Eliminating the verb "to be" and all variants, so-called copula-free text, is called English Prime. Words eliminated include: be, being, been, am, is, isn't, are, aren't, was, wasn't, were, and weren't.

Many people have found reason to criticize E-Prime, but you can use strict E-Prime as a temporary exercise, and afterwards to merely reduce use of the verb rather than eliminate it. Some usages of the verb are in fact the best way to express an idea, but only some: perhaps 1 out of 10 occurrences in many adults' writing.

Instead of "a prediction of…"/"the prediction of…"

say "if entity X predicts that…".

Basically, if you can use a verb, do so. On words that end with -ion, like "prediction", think twice.

Eliminate "think" and replace them with evidentials:

• I conclude
• I define
• I expect
• I generalize
• I hear (hearsay)
• I know by cultural means
• I know by internal experience
• I observe
• I opine
• I postulate
• I state
• I experience
• I remember
• I particularize

… my additions (the above list comes from Lojban):

• I suspect
• I believe
• I deduce
• I infer (kind of a synonym to generalize)
• I induce
• I abduce (kind of a synonym to infer)

Be a bit averse not just to "think" but also "believe". It can be vague, when it is possible to use a word that clarifies how you believe it, e.g. "I hear". Even for what I take as the most straightforward meaning of "I believe", I expect it's clearer to say "I expect" instead.

Linguists of English sometimes refer to using native (Germanic-derived) words instead of Latin-derived words as speaking a modified language called Plain English, even if there's more to speaking plainly than just that.

Using the long Latin-derived words lets you sound smart, needn't be a bad thing – akin to dressing well. It may also obfuscate that you don't know what you're doing, useful when you seek employment or publication. But when your highest desire is that someone understand and learn from what you are saying, speak Plain English. If scientists have a responsibility to reach out to society, speaking plainly is a high skill of science … and such language ought to be favoured in publications, when you think about it.

Consult the List of Germanic and Latinate equivalents in English.

(<2022-Mar-17> I'm not sure if the quote is about Native over Latin, or just simple language.) Some precedent for mandating simple language, from lawyers of all people:

In legal writing, David Mellinkoff, a professor at the UCLA School of Law, is widely credited with singlehandedly launching the plain English movement in American law with the 1963 publication of The Language of the Law.[15][16] In 1977, New York became the first state to pass legislation requiring plain English in consumer contracts and leases.[17] In 1979, Richard Wydick published Plain English for Lawyers. Plain English writing style is now a legal duty for companies registering securities under the Securities Act of 1933, due to rules the Securities and Exchange Commission (SEC) adopted in 1998.[18] In 2011, PLAIN (Plain Language Action and Information Network) published Federal Plain Language Guidelines.[19]

If plain language is so important in legal writing, then in scientific writing… given that journalists and even many antiscientists (like anti-vaxxers) now read abstracts for themselves and it has an impact on society… we should be able to mandate it too. The plainer it is, the more people might get started doing science themselves.

Science started out as something only the most fortunate members of society wasted their time on, which is why the language du jour is still so opaque, as if to keep out the dregs. I opine that everyone can contribute something to our public knowledge pool. Not just those getting an university degree, but those in high school. That would be possible if the language got points for being the opposite of opaque and there was a review process not reliant on the honor system, then if a kid finds something genuinely new, or even just replicates a study, s/he can just write something up, have the teacher edit it, and submit it, adding to the public reproducible knowledge pool of humankind.

NOTE: this is more about style than transparency, though good style makes it easy to read which makes it, in truth, more transparent.

Be allergic to "really" and "very" in particular. Instead of "really good", write "magnificent". Instead of "really gross", write "revolting". You may even want to avoid all intensifiers, like quite, rather, totally, remarkably, amazingly, most, actually, literally. As a tip-off, watch words that end in -ly.

The Emacs package artbollocks-mode flags them as weasel words.

Instead of "not doing something", write "avoiding something". Instead of "not having", write "lacking". Instead of "not picking up the soap from the floor", write "leaving the soap on the floor". It's less cognitive work for the reader to parse.

In "avoid", we have a verb, perhaps the simplest unit of language – avoiding is something you actively do, full stop. The negation "not" does not even exist in all languages. It takes at least one extra mental step to imagine "doing something, except not".

We can easily parse a single negation, but on any complex topic we'll soon speak in multiple negations if we don't watch ourselves. Then this habit pays off.

• Unpacking beliefs
• www.greaterwrong.com/posts/jbE85wCkRr9z7tqmD/epistemic-legibility

Is there a boilerplate preamble for science papers?

• Writing – just write well

• Communication problems
• Science quality assurance
• Perceptually uniform colormap

• Wilful ignorance
• Alternative Science Respectability Checklist
• Science quality assurance