Writing Workshop: Choose Good Words

[Hemingway] has no courage . . . He has never been known to use a word that might cause the reader to check with a dictionary. -William Faulkner

Poor Faulkner. Does he really think big emotions come from big words? -Ernest Hemingway

We've arrived at the part of the writing workshop where we talk about revising. For the next several chapters, I will assume that you already have a bad draft of whatever you are writing, and you are ready to revise it into a better draft. These chapters are about polishing up words, sentences and paragraphs.

[Note: Most of the material from these writing workshop blog posts, plus a lot more never-blogged material, is now available in my book, The Writing Workshop: Write More, Write Better, Be Happier in Academia.]

By the way, please don’t feel bad if your writing process doesn’t proceed in a series of neat steps. This blog might give you the impression that writing proceeds in neat stages of plan, outline, draft, revise. But that's just a convenient way for me to organizing these posts. My own writing process is more like plan, outline, draft, outline, draft, revise, outline, draft, plan, draft, outline, draft, revise, draft, plan, revise, continue until I run out of time or can't stand to look at it anymore, submit.

The Curse of Knowledge

The biggest challenge of academic writing is bridging the gulf between the reader's knowledge and your own. You've probably thought about your research topic every day for years; the reader probably hasn't. But in order to communicate, you have to write things the reader can understand. 

Here's the problem: We all tend to assume that other people have the same knowledge we do.  This is the curse of knowledge, and it's just a fact about human thinking. Even if you are consciously aware that your readers haven't spent anything like the amount of time you have on this topic, you still probably overestimate what they know about it, because you are truly unaware of all the assumptions and background knowledge that inform your own thinking.

A few years ago in the writing workshop, the class was giving feedback on a writing sample submitted by my graduate student, whose name was James. Another student in the class read a sentence aloud.

‘There were equal numbers of area-congruent and area-incongruent trials at each of seven discriminability ratios.’

The student said, “So, this is an example of . . .” And I truly thought he was going to say, '. . . a perfectly clear statement' or ' . . . a sentence that everyone understands.'

Instead he said, “So, this is an example of a totally incomprehensible sentence. It might as well be in a foreign language.” What? I saw the other students nodding in agreement. What were they talking about? How could that sentence have been any clearer?

The moment stuck in my mind not just because it was ironic—I was teaching a writing class and couldn’t recognize unclear writing from my own lab—but also because it was such a clear demonstration of the curse of knowledge. The sentence looked fine to James and me because area-incongruent trials and discriminability ratios were things we talked about every day. But of course the other grad students in the workshop worked on other topics, so the words weren't familiar to them.

Over the years, I’ve seen plenty of incomprehensible academic writing. But the authors weren’t trying to be jerks and exclude their readers—they were just experts. In most cases, what they wrote was clear to them; they either didn't realize that it wouldn't be clear to readers, or they didn't know how to make it clear. This is the curse of knowledge.

If you are an expert, writing in your own area of expertise, you are actually teaching. Because you know so much more about the topic than your readers do, you can’t say what you want to say until you help readers build up the relevant concepts. That’s why it’s so important that you learn to describe your research at different levels. Your readers' level of background knowledge tells you what gap you have to bridge; it determines how much you have to teach your audience before they can understand what you want to tell them.

Because readers already have to work hard just to understand what you're talking about, you will naturally want to reduce their cognitive load (the mental work they have to do) as much as you can. This chapter is about how to choose words that make things as easy as possible for the reader. So when I say that one word is better than another, I just mean that it is easier for the reader to process.

Choose Good Words

Short is better than long.

When we read English, even silently, we use our verbal working memory (also called phonological working memory) to hold the words in our minds for processing. But our working memory can only hold as much as we can say in about two seconds. So the length of each word (in syllables) really matters. The shorter each word is, the more words you can hold in mind. And the more words you can hold in mind, the easier it becomes to interpret the meaning of the entire sentence.

[Note from Prof. Lisa Pearl: This is super-important when you’re trying to connect words together into the larger conceptual units that sentences encode. This is one reason why it can be hard to follow people who stutter and have a long delay between each word—the memory of that word is decaying as you wait for the next one, and you can’t weave it together into a larger conceptual chunk until you get enough word pieces.]

Keeping words and sentences short is one of the keys to making written language understandable. The Flesch-Kincaid Readability Tests—used since the 1970s to estimate the readability of any piece of writing—are calculated from nothing more than the number of words per sentence and the number of syllables per word. Shorter words and shorter sentences equal higher readability. So choose short words instead of long ones whenever you can.

Common is better than rare.

The frequency of any word is a number that expresses how often that word is used. Thanks to the internet, we now have huge databases like the one at https://www.wordfrequency.info/intro.asp where you can look up the frequency of any English word. (I’m not suggesting that you consult this kind of database when you write, but it illustrates what word frequency is.)

The more often we use a word, the more easily we can retrieve its meaning from memory. So the meanings of common words are easier to retrieve than the meanings of rare ones. Thus, you should choose common (high-frequency) words over rare (low-frequency) ones whenever possible.

Concrete and specific is better than abstract and vague.

Concrete words are specific, definite, and vivid rather than vague, general and abstract. Concreteness is closely related to imageability—the ease of forming a mental image associated with a word. The more concrete a word is, the faster and more accurately it is processed; concrete words are also easier to learn and remember than abstract ones. Just as there are databases of word frequencies, so are there databases of word concreteness like this one http://crr.ugent.be/archives/1330, where you can look up the concreteness of just about any word.

Concreteness is hugely important for clear communication. It’s also a huge problem for academic writers, because we write about abstract things like ideas, arguments, evidence and analysis. Learning how to explain abstract things in a concrete way is arguably the highest skill of academic writing.

At the word level, you can make your writing more concrete by choosing the most specific word available. For example, earlier today I heard a podcast where two journalists were interviewing a former senator. The senator was talking about what’s wrong with Washington, and she kept using the word ‘people.’ Her comments went something like this.

‘People want to know that you have their back. They want to know that you care about what’s happening to them, and you’re working to make their lives better. And people in my state understand that you can’t get anything done if you won’t compromise. That’s just the reality. But compromise is hard, because people are afraid to take tough votes. People from the deep blue and deep red states are afraid that their base will turn on them if they compromise. And you know, I get that. Because people don’t call my office and say, ‘Senator, please compromise.’ Compromise isn’t what gets people excited. And honestly, journalism is part of the problem, too. Celebrity gossip is always going to get more clicks than a budget bill, so even when we do get something done, people write about the celebrity. They don’t say, ‘Hey, the senate did something good today.’”

The senator wasn’t trying to be unclear, and she probably didn’t realize that she used the word ‘people’ to mean four different groups of people within just a few sentences. If she had replaced each instance of ‘people’ with a more specific word, her comments would have been much clearer:

‘People Voters want to know that you have their back. They want to know that you care about what’s happening to them, and you’re working to make their lives better. And people voters in my state understand that you can’t get anything done if you won’t compromise. That’s just the reality. But compromise is hard, because people senators are afraid to take tough votes. People Senators from the deep blue and deep red states are afraid that their base will turn on them if they compromise. And you know, I get that. Because people voters don’t call my office and say, ‘Senator, please compromise.’ Compromise isn’t what gets people the public excited. And honestly, journalism is part of the problem, too. Celebrity gossip is always going to get more clicks than a budget bill, so even when we do get something done, people journalists write about the celebrity. They don’t say, ‘Hey, the senate did something good today.’"

As this example shows, there is sometimes a tradeoff between length and frequency on the one hand, and concreteness on the other. ‘People’ is a shorter and more common word than most of the replacements (‘voters,’ ‘senators,’ ‘the public,’ and ‘journalists.’) But in this case I think the other words are better than 'people' because they are more specific. It's worth having slightly longer and less common words to avoid the problem of using the same word to refer to different things.

As academic writers, we can’t avoid abstract words completely. But we can swap in more concrete words whenever possible. For example, once you’ve established that the stimuli in your experiment were puppets and the participants were children, you don’t have to keep writing sentences like, ‘Each participant saw one of two stimulus items.’ You can help readers picture the experiments more clearly by writing, ‘Each child saw one of two puppets.’

Practice: Fancy words

The first column of the table below contains fancy (long, rare, or abstract) words and phrases; the second column contains plain (shorter, more common, or more concrete) words that could potentially work instead. It’s up to you as writer to decide when a bad word can be replaced by a good one; this exercise is just to help you identify places where substitutions might be made. We sometimes do these exercises as a game in the writing workshop—I read the bad word aloud; workshop members call out better substitutes.

Fancy word	Plain word
adjacent	next
advantageous	helpful
aggregate	total
alleviate	ease
allocate	give
alternatively	or
ameliorate	fix
anticipate	expect
apparent	clear
ascertain	learn
attempt	try
beneficial	helpful
component	part
conceal	hide
concerning	on
consequently	so
contains	has
currently	now
discontinue	stop
emphasize	stress
encounter	meet
equitable	fair
demonstrate	show
evident	clear
exclusively	only
exhibit	show
facilitate	help
frequently	often
however	but
inception	start
initial	first
monitor	watch
necessitate	need
nevertheless	still
notify	tell
numerous	many
objective	aim
obtain	get
option	choice
perform	do
permit	let
portion	part
possess	have, own
provide	give
purchase	buy
remain	stay
require	need
selection	choice
subsequent	next
subsequently	then
sufficient	enough

Positive words are better than negative ones.

In order to interpret a negative statement, the reader must represent some idea (e.g., This tea is delicious) and then negate it (e.g., This tea is not delicious). This two-step process places a greater cognitive load on the reader than the equivalent positive statement would (e.g., This tea is bitter). So you can often make writing clearer by recasting negative statements as positive ones.

English has many words that include negation or opposition as part of their meaning. When you revise, get in the habit of noticing prefixes that signal negation or opposition and asking whether the negative statement can be swapped out for a positive one. (Sometimes the substitution will work; sometimes it won’t.)

Practice: Restating negatives as positives 

Prefix	Negative	Possible substitute
DE-	deactivate	turn off, close
	decompose	rot, break down
	deconstruct	take apart
	decontaminate	clean, purify
	decrease	shrink
DIS-	disagree	argue
	disbelieve	doubt
	discontinue	stop
	dishonest	lying
	dishonor	shame
	disinfect	clean, purify
	displease	annoy
	distrustful	suspicious
IN-, IL-, IM-, IR-	illegible	sloppy
	illegitimate	bogus
	illogical	wrong
	immature	childish
	impatient	antsy
	imperfect	flawed
	inaccurate	wrong
	incorrect	wrong
	insignificant	petty
	inconsistent	off and on
	irrelevant	moot
	irresponsible	risky
	irregular	strange
NON-	noncomprehension	confusion
	nonconformity	oddity
	nonobvious	subtle
	nonresident	outsider
	nonstop	constant
	nontrivial	important
	notwithstanding	in spite of, still
UN-	unaided	alone
	unafraid	brave
	unanticipated	surprising
	unbelievable	shocking
	uncertain	dubious
	unconventional	fresh
	undamaged	whole
	unexpected	novel
	unfocused	scattered
	unforgettable	memorable
	unforeseeable	random
	unkind	mean, harsh, cruel
	unstimulating	boring
	unrelated	separate
	untrue	false
	unusual	rare

 

Explain Technical Terms as needed

Academic writing is full of technical terms. For experts, technical terms are convenient tools that actually reduce our cognitive load by packing a lot of information into just a few syllables. But for people who don't have the knowledge to unpack them, technical terms are a barrier to understanding

Consider the term ‘discriminability ratio,’ which appeared earlier in this chapter. I talk about discriminability ratios all the time, so it doesn’t sound strange to me. But I’ll assume that at least some people reading this book are not numerical cognition researchers, so I’ll explain what it means.

People and other animals share a perceptual system that allows us to tell different numbers of things apart. For example, if you look at two apple trees and one has just a few apples while the other has a lot of apples, it’s easy for you to see that the second tree has more. But if the two trees have similar numbers of apples on them, it’s hard.

How hard it is depends on how similar the numbers of apples on the two trees are—not their absolute difference, but their ratio. If one tree has twice as many apples as the other, it’s easy to tell them apart. But if one tree has only 10% more apples, it’s hard.

To tell things apart is to 'discriminate' them. Things that are easy to tell apart are said to be 'highly discriminable,' or to have 'high discriminability.' Things that are hard to tell apart have 'low discriminability.'

If one tree has 5 apples and the other has 10, that’s a ratio of 1:2, which is easily discriminable. The discriminability of 5 and 10 is the same as the discriminability of any other pair of numbers with a 1:2 ratio. In other words, 5 and 10 are as discriminable as 8 and 16 or 10 and 20 or 24 and 48. They all have the ‘discriminability ratio’ 1:2.

If one tree has 63 apples and the other has 70 (a ratio of 9:10), it’s much harder to see which has more. But all pairs of trees with a 9:10 ratio are equally discriminable. So if one tree has 45 apples and the other has 50, they are just as hard to tell apart as if one tree has 90 and the other has 100. Those pairs all have the ‘discriminability ratio’ 9:10.

When we want to study people’s accuracy at estimating numbers, we give them some trials with easy-to-discriminate ratios, and other trials with hard-to-discriminate ratios. So when we say that participants completed trials at each of seven discriminability ratios, it means that they were given trials at seven different levels of difficulty.

That explanation took 351 words (531 syllables). ‘Discriminability ratio’ might be a fancy term, but it’s only 10 syllables long. If I were writing a conference abstract with a 500-word limit, the explanation would take up way too much space. But if I were writing something like a handbook chapter for non-experts, it would be perfect.

You are an imprisoned chess master.

Think of it this way: You are an expert chess player locked in a prison cell all by yourself. One day you discover that Annie, the prisoner in the adjoining cell (whom you can talk to, but cannot see) is also an expert chess player. Hooray! You each draw a chess board on the floor of your cell, and you make origami chess sets out of black and white chewing gum wrappers. Now you can play. Naturally, you both describe your moves using standard algebraic notation—the system that all chess players use today. Your game with Annie sounds like this:

YOU: d4

ANNIE: Nf6

YOU: c4

ANNIE: e5

One day Annie gets a pardon from the governor and goes home. She is replaced by a new prisoner, Bridgette. Bridgette also loves to play chess, but she has been behind bars since the mid-1970s. Back then, she learned to play using descriptive notation. This system is clumsier than algebraic notation—it takes longer to describe the moves, and occasionally the same description can refer to two different possible moves. But this is the system that Bridgette understands, and you want to play with her, so you use it. Your game with Bridgette sounds like this:

YOU: Pawn to Queen 4.

BRIDGETTE: Knight to King Bishop 3.

YOU: Pawn to Queen Bishop 4.

BRIDGETTE: Pawn to King 4.

After a while, Bridgette completes her jail sentence and goes home. Your new neighbor is Cassidy—a person who knows how to play chess, but has never played seriously and doesn’t know any of the notational systems. Describing your moves to Cassidy is much more work than describing them to Annie or Bridgette. But if you want to play chess, that’s what you have to do. When you play with Cassidy, it sounds like this:

YOU: OK, first move . . . The pawn in front of my queen moves up two.

CASSIDY: OK, then the knight on my left moves up and toward the center.

YOU: OK, got it. Now the pawn in front of the bishop near my queen moves up two.

CASSIDY: OK. The pawn in front of my king moves up two.

Eventually, Cassidy is transferred to another cell block and replaced by David, who has no knowledge of chess at all. But since neither of you has anything better to do, you offer to teach him. Playing chess with David is a slow, difficult process full of misunderstandings and corrections.

YOU: OK so for my first move, I am taking the pawn that’s in front of the queen, and moving it forward two spaces.

DAVID: Wait, I thought a pawn could only move one space.

YOU: Usually that’s true, but on the first move it can move two spaces. Only on the first move.

DAVID: Oh, OK. And knights can jump over other pieces, right?

YOU: Right, they’re the only piece that can do that.

DAVID: OK, then I’d like to move my knight, the one on my left, up two spaces. So it will jump over the pawn and land in front of the pawn.

YOU: Don’t forget, it also has to move one square to the right or left. It can’t just move two spaces forward; it has to move in an ‘L’ shape.

DAVID: Oh yes. OK, then it will move one space toward the center. So it’s in front of the pawn that’s in front of the bishop. The bishop next to the king.

YOU: Got it. Now I’m moving the pawn that’s in front of the bishop by my queen, and moving it forward two spaces.

DAVID: Wait, you said you could only do that on the first move. This is your second move.

YOU: Not the player’s first move, the pawn’s first move. Each pawn can move two spaces, but only the first time it moves.

DAVID: Oh, OK! So can I move the pawn that’s in front of my king two spaces?

YOU: Yep.

DAVID: OK, I’m doing that.

All of those conversations communicated the same four moves. But if we count the syllables in each conversation, we see that you and Annie only had to exchange 9 syllables to describe those moves to each other. With Bridgette it took 22 syllables; with Cassidy it took 59; and with David it took 284.

Technical terms used by experts are like a notational system in chess. They are elegant and precise, conveying exactly the right information as compactly as possible. But you can’t use the terminology of your field with a novice, any more than you could chess notation with Cassidy or David.

Figuring out how to say something in non-technical terms is hard work. In writing the example above, I enlisted the help of my 13-year-old son, an avid chess player. He had no problem describing the moves in standard algebraic notation, but when we got to the descriptive notation, he was unimpressed. “Why did people ever say it that way?” He complained. “It takes so long. You don’t have to say ‘pawn’ when you’re moving a pawn. You don’t have to say anything—everyone knows it’s a pawn. It has to be a pawn. There’s literally no other piece that could move there.”

When we got to the conversation with Cassidy and I asked him to describe the same moves without using any form of notation, he found it difficult and tiresome. How can players focus on strategy or tactics if they have to waste all their time describing the moves? When we got to the conversation with David, my son got up and walked away in disgust.

I understand his frustration—I see it every week in the writing workshop. Talking about your area of expertise with non-experts is difficult. Non-technical terms take up a lot of space, and don’t convey meaning as precisely as the technical terms. But to bridge the gap between the reader’s understanding and your own, you have to speak their language. The curse of knowledge makes it hard to know which terms will cause your readers trouble, and the best way to find out what's unclear is to ask someone to read it and tell you. But in editing your own work, there are some red flags that you can learn to look for.

Ambiguous terms

If you know that different groups of readers may use the same term in different ways, you should definitely explain how you are using it. Words like 'theory,' 'bias' and 'significance' mean different things to scientists and non-scientists, so you have to be careful with them when writing for a general audience. Either avoid the confusing term altogether, or define what you mean by it.

Researchers in different fields also use words differently. For example, in my work on children’s number-concept development, I interact with people who study the brain and also with people who study education. In brain research, the term ‘number sense’ refers to an ancient perceptual system that humans share with other animals. In education research, ‘number sense’ means not only innate number perception, but also skills we learn in childhood, like counting and simple arithmetic. This situation is obviously ripe for confusion, and I try to avoid the term ‘number sense’ when I write. But when I have to use it, I define it.

Abbreviations

An abbreviation is a shortened form of a word or phrase (e.g., 'Prof.' for 'Professor'). An acronym is an abbreviation made from the initial letters of two or more words. If you want to get really geeky, you can distinguish acronyms from initialisms: Acronyms are made from initial letters and pronounced as words (e.g., POTUS for 'president of the United States'); initialisms are formed from initial letters but not pronounced as words (e.g., 'PhD,' 'FBI,' HMO). Some acronyms are so common that after a while, people treat them as words. This happened with 'radar' (radio detection and ranging), 'scuba' (self-contained underwater breathing apparatus), and 'laser' (light amplification by stimulated emission of radiation).

The reason to look out for acronyms, initialisms and other abbreviations is that they are often specific to a particular community. So experts and insiders know them, but non-experts and outsiders don't. The solution is simple: Define each abbreviation, acronym or initialism the first time it appears, unless you are pretty sure that your readers will know what it means. Of course, the curse of knowledge makes it hard to guess which abbreviations will be a problem for your readers, but you can take your best guess. Then, ask someone to read a draft and tell you which terms are unclear.

Nominalizations

Just as English has a lot of prefixes that can make words negative, we also have a lot of suffixes that can make things into nouns. The technical word for noun is 'nominal,' and the verb for making something into a noun is 'nominalize.' The new noun that you make by adding one of these suffixes to a word is called a 'nominalization'.

Some nominalizations are really helpful. For example, the suffix -ER turns a verb into a noun meaning the person who does that action; the suffix -EE turns a verb into a noun meaning the person who receives the action. So we can talk about ‘advisers’ and their ‘advisees,’ ‘employers’ and their ‘employees,’ ‘interviewers’ and ‘interviewees,’ and so on. It’s much quicker to say, ‘the interviewee was nervous, but the interviewer was kind’ than to say, ‘The person being interviewed was nervous, but the person performing the interview was kind.’

The suffixes -ER and -EE are familiar and easy to understand. But in other cases, nominalizations add a layer of abstraction without adding meaning. The two sentences below have the same meaning, but the second sentence is shorter and clearer.

Emily had the responsibility of supervision over ten research assistants. (24 syllables)

Emily was responsible for supervising ten research assistants. (19 syllables)

It’s good to get in the habit of noticing nominalizations and asking whether each one should be turned back into its original verb or adjective, because nominalizations are often bad words: longer, more abstract and less common than the original verb or adjective. Of course, sometimes the nominalization has a different meaning from the original verb or adjective and can’t be swapped out. For example, the substitution below doesn’t work.

I can give him a strong reference.

I can refer him strongly.

Practice: Turning nominalizations back into verbs

Suffix	Nominalization	Verb
-ANCE, -ENCE	acceptance	accept
	admittance	admit
	adherence	adhere
	attendance	attend
	governance	govern
	preference	prefer
	reference	refer
-ING	being	be
	doing	do
	writing	write
-MENT	abatement	abate
	advancement	advance
	agreement	agree
	allotment	allot
	argument	argue
	assessment	assess
	assignment	assign
	commitment	commit
	curtailment	curtail
	displacement	displace
	enhancement	enhance
	enjoyment	enjoy
	excitement	excite
	incitement	incite
	government	govern
	movement	move
	payment	pay
	replacement	replace
-TION, -SION	activation	activate
	admission	admit
	anticipation	anticipate
	cancellation	cancel
	citation	cite
	composition	compose
	conclusion	conclude
	construction	construct
	contamination	contaminate
	conversion	convert
	creation	create
	demonstration	demonstrate
	deviation	deviate
	discussion	discuss
	division	divide
	equation	equate
	intention	intend
	investigation	investigate
	invitation	invite
	nomination	nominate
	participation	participate
	pollution	pollute
	population	populate
	promotion	promote
	quotation	quote
	recommendation	recommend
	relation	relate
	resignation	resign

Practice: Turning nominalizations back into adjectives

Suffix	Nominalization	Adjective
-ANCE, -ENCE	ignorance	ignorant
	importance	important
	independence	independent
	innocence	innocent
	invariance	invariant
	magnificence	magnificent
	permanence	permanent
	prominence	prominent
	residence	resident
	reticence	reticent
	salience	salient
	significance	significant
	silence	silent
-ITY, -TY	ability	able
	capability	capable
	certainty	certain
	morality	moral
	mortality	mortal
	rarity	rare
	responsibility	responsible
	salinity	saline
	scarcity	scarce
	solidity	solid
	uncertainty	uncertain
	universality	universal
-NESS	bitterness	bitter
	dizziness	dizzy
	fuzziness	fuzzy
	goodness	good
	greatness	great
	happiness	happy
	illness	ill
	lateness	late
	openness	open
	roughness	rough
	sickness	sick
	steadiness	steady
	sweetness	sweet
	ridiculousness	ridiculous
	ugliness	ugly
	wellness	well
-Y, -CY	adequacy	adequate
	ascendancy	ascendant
	difficulty	difficult
	honesty	honest
	latency	latent
	legitimacy	legitimate
	modesty	modest
	normalcy	normal
	truancy	truant

There’s a reason that academic writing is full of nominalizations: When we make a verb or an adjective into a noun, we can zoom out and talk about the action itself (in the case of a verb) or the quality itself (in the case of an adjective) in the abstract. To take an example from earlier, the word ‘discriminability’ is a nominalization meaning ‘ease of being discriminated.’ It packs a lot of meaning into a few syllables, which is good for experts but bad for non-experts.

Practice: Getting rid of nominalizations

With nominalization	Without nominalization
There is an abundance of evidence leading to the suggestion that X.	There is abundant evidence to suggest X.
An adjustment was made to the protocol	The protocol was adjusted.
There was no anticipation of the non-replication.	No one anticipated that the finding would not be replicated.
A brief assessment of X was carried out.	X was briefly assessed.
An experimenter was on hand to provide assistance, in the event that such assistance became necessary.	An experimenter was on hand to assist if necessary.
Researchers tend to work under the assumption that X.	Researchers tend to assume X.
We made an attempt at identification of . . .	We attempted to identify . . .
We now have the capability of measurement of X.	We are now capable of measuring X.
We have a reasonable certainty that X.	We are reasonably certain that X.
The completion of the survey in under five minutes was grounds for exclusion from the analyses.	Data from participants who completed the survey in under 5 minutes were excluded and not analyzed further.
Our conclusion is X.	We conclude that X.
This is a demonstration of X.	This demonstrates X.
The first treatment group received the designation ‘T1’.	The first treatment group was designated ‘T1’.
Researchers made a determination . . .	Researchers determined . . .
What is the explanation for this observation?	How can we explain what was observed?
The implementation of the protocol was carried out by a research assistant.	The protocol was implemented by a research assistant.
The necessity of further research is obvious.	Further research is obviously necessary.
There was a predetermination of the arrangement of the stimuli.	The stimuli were arranged in a predetermined order.
There was a strong preference for X among the majority of participants.	Most participants strongly preferred X.
Participants showed a preference for X over Y.	Participants preferred X over Y.
There is a high probability that X is true.	X is probably true.
There is a prohibition against the disclosure of participants’ private data.	It is prohibited to disclose participants’ private data.
The utilization of such methods is an inadvisability.	It is inadvisable to utilize such methods.

Noun compounds and clusters

English lets us use nouns like adjectives, to modify other nouns. When two nouns appear together and the first one modifies the second one (e.g., 'bus stop'), it’s called a noun compound. 'Discriminability' is a nominalization; 'discriminability ratio' is a noun compound. Like nominalizations, noun compounds are an efficient way to pack information into a few syllables. But because they are just a couple of nouns stuck together, they don’t offer many grammatical clues to help readers figure out their meaning.

Take 'bus stop,' for example. We know from real-world experience that a bus stop is a place designated for a bus to stop and let passengers on and off. But if you didn’t know that, you could imagine several other meanings. A bus stop could refer to an object that is placed behind the wheels of a bus to stop it from rolling downhill—like a door stop, but for a bus. Or it could refer to one instance of the bus stopping, as in, ‘the fugitive slipped away during an unscheduled bus stop.’ Or it could refer to the police stopping a bus for a traffic violation—a traffic stop for a bus.

That’s just two nouns, but the fun doesn’t stop there. English lets you throw as many nouns on the pile as you want. And if you have some spare adjectives lying around, go ahead and throw them on, too. English is very flexible that way.

This flexibility can give rise to monster noun phrases, where a head noun is modified by three or four (or more) nouns and adjectives. These phrases are called ‘noun stacks,’ ‘noun strings,’ or ‘noun clusters.’ And if you think a noun compound like 'bus stop' has a lot of possible meanings, imagine how many possible meanings a longer cluster has. Here are some examples collected by technical writer Mike Pope (2011).

• data bound control table row action links

• failed password security question answer attempts limit

• reduced minimum OS partition space available requirement

[Note from Prof. Lisa Pearl: Another reason we’re often tempted toward noun clusters is that we have a hard word limit. Noun clusters are far more compact word-wise than the equivalent, easier-to-understand rephrasing. To see this, try busting any of those clusters above, and see how many words it takes you.]

Just as nominalizations flag possible bad language, so do noun compounds and clusters. To make your writing easier to read, learn to notice them, and ask whether their meaning is clear or ambiguous. If the meaning is ambiguous, try breaking up the compound or cluster to provide more clues about meaning. Start by moving the main noun from the end to the beginning. Then rearrange words, replace nominalizations, add prepositional phrases, and add hyphens to make the meaning clearer.

Practice: Busting clusters

Participant recruitment improvement training prioritization efforts	Our efforts to prioritize the training that improves the way we recruit participants.
A high value data exclusion minimization plan	a plan to minimize the number of high-value data we exclude
service level agreement achievement percentage.	percentage of service-level agreements achieved1
MHS has a hospital employee relations improvement program.	MHS has a program to improve relations among its employees.2
NASA continues to work on the International Space Station astronaut living-quarters module development project.	NASA is still developing the module that will provide living quarters for the astronauts aboard the International Space Station.2
underground mine worker safety protection procedures development	developing procedures to protect the safety of workers in underground mines3
draft laboratory animal rights protection regulations	draft regulations to protect the rights of laboratory animals3

Putting it all together

In his brilliant essay Politics and the English Language, George Orwell (1968) satirically revised a famous verse from Ecclesiastes by doing the opposite of what is advised in this chapter. The result seems to suck all the life out of the original verse.

Original verse:

I returned, and saw under the sun, that the race is not to the swift, nor the battle to the strong, neither yet bread to the wise, nor yet riches to men of understanding, nor yet favour to men of skill; but time and chance happeneth to them all.

Orwell’s rewrite:

Objective consideration of contemporary phenomena compels the conclusion that success or failure in competitive activities exhibits no tendency to be commensurate with innate capacity, but that a considerable element of the unpredictable must invariably be taken into account.

Like many other advocates of plain English, Orwell was frustrated with authors who use big words when they could use small ones, and fancy sentence constructions when they could use plain ones. Orwell cared about the politics of bad writing—the way it shuts readers out.  I love Orwell's work and I'm sure he's right that people sometimes write unclearly because they have something to hide. But that's not what I see in the writing workshop.

The problem I see people struggling with in the writing workshop is the curse of knowledge. It is hard to write clearly. It's hard to guess what readers won't understand, and it's even harder to figure out what you can say that they will understand. When you get rid of nominalizations, noun clusters and technical terms, it can feel like you are sacrificing elegance and precision—trading in the right words for clumsier, stupider words. Like my son with the chess notation, you may feel like throwing up your hands and walking away.

But I hope you won't give up, for three reasons. First, because your career depends on your being able to communicate  clearly. Almost no one knows as much as you do about your topic. Even the experts who review your journal articles and grant proposals need the equivalent of descriptive chess notation, rather than algebraic.

Second, it's an act of generosity to share your knowledge with others.

Third, when you get used to these word-level revisions, you may actually enjoy them. I do.  Replacing bad words with good ones is easy and pleasant, like doing a crossword puzzle or a sudoku. It feels like scrubbing the crud off my sentences, leaving them clean and shiny and beautiful. I hope it feels that way to you, too.

You can download 'worksheet' versions of the exercises in this post here.


Go to Next Post

Go to List of Writing Workshop Posts