your field will change once people get to know your work.
Don’t be intimidated by this request. Changes rarely happen overnight. Know
your strength, and talk about it with confidence.

Truth is impartial. It shouldn’t depend on the person who pursues it. But
science is more than absolute truth. How individual researchers approach the
same topic from different angles, how they test their hypotheses and
interpret the results, make science colored with all kinds of human factors.
We try to write as impartially as we can, but meanwhile, don’t be afraid
of expressing your opinions---they might be wrong, but you’ll never know
without them being articulated. “Let who you are … what you believe, shine
through every sentence you write.” (John Jakes)

Now a bit off topic here. When addressing reviewers’ comments, there are
two typical tendencies. One is to be overly flattering in hopes that the
reviewers or editor will simply let it go. The other is to answer with
strong emotions, almost getting personal.  Neither is the appropriate
approach, especially the second one. I know you might be indignant for being
misunderstood or mistreated; you might be shocked at the reviewers’
hostility or stupidity, but it does you no good to start a heated debate due
to the nature of peer reviews. When you can’t exchange arguments with the
other party in a freely and timely manner, things are frequently
misinterpreted. Remember, don’t get personal. Just answer the questions, as
calmly and concisely as you can. They may have acted unprofessionally, but
you wouldn’t. And no need to be flattering either. Acknowledge their
expertise, and say that the manuscript has been greatly improved due to
their help. Note the last sentence is the key. It makes you and the
reviewers collaborators, not enemies.


All right! That’s all I wanted to say. In short, you have to love what you
do. And why not? Science is not only about inventing fancy machines or
conquering diseases. Through scientific activity we learn to free our mind,
we try to find out where we are from, and survive as better human beings.