John’s guide to tricky English grammar rules

The title is pretty self-explanatory: these are some rules and examples of tricky, confusing, frequently abused/ignored, or difficult-to-remember aspects of English grammar. Many of the example sentences are about science or medicine because I edit biomedical research manuscripts written by foreigners to improve their language and grammar to the level of native English speakers, so I’ve used their errors to teach from. I should also note that many of these issues are more properly classified under style or usage.

Table of contents
1. Serial commas and the “Oxford” comma
2. Numbers: Words vs. numerals
3. Hyphens and en dashes
4. Literally vs. figuratively
5. Apostrophe-S vs. S-apostrophe
6. Biannual or semiannual? Biweekly or semiweekly?
7. As such
8. Should everyday, sometime, anymore, anytime, etc. be written as one word or two?
9. Lay, lie, laid, lied
10. Who vs. whom
11. Is it okay to split infinitives?
12. Can since be used to mean because?
13. Can And and But be used at the beginning of a sentence in formal writing?
14. State-of-being verbs take adjectives, not adverbs
15. What does beg the question mean?
16. Between is not limited to relationships between two nouns
17. Farther vs. further
18. Disinterested vs. uninterested
19. Like vs. as
20. Parallelism with the word both or either
21. If X had happened vs. *If X would have happened (past perfect vs. conditional)
22. Into vs. in to, onto vs. on to
23. Eager vs. anxious

1. Serial commas and the inclusion of the final (or “Oxford”) comma
I start with this because it is important to me. Or, maybe, it’s important to me that people agree with me. In truth, those are probably the same thing. Serial commas are the commas that come in between items of a list. Perhaps you were taught that before the and or or preceding the final item of a list, you shouldn’t put a comma because the and or or fulfills the function of this comma in addition to alerting you that the last item is coming up. I think this is a bunch of hokum. Despite the absolute requirement of the final comma by nearly every style guide you can find, including Strunk & White, Garner’s Modern American Usage, the Chicago Manual of Style, the United States Government Printing Office, the American Psychological Association, the Purdue OWL,, Wilson Follett’s Modern American Usage,, the Harvard University Press, The American Heritage Guide to Contemporary Usage and Style, and the Oxford University Press, the minds of many schoolchildren seem to have been poisoned with the no-final-comma rule, much to the English-speaking world’s detriment.

The only reason there could be to omit the final comma is to save space, as most famously prescribed in various (print) newspapers’ style guides, such as the New York Times and The Times (of London). Some style guides, such as The Economist and The Guardian, say to omit the Oxford comma unless clarity requires one.

The problem is that omitting the final comma 100% of the time leads to awkward, confusing, and even ambiguous sentences, whereas inserting a comma before the last item is always at least as clear (see the appositive caveat below); is, in some cases, as the examples below show, clearer or at least easier to follow; and, in some cases, is the only way to definitively express the meaning of the sentence. According to definition, if it isn’t applicable at all times and in all situations, it isn’t a rule; it’s a preference, and a misguided one at that. Instead of requiring the Oxford comma be omitted except when it is clearer to include it (which amounts to a style preference with no logical backing), I think it is clearer, more practical, and more considerate to the reader to include the Oxford comma without exception. As Grammar Girl says, “Although the serial comma isn’t always necessary, I favor it because often it does add clarity, and I believe in having a simple, consistent style, instead of trying to decide whether you need something on a case-by-case basis. I also think using the serial comma makes even simple lists easier to read. Really, unless space is incredibly expensive, I can’t imagine why anyone would decide the best method is sometimes leave it out and sometimes add it in.”

a. The following sentences or sentence fragments do not have different or ambiguous meanings without the final serial comma, but they’re really bulky and are therefore easier to follow with a final comma:

The authors are affiliated with the Medical School, the Department of Cellular and Developmental Biology and Laboratory and Industrial Products, Inc.

If you can’t admit this sentence reads very awkwardly with so many ands and would be improved by the insertion of a final comma, you have much worse problems than following the wrong comma style.

This is characterized by the disruption of polarized tubular epithelial cell morphology, de novo mesenchymal gene expression and actin reorganization and increased cell migration and invasion.

There are four noun phrases separated by three ands, and not a comma in sight. Foolish and awkward. The middle pair of noun phrases (mesenchymal gene expression and actin reorganization) are both phenomena that happen for the first time (de novo) during the process in question, and the last two noun phrases (cell migration and invasion) are both modified by increased. Put a comma before the second and, and it looks like a normal sentence that an English speaker might be proud of writing.

EMDB was established by the Protein Databank in Europe (PDBe) at the European Bioinformatics Institute, the Research Collaboratory for Structural Bioinformatics (RCSB) at Rutgers and the National Center for Macromolecular Imaging (NCMI) at Baylor College of Medicine.

The other group of BMDMs was co-treated with CHX and IC for 2 h and 11 h, 4 h and 13 h or 6 h and 15 h, respectively.

A final comma would make the separation between these long, bulky noun phrases a little easier to follow.

b. Many sentences I’ve encountered are asymmetric and unbalanced because some items of the list are much longer than others and the last two items aren’t separated by a comma. This leads us to temporarily group the last pair of items together when they should be separate. For this reason, the lists are awkward and occasionally difficult to follow, and they disrupt the reader’s expectation of clarity, cadence, and flow.

The pits are then removed from the flesh of the fruits, dried and planted immediately.

pods with seeds, pods without seeds, seeds and seeds without the embryonic axis were harvested

The cells were fixed with 4% paraformaldehyde for 20 min at room temperature, rinsed and blocked with 20% goat serum for 60 min at room temperature.

fixed in paraformaldehyde, decalcified and embedded in paraffin

the position and shape of the occluder and of the residual shunt around the occluder, the ventricular size and heart systolic function should be observed

The criteria for diagnosing PA were based on the comprehensive elevation of histological and clinical symptoms, their alleviation by anti-rejection therapy and graft biopsy.

OCT can generate three kinds of images: traditional (similar to the ones generated by ultrasound), color and inverse (demineralized areas appear in hues of gray or black on a white background)

The aqueous layer (1 mL) was precipitated in 1 mL of isopropanol on ice for 1–2 hours, vortexed and centrifuged for 30 min at 4°C and 13,000 rpm.

Data collection and analysis, biopsy and surgery

The tissue was deposited on a single histological slide, identified and fixed with sprayed solution.

I think I understand the author’s point, agree and would like to share my understanding for review.

Thirteen adolescents aged 13–18 years with poorly controlled type 1 diabetes, 19 parents and 9 health-care professionals participated.

The problem with all of those lists is that the last and/or second-to-last item is much shorter than at least one other item in the list, leading our minds to group two separate items together because of their shortness and the lack of a comma separating them. A comma between each pair of items in the list would put all items on equal standing regardless of length, would give each item its own slot between commas or bookending the list, would make the separation between items clear, and would make each list more symmetrical, balanced, and measured. The imbalance created by the disparate lengths of the list’s items can be largely or even completely eliminated by the placement of a comma to mark the separation between each pair of items. The way they are, those lists often require re-reading so our minds can perform the mental separation that the authors were too dense to provide on the page. It is simply impossible to deny that those lists would be made clearer and less awkward with an Oxford comma.

c. These are good examples of how the consistent use of the Oxford comma can avoid temporary confusion as to the meaning of a sentence:

In addition to endogenously synthesized cholesterol, the absorption of dietary cholesterol and the reabsorption of biliary cholesterol in the small intestine also contribute to the regulation of the plasma cholesterol level.

If the author of this sentence is not a habitual Oxford comma user, the reader doesn’t know if endogenously synthesized cholesterol, the absorption of dietary cholesterol and the reabsorption of biliary cholesterol in the small intestine is a list of three items, or if the comma separates an introductory phrase from an independent clause beginning with two noun phrases. It turns out it was the latter. If the reader knows the author is consistent in his use of the final serial comma, there would (should) be no confusion.

Cases and controls did not differ significantly in age, weight, history of gastric reflux or GI bleeding, type of AB therapy or use of concomitant NSAIDs, anticoagulants or corticosteroids.

That is a ridiculous sentence that is difficult to follow because of the lack of Oxford commas before the last two ors. You might say, “Ah! But this is a strawman! The problem is the lack of semicolons, not Oxford commas!” No, that sentence is perfectly clear with Oxford commas but no semicolons:

Cases and controls did not differ significantly in age, weight, history of gastric reflux or GI bleeding, type of AB therapy, or use of concomitant NSAIDs, anticoagulants, or corticosteroids.

As is made clear by the Oxford commas, the only individual item in that list that contains commas is the last one. In such cases it is almost never necessary to separate the items of the list with semicolons.

After screening, wash-out of RAAS blockers if needed and informed consent, patients were….

I don’t know about you, but that sentence was temporarily confusing the first time I read it, and today when I browsed my ever-growing list of Oxford comma–needing sentences to find some more good ones to add to this guide, this sentence initially confused me again. I’d wager that few people would clearly understand this sentence on first read without the Oxford comma but that many people would (scientists, anyway) with it. It’s true that “if needed” could be surrounded by commas, but this is not necessary. What is necessary for ease of reading is the comma after it.

cyclosproine A (CsA)/tacrolimus (TAC), MMF and prednisone

That is supposed to be a list of three different treatments. If that isn’t clear to you, it isn’t because you aren’t a scientific researcher and aren’t familiar with the terminology; it’s because the authors left out a necessary comma. The three treatments are supposed to be an option of CsA or TAC, MMF, and prednisone. However, without the comma, it looks like it could be two treatments: CsA or TAC, or MMF and predisone (or even CsA+TAC or MMF+prednisone).

Heat-shock proteins play important roles in maintaining proteins in their functional conformations, stabilizing and preventing aggregation of proteins and refolding denatured proteins, which are important for cell survival in stressed conditions.

That is supposed to be a list of three roles played by heat-shock proteins, but it looks kind of like a single function (maintaining proteins in their functional conformations) followed by two sub-functions in apposition to that first function. In other words, it looks like you could replace the first comma with by and convey the same meaning. This ambiguity might not be completely eliminated by an Oxford comma here or in every other sentence containing a list, but it would make ambiguity much less likely.

d. The meanings of the following sentences are impossible to determine definitively without a properly placed final comma:

My favorite dresses are yellow, red and black and green.

It is impossible to know if the last two dresses were a red one and a black and green one or a red and black one and a green one. The Oxford comma easily solves this ambiguity. It’s true that English has no (perfect) way of clarifying a similar sentence with only two dresses (“My favorite dresses are yellow and red and black”), but that doesn’t mean we shouldn’t clarify three-item lists with Oxford commas when we do have an opportunity to.

The team focused on hiring and training, setting and meeting deadlines, reorganization and expansion and visibility in the retail market.

Which efforts were focused on the retail market? Visibility, or expansion and visibility? Everything the hypothetical team or company does isn’t necessarily directed at the retail market, so some of their efforts could be business-to-business, and some activities could simply apply to internal affairs. Without the final comma, it is not clear what the expansion and visibility efforts refer to.

Plant peroxisomes play a physiological role in the biosynthesis of the signaling molecule jasmonic acid, β-oxidation of indole-butyric acid and sulfur and polyamine metabolism.

Which process does the sulfur belong to? Is it part of the β-oxidation, or does it go with polyamine metabolism? You have to know what β-oxidation is to know that it can’t be applied to sulfur, but this comes from technical knowledge, not from anything the sentence conveyed. Without prior, specialized knowledge, it is not possible to determine the meaning of this sentence as written. I know no sentence is written in a vacuum, and you have to have some knowledge of the context and the meaning of the words to understand any sentence, but that doesn’t mean the sentences shouldn’t be punctuated properly to express their meaning correctly regardless of the audience. The above sentences don’t convey their intended meaning regardless of your specialized knowledge, and the reason for this is their lack of the final comma.

HCC subgroups were stratified according to tumor stage, vascular invasion or lymph node metastasis or tumor size.

This is yet another real-life example from a paper that I have lost track of (if you find it, send me the URL so I can cite it). Vascular invasion and lymph node metastasis are supposed to be two alternate criteria used to stratify subgroups, and tumor size was the last one. Without the serial comma, it is impossible to express this definitively.

Here are two other real-life examples whose sources I’ve lost track of:

document content, legal aspects and procedures and registration

PMCA1, NCX1, and NCKX4 could be the targets of miR-714, miR-762 and 135a*, and miR-712*, respectively.
[This last one did have both Oxford commas, thankfully. It’s a great example of how they can be absolutely necessary.]

In summary: just insert the damn Oxford comma. It’s not a waste of space, it’s usually clearer, it makes the list more symmetrical and balanced, and once in a while it will be the only way to clearly convey your meaning.

Finally, let’s talk about the appositive caveat I mentioned above. Wikipedia’s article about the serial comma gives a good example of the ambiguity that often results from forgoing the final comma…

Consider the possibly apocryphal book dedication quoted by Teresa Nielsen Hayden:

To my parents, Ayn Rand and God.

There is ambiguity about the writer’s parentage, because Ayn Rand and God can be read as in apposition to my parents, leading the reader to believe that the writer refers to Ayn Rand and God as his or her parents. A comma before and removes the ambiguity:

To my parents, Ayn Rand, and God.

…and a similar example of ambiguity that can result from including the final comma:

To my mother, Ayn Rand, and God

The serial comma after Ayn Rand creates ambiguity about the writer’s mother, because the proper-noun phrase Ayn Rand could be read as in apposition to my mother (with the commas fulfilling a parenthetical function), resulting in the interpretation “To my mother (who is Ayn Rand) and to God”.

While it is true that the latter ambiguity could be solved by the author having some sense and including or omitting parentheses and an additional “to” to improve clarity, this doesn’t change the fact that it is at least possible to introduce ambiguity with either the omission or the inclusion of the Oxford comma due to the comma’s other function as the offsetter of appositives.

Therefore, this appositive problem solves nothing in the Oxford comma debate, leaving the Oxford comma with a score of Everything Else–0–1 against omitting the comma. In this sense, one can conclude that the Oxford comma is “always at least as clear”, as I wrote above, if not in every case, then at least with every issue.

Clearly, the inclusion of the final comma in lists is more versatile—is, in fact, universally appropriate—whereas the absent-comma option quite frequently yields sentences that are harder to follow or even impossible to interpret, making it less appropriate some percentage of the time and therefore not suitable to be called a rule at all. We therefore do have a serial-comma rule that we can follow all the time, so it is foolish to institute a personal preference of including or omitting it on a case-by-case basis.

2. When to write out the word for a number and when you can use the numeral
Unfortunately, there is no consensus on this topic, and there never will be. It’s important not to find one person or web page who tells you what you want to hear and go with that. For instance, some people who act like they know what they’re doing don’t even know the definitions of the words number and numeral. (Those morans got it exactly backwards. I’d be embarrassed if I were them.) A number is an abstract concept for a quantity, and a numeral is a written symbol for expressing that quantity. Roman numerals and Arabic numerals are two examples of different ways of expressing the same mathematical concepts (numbers). A word, made up of letters, is another way to express a number in writing.

In the most formal writing, it is basically never appropriate to use a numeral. If you’re doing that kind of writing, you’re probably not reading this because you’re already an accomplished writer, grammarian, and/or editor. Other style guides allow numerals for years and really long numbers (in the millions or billions, for example).

In scientific or technical writing, different publishers have different preferences, but those might not always be available or exhaustive. In the absence of an available style guide or clear rule for your specific needs, here are the rules that I think are the most agreed-upon and best for when numerals are allowed in formal scientific writing, so that you don’t have to spend too much time worrying about whether you’re being formal or consistent enough:

  • Use numerals for all numbers with more than one digit. This includes decimals and fractions.
  • Any time you use a numeral for a certain noun (for instance, when that noun has a quantity of 10 or more), you can use a numeral for the same noun in the same sentence, even if the other number is less than 10. Example sentence:

    The two largest groups contained 12 and 8 people, respectively.

    Other things being equal, you would normally write eight instead of 8, but “people” is modified by both 12 and 8 in this sentence, so to be uniform, it’s okay to break this rule and write the smaller number the same way you write the larger number. On the other hand, “groups” is only modified by one number (two), so it must be written as a word. On the third hand, it is acceptable to maintain document-wide uniformity and write eight the same way you do every other single-digit number in every other sentence.

    A similar type of situation is the use of an integer and a decimal together:

    a 1.7–2-fold increase

    Normally you might write 2-fold as two-fold (though there are style guides that treat fold as a unit like % or other measured quantity), but in combination with—even in the same sentence as—that decimal, you must use a numeral.

  • An abbreviated unit (kg, mM, h, %, etc.) requires a numeral; inversely, spelled-out numbers require the unit to be written out in full. This is only relevant at the beginning of sentences. That is, if you have to spell out the word at the beginning of the sentence, you cannot use an abbreviation for the unit. This might feel a little awkward when you have to write Ten microliters of PCR product was added, but it looks a hell of a lot better than beginning a sentence with 10 μL. This often results in having two equally acceptable alternatives:

    Ten microliters of PCR product was added…
    PCR product (10 μL) was added…

  • When the number modifies any measured or counted unit of time, it is acceptable to use a numeral. Often units of time are abbreviated, but you or the publisher might not want to abbreviate them, such as hour, day, week, or month. This might only be relevant to scientific publications, where many numbers and measurements are used, whereas other writings don’t contain a lot of numbers and would look fine with occasional numbers written as words. In scientific papers, this rule helps avoid the awkwardness of using abbreviations in such examples as “at the 1-mo time point” or “the 3-d samples”. It should be acceptable to use a numeral with a non-abbreviated unit, like 1-month and 3-day, which is clearer and less weird-looking, is often more uniform with all the other single-digit numerals used in the methods and results sections, and is consistent because units will always be preceded by numerals.

    Other counted nouns in scientific papers, such as group or mouse, are not units, so their numbers follow the other rules of formal writing.

  • In the name or label of a group, time, disease, or several other things (note that these come after the noun):

    type 2 diabetes
    day 2 [the publisher might require number-words before units of time, like two days, but never in the label of a time point]
    group 1
    row 2
    Section 3

3. Hyphens and en dashes
The functions of the hyphen, the en dash, and the em dash can be summarized easily: the hyphen compounds, the en dash connects, the em dash separates. Therefore, if you don’t wish to compound two things into one, then don’t use a hyphen.

a. The function of the hyphen

The hyphen is a compounder or combiner. It is very versatile and is woefully underused in the modern English-speaking world.

  • When a two-word phrase consisting of an adjective and a noun is used as a single adjective, it should usually be hyphenated:

    one-way street
    real-world example
    5-mL solution

  • The hyphen between the adjective and the noun can be unnecessary or even incorrect when the adjective–noun pair is a single, two-word entity, almost like a proper noun or a name:

    Nervous system development
    Renal tubule formation

  • Also, two nouns that are used as a single descriptor (adjective) can often be hyphenated, though I am unable to discern a rule defining exactly when a hyphen is appropriate between the two nouns and when it isn’t. (My rule is: it’s appropriate when it’s clearer.) Often we might think of the first noun as fulfilling the function of an adjective, the second noun as a noun, and the combination of the two as a single, compound adjective.

    Air traffic was so dense that afternoon that air-traffic control could hardly cope. (Hat tip: BBC.)
    Density-gradient centrifugation
    Insertion-mutant populations

    She has been with her computer-programmer boyfriend for two years.
    I have a lot of science-nerd friends.
    The DNA-methylation differences were more subtle than the protein-phosphorylation ones.

    This doesn’t seem to be always appropriate or necessary, though:

    Skeletal muscle cell differentiation
    Cell cycle regulation (I also often see cell-cycle regulation)
    Data analysis program (but data-analyzing program)
    Brain stem reflexes
    Flow cytometry analysis (though I strongly advocate flow-cytometric analysis, which is noun–adjective, so there’s that)

  • Infinitive phrases and prepositional phrases almost always must be hyphenated when they precede their noun but must remain un-hyphenated when they follow their noun:

    The teacher gave us some easy-to-remember rules.
    The rules were easy to remember.
    The Atlanta-to-Houston flight
    The flight from Atlanta to Houston
    He had a lot of built-up anger.
    His anger built up for a long time.
    The on-call physician is Dr. Smith.
    Dr. Smith is on call.

    Here’s an example that’s always bothered me. Why do radio stations claim they are “new”? Or that they are the “new #1 hit music station”? I know stations that have literally been around for four or five years that still use that same damn slogan. But the grammarian in me thinks, Well, if you hyphenate the words properly, the statement could still be technically correct…

    A new-#1-hit music station is a music station that plays new #1 hits. (Presumably they play other things, too, seeing as how there is only one #1 hit at a given time.)
    A new, number-1, hit-music station is a radio station that plays hit music, is new, and is the number-1 station in town.

    If you can point to a single radio marketing employee in the entire United States who knows the difference between the two or could punctuate either one of those sentences properly, I have a newborn unicorn and a shiny pot of gold to give you.

  • Adverbial modifiers (adverb–adjective or adverb–adverb combinations) should not be hyphenated, except in cases where the adverb can also be an adjective in other situations (e.g., well, better, best, fast, high, hard, little, many others) and the compound modifier precedes its noun:

    A highly talented violinist
    A finely tuned machine
    The most often recommended remedy [adverb–adverb–adjective!]
    The least easily recognizable member of the group
    A fast-acting drug
    The best-recommended remedy
    Over-broad but overly broad (a preposition made into an adverb!)
    A poorly defined mechanism
    A well-defined mechanism [well-defined comes before the noun, so it should be hyphenated]
    A mechanism that is well defined [well defined comes after the noun, so it cannot be hyphenated]
    A relatively unknown protein
    A better-known protein
    A protein that is better known
    The best-known member of the group
    He is best known for his accomplishments in…
    The well-known chef opened a new restaurant.
    The new restaurant’s chef is well known.

  • When numbers like the following are written out in words, they must be hyphenated (the word “and” is also verboten!):

    One hundred sixty-four
    Three thousand four hundred ninety-nine

  • Any usage of the prefixes self or ex must be hyphenated:


  • When a prefix ends with the same letter that the base word begins with, or the non-hyphenated compound word is likely to be read wrong for another reason, it’s usually best to hyphenate the compound word:

    anti-inflammatory instead of antiinflammatory (though proinflammatory is okay)
    pro-oncogenic instead of prooncogenic (though antioncogenic is okay)
    de-ice instead of deice, which is never okay
    re-sign instead of resign; these are two different words with different meanings
    post-translational looks a lot better than posttranslational, though I guess both are fine
    reelect and re-elect both seem acceptable

  • Dangling hyphens are good to use when multiple prefixes precede the same base noun but you don’t want to repeat the base noun every time, or when you want to attach a single prefix to multiple base nouns but don’t want to repeat the prefix every time.

    macro- and microeconomics
    pre- and post-WWII
    Ron Paul’s anti-Obama, -McCain, -Clinton, and -Bush

    I particularly liked this example from the EDline editors mailing list, cited by Wikipedia:

    …a large number of adjectives…were used to describe [ships protected by iron or steel armor]: iron- or steel- or armor-plated, -cased, -clothed, -sided, and many others….

That Wikipedia article on the hyphen is quite good and long, so I recommend reading it for a fuller description and explanation of its uses.

b. The function of the en dash
You thought the hyphen was underused; most people don’t even know the en dash exists! Such a shame.

The en dash (–, HTML character code – or –) is a connector and a comparer. It’s the neglected middle child between the hyphen and the em dash. (The em dash is what we normally think of when we think of a dash, —.) So named because it was roughly the width of the capital letter N in many fonts, the en dash is used to signify a range, to connect things, and to replace the hyphen in compound adjectives of more than two words (see part c.) Wikipedia has a really good article on dashes.

  • In the more traditional, first two uses, the en dash replaces the word to or and:

    allowed to grow for 2–3 days, not 2-3 days, because you mean 2 to 3
    dose–response curve, not dose-response curve, because it is a dose-and-response curve; there is no such thing as a dose-response
    cell–cell communication, not cell-cell, because you mean cell-to-cell; there is no such thing as a cell-cell. Remember that the hyphen would combine the two words into a single cell-cell, whereas the en dash connects two different cells.
    Atlanta–Houston flight, not Atlanta-Houston flight, because it is an Atlanta-to-Houston flight; there is no such place as Atlanta-Houston. The hyphen would combine the two places into one region, like Alsace-Lorraine or Schleswig-Holstein; the en dash connects these two separate places.
    blood–brain barrier; there is no such thing as a blood-brain
    President Jimmy Carter (1977–1981)
    the Supreme Court’s 5–4 decision
    a score of 31–27

  • Lastly, the minus sign looks identical to the en dash on most computers and word processors, so it is best to write negative numbers with an en dash instead of a hyphen:

    a change of –20%

    It is important to note that if you’re giving a range of negative numbers, or any other times it might be confusing to the reader, you can’t use the en dash to specify a range:

    –15 to –20 (negative fifteen to negative twenty)
    –15 to 20 (negative fifteen to twenty)

c. Compound modifiers that require an en dash to perform the function of a hyphen

  • The following types of constructions require an en dash and cannot use a hyphen because one side or another of the compound adjective has two words or already contains a hyphen:

    embryonic stem cell–focused research [With a hyphen, it would be conveying the idea of the noun phrase “cell-focused research” being modified by “embryonic stem”, which doesn’t really make sense.]
    germ cell–derived
    high pressure–sensitive component
    mouse organelle–specific component [It is specific to mouse organelles, not a mouse component that is specific to organelles]
    non–government-based [Meaning not government-based]
    maltose-binding protein–tagged antigen [Meaning an antigen tagged with maltose-binding protein, not an antigen that is maltose-binding and is protein-tagged]
    small RNA–dependent scaffold [the scaffold isn’t small and dependent on RNAs; it is dependent on small RNAs]
    inter–stress fiber space [I don’t know what “inter-stress” would mean, but “inter” is a prefix to a two-word phrase, so it must be followed by an en dash, not a hyphen]

  • Contrast the first example with a re-writing that uses “ESC” as an abbreviation for “embryonic stem cell”. It has the same meaning but requires a hyphen, not an en dash, because the hyphen is preceded by only one word:

    ESC-focused, not ESC–focused

    I’m not 100% positive that this is right, but I think it’s best to use a hyphen instead of an en dash in these types of phrases where the full term and its parenthetical abbreviation (or a clarifying word in parentheses) precede a hyphen, if and only if they are both one word:

    MitoTracker (MT)-stained cells
    tyrosine (Tyr)-phosphorylated proteins
    vehicle (DMSO)-treated rats

    I don’t think it would be wrong to use an en dash instead in those examples.

  • If your phrasing calls for a dangling hyphen or en dash, simply use them the way you normally would if they weren’t dangling:

    the high pressure– and heat-sensitive gauge, not the high-pressure- and heat-sensitive gauge
    the angiotensinogen II– and VEGF-induced effects 

  • There’s also the classic example of Bart Simpson asking the following to Mrs. Krabappel:

    How would I go about creating a half-man–half-monkey–type creature? Or: How would I go about creating a half-man, half-monkey–type creature? 

    Either way uses an en dash before “type”.

The take-home message, which bears repeating: The en connects, the em separates, and the hyphen compounds.

4. Don’t use literally when you literally mean figuratively
This is one of my biggest pet peeves in the world because it is so ridiculous. Why do people say “literally” when they mean its exact opposite, “figuratively”? Here are some examples I’ve encountered that I can remember:

My boyfriend literally hit the ceiling.
We have literally created a monster.
This is literally a tortoise–hare situation. (Michael Wilbon, Pardon the Interruption, December 14, 2009, on the late-surging Chargers posing a potential threat to the 13-0 Colts, should they meet in the playoffs.)

People, please learn what the word “literally” means, and don’t use it foolishly.

5. When to add apostrophe-S (’s) after a noun, and when to add only an apostrophe
Despite what complex and specific rules your grade-school textbook or teacher might have tried to make you memorize, only two simple rules are needed to determine whether to add ’s or just the after a noun that ends in s:

If the word is one syllable, always add ’s.
If the word is multiple syllables, then spell it how you pronounce it.

Don’t let anybody tell you those rules are insufficient or too lenient. If a general prescriptivist like me tells you to ignore the other pseudo-rules that people may throw at you, you know it’s OK to. Many times, we would add ’s to a two- or three-syllable word because the word ends in a hard S (the voiceless S or voiceless alveolar fricative or [s]) or in spite of the fact that it ends in a soft S (or [z] or voiced alveolar sibilant fricative) because it is short enough to sound better with that extra ’s syllable added to the end of it.

Some examples:

Charles’s (Charles always counts as one syllable, so Charles’ is always wrong)
Dickens’s (I’d prefer to pronounce it this way, possibly because I err on the side of adding the ’s to proper nouns whenever it seems remotely reasonable)
Mr. Rogers’s Neighborhood and Mr. Rogers’ Neighborhood are equally correct because you could pronounce it either way
Socrates’s sounds better than Socrates’ in every phrase I utter to myself, but I can’t really explain why
the class’s desks
the students’ desks
Illinois’s constitution (you would add the s sound, which is absent in the state’s name, so Illinois’ seems pretty foolish and counterintuitive to me)
Mr. Jones’s car
the Joneses’ home
men’s clothing (mens is not a word, so mens’ is always wrong)
Venus’ fly trap and Venus’s fly trap seem equally correct in writing, though we only pronounce it the first way. But in other Venus possessives that aren’t common names of plants, we would add the extra ’s syllable in speech. You wouldn’t pronounce “Venus’ orbit” identically to “Venus orbit”, would you? No. So add the ’s.
for goodness’ sake probably never needs an ’s; we certainly wouldn’t pronounce it that way
for appearance’ sake and for conscience’ sake are never anything remotely approaching acceptable, despite what the idiots at the New York Public Library’s Guide to Style and Usage would have us believe. If it doesn’t end in s, then of course you have to add ’s to make it possessive!

6. Should I use biannual or semiannual? Biweekly or semiweekly?
This is an example of when Latin roots do matter in English usage. At least, they should.

The Latin prefix bi- means two, and the prefix semi- means half. Therefore, a biannum is two years and a semiannum is a half-year. A biweek is two weeks (a fortnight!) and a semiweek is a half-week.

From this it is easy to see that biannual means happening once every two years, whereas semiannual means happening every half-year (twice per year). Biweekly means happening once every two weeks, and semiweekly means happening once every half-week (twice per week). The same for bimonthly and semimonthly.

It’s not that hard to remember, and not that hard to understand. We have two perfectly clear prefixes that mean two different things, so use them correctly. Would you want to publish a paper that said a cancer drug should be administered semimonthly when you meant every two months? Increasing the frequency of drug administration 4-fold might very well kill the patients. Or, in the other direction, decreasing the frequency of administration of a drug 4-fold might eliminate the drug’s effect. What would people think of you if you couldn’t get such a basic difference straight? What would you think of someone who couldn’t learn the meanings of two simple words and use them correctly in a scientific paper?

7. As such
As such is not a generic synonym of “therefore” or “consequently” or anything like that. It is bound by one specific grammatical rule: the such has to refer to an earlier noun or noun phrase. In other words, before writing as such, ask yourself: “As what?” The such is the answer to what. If the previous thought specifies a noun that such would refer to, then it is okay to write as such; if not, then use “therefore” or something similar.

Correct: I am a scientific editor for non-native English speakers, and as such [as what? as an editor for non-native English-speaking scientists], I’ve collected mainly biology-related examples of grammatical errors.
Incorrect: I am a scientific editor for non-native English speakers, and as such, my examples of grammatical mistakes are mostly biology-related. [This doesn’t make sense because “my examples” does not refer to any antecedent noun.]

Correct: She is the only player on the team who is any good at pitching. As such [as what? As the only one who’s good at pitching], she doesn’t get to play shortstop or third base, her favorite positions, as much as she wants.
Incorrect: No one else on the team is good at pitching. As such [as what? as the concept of being good at pitching? as the concept of being bad at pitching? as no one else? there is no antecedent that “such” could refer to], she doesn’t get to play shortstop or third base, her favorite positions, as much as she wants.
Incorrect: She is the only player on the team who is any good at pitching. As such, a girl with a much weaker arm usually has to play third base. [This doesn’t make sense because “As such” does not connect “a girl with a much weaker arm” to any preceding noun.]

In summary, as such connects (equates) one noun to a previous one. If the two nouns aren’t identical and the statement you made about the first noun doesn’t lead directly to the second statement being true, then as such is wrong.

8. Should I write everyday, sometime, anymore, and anytime as one word or two?

Let’s take these in alphabetical order.

  • any more vs. anymore
    The single-word version anymore has a temporal meaning, expressing “still”, “now”, or “nowadays”. It is an adverb that is synonymous with “any longer” and usually appears at the end of a clause: 

    I don’t like tea anymore.
    I hope we don’t have to attend those boring meetings anymore.
    I wonder if it will be cold anymore this spring.

    In contrast, the two-word adjective phrase any more has a quantitative meaning, referring to the amount or number of stuff or things under discussion:

    I don’t want any more tea.
    I hope we don’t have any more boring meetings.
    I wonder if we will have any more cold days this spring.

    Take-home message: if you could replace it with “any longer”, the one-word version is correct; if you are referring to a quantity or number of a specific noun, the two-word version is required.

  • any one vs. anyone
    This one is easier because anyone is synonymous with “anybody” and “any person”, whereas any one always refers to “any single one of [some nouns]”: 

    I don’t know anyone here.
    I don’t know any one of those people.
    Does anyone still watch that show?
    Is any one of those shows worth watching?

  • any time vs. anytime
    Lynn Gaertner-Johnston wrote a guide to any time vs. anytime that should be sufficient for everyone and every context

    The first distinction to remember is that any time is an adverb phrase used to mean any amount of time:

    Do you have any time to meet tomorrow?
    Sure, that won’t take any time at all.
    I haven’t spent any time practicing lately.

    In contrast, anytime is an adverb used to mean whenever or at any time:

    I can meet with you anytime tomorrow.
    That should be easy to do anytime.
    Anytime I can practice, I do.

    The second distinction between these two adverbs is that following the preposition “at”, only the two-word any time can be used. Because of this rule, the two-word any time can mean any length of time (distinction 1) or, only after “at”, any point in time (whenever). The take-home message: if preceded by “at”, only any time is acceptable; otherwise, use any time to mean any length or amount of time and anytime to mean any point in time or whenever.

  • every day vs. everyday
    This one should be easier than people seem to make it because the two-word every day is an adverb, whereas the one-word everyday is only an adjective: 

    It’s part of my everyday routine.
    It’s part of my routine every day.
    That’s not an everyday occurrence.
    That doesn’t occur every day.

  • The take-home message: Because everyday is only an adjective, it can only appear before a noun, modifying that noun. If you want to put one or the other at the end of a clause or sentence, saying when or how often something happens, the two-word adverb every day is needed.

  • some time(s) vs. sometime(s)
    This distinction is very similar to any time vs. anytime. First, let’s consider the versions without the final s

    Some time is an adjective–noun pair that means some amount or length of time:

    This is going to take some time.
    Let’s take some time tomorrow and talk about it.
    We’ll have to make up some time by resting less.

    The one-word adverb sometime means at some point in time or at a yet unspecified time:

    This has to get done sometime.
    Let’s talk about it sometime tomorrow.
    We have to stop and rest sometime.

    There is also another way to use sometime as an adjective, before nouns. In this sense sometime means:

    1. having been formerly, former: The diplomat was a sometime professor of history at Oxford.
    2. being so only at times or to some extent: Traveling so much, he could never be more than a sometime husband.
    3. that cannot be depended upon regarding affections or loyalties: He was well rid of his sometime girlfriend.
    [examples shamelessly stolen from

    Now let’s consider the versions with the final s. The two-word adjective–noun phrase some times is simply the plural version of some time:

    Most times work for me, but some times are impossible.
    Most runners performed as expected, but some times seem dubious. [This one’s a little contrived, but you get the point.]

    The one word adverb sometimes means occasionally, at times:

    I only like drinking sometimes.
    I sometimes bike to work.
    Sometimes grammar can be fun!

    In summary, the two-word phrase some time(s) is a normal, run-of-the-mill adjective–noun pair, whereas the one-word adverbs sometime and sometimes tell when or how often things happen, with sometime also being available as an adjective for fancy-pants writers who probably don’t need this grammar guide.

9. How to use the verbs lay, lie, laid, and lied

The first lesson to learn about the verbs lay and lie is that lay is a transitive verb and lie is an intransitive verb. A transitive verb must take an object, meaning the verb must be performed by an actor upon some other noun:

Lay down your book.
When are you going to lay that issue to rest?
As I lay myself down to sleep.

An intransitive verb does not take an object, meaning the subject just performs the verb without acting upon any other noun:

The book is lying on the table.
I’m lying in my bed.
As I lie down to sleep.

In the third example of each verb above, the presence or absence of an object (myself, my body, me) determines what verb can be used, even though they mean basically the same thing. If you said, “I’m laying on my bed,” someone might properly ask, “Laying what on your bed?”

The only reason any of this gets difficult is because English is stupid and the past tense of lie is lay. Well, the past tense of (to tell a) lie is lied, but the past tense of lie (down on something) is lay (down on something).

Last night as I lay in my bed
As soon as the dog lay down on the floor
The treasure lay undiscovered for centuries.

The past tense of lay is laid, so that shouldn’t be hard for anyone to figure out how to use.

Note, finally, the past participle of lie is lain:

Having lain down to sleep
After the dog had lain down on the floor
The treasure has lain undisturbed for centuries.

In summary, yes, you’d be right to label English as a stupid, overly difficult language for making lay the past tense of lie and to complain about this not becoming regularized (i.e., the past tense of lie always being lied regardless of meaning) sometime soon.

10. Who vs. whom

Who and whom are both pronouns, but who is a subject and whom is an object. The best way to determine whether to use who or whom is to either ask or answer who is meant by either pronoun: if you would substitute the who or whom with a subject pronoun like I, he, she, we, or they, then who is needed; if you would substitute the who or whom with an object pronoun like me, her, him, us, or them, then whom is needed.

The patients, most of whom were examined by MRI [You could say “most of them were examined”]
The patients who were examined by MRI [You could say “they were examined by MRI”]
He’s started dating a girl whom he met in class. [Whom did he meet? He met her.]
Who is that girl he’s dating? [Answer the question: “She is that girl he’s dating.”]

The decision between who/whoever and whom/whomever can get tricky when the pronoun is the subject of a verb (i.e., the who/whoever is doing something in the sentence in addition to being the object of an action). In those cases, who/whoever is required, although very slight wording changes would make whom/whomever correct.

I couldn’t decide whom to vote for. [Whom would you vote for? You would vote for him or her.]
I couldn’t decide who was the best candidate. [Who was the best candidate? He or she was.]
You can go with whomever you choose. [Whom would you choose? You would choose him or her.]
You can go with whoever you think is best. [Who do you think is best? You think he or she is best.]
Ask whoever is in charge. [Who is in charge? He or she is.]
You can just ask whomever you want. [Whom do you want to ask? You want to ask him or her.]

In these last three who examples, even though the who clauses are serving the function of objects, you cannot separate the who from its verb; the who must remain in the nominative case, meaning the subject form, not the object form, because the who is performing an action. Subjects perform actions, not objects. Rewording those sentences by removing the who’s verb will convert it back into a normal object and into the whom version.

It should also be noted that basically no grammarians require clause-initial whom, so a lot of sentences in which the rule requires whom would not be considered wrong if who were used, e.g., “Who is he dating?”, “Who would you vote for?”, and “Who would you choose?”

11. Is it okay to split infinitives?

Of course it is. It has never been wrong to split infinitives in English. This was a pseudo-rule that has fixed itself as a misconception in the minds of many young schoolchildren and bad writers.

A split infinitive is the infinitive form of a verb (to + verb) with some word in between the to and the verb. Because it is a waste of my time to even explain any more about this or give a brief history lesson about why some overzealous teachers of grammar ever thought it was wrong to split infinitives, I’ll limit this section to shamelessly borrowing other people’s examples of why it can be necessary to split infinitives occasionally:

The greatest difficulty about assessing the economic achievements of the Soviet Union is that its spokesmen try absurdly to exaggerate them; in consequence the visitor may tend badly to underrate them. (Fowler, 2nd ed., 1893)

We expect our output to more than double in a year.

She decided to gradually get rid of the teddy bears she had collected. [If you think “get rid gradually” or “rid herself gradually” sounds better than the wording of this example sentence, then you have far worse problems than following irrelevant, useless pseudo-rules of grammar.]

Berkeley, Wordsworth, Shelley are representative of the intuitive refusal seriously to accept the abstract materialism of science. [I mean, honestly, come on.]

My first three books…mark the beginning of this development, for it was only as I was writing them that I began systematically to explore childhood

If no appropriate inputs are presented during this period in their lives, songbirds lose the ability fully to supplement what was not acquired, and their later behaviour is significantly affected.

He was never ashamed publicly to bear witness

To re-emphasize: It was never wrong to split infinitives, and it is often wrong and counterproductive to follow the pseudo-rule against it.

12. Can since be used to mean because/inasmuch as/seeing that?

Yes, but be sure no confusion could result from this, because since also has a temporal meaning. The use of causative since is best and most common at the beginning of a sentence, but it can be equally correct and, at the same time, equally tricky in the middle of a sentence, again due to its primary meaning of “in the time that has passed”.

Here are some examples of sentences with since in perfectly fine causative roles:

Since the knockout mice have no embryonic defects, that gene is not necessary for development.
However, since they function as adjectives, participles modify nouns or pronouns. [Copied from the Purdue OWL, a highly esteemed grammar guide.]
Since the store was really crowded, I tried to get in and out as fast as possible.
We might as well walk home since it’s nice outside.
He couldn’t understand why she broke up with him, since it was going so great the week before.

There are, however, many instances in which since can be ambiguous between its temporal and causative meanings (which is no doubt why some strict grammarians insist on its never being used causatively). In these cases, change since to because or re-word the sentence so that since becomes a preposition instead of a conjunction.

Since the CEO died, the company has been making excellent profits. [Is it because the CEO died, or just since that time? Here, Since is being used as a conjunction (before a clause), so if you mean its temporal sense, change the clause “the CEO died” to a noun phrase: “Since the CEO’s death”. If you mean Because, then write “Because the CEO died” or “Because the old CEO is gone” or “Due to the CEO’s death”.]
Since we have been investigating the effects of this drug, we have received a lot of attention from notable researchers in our field. [Does this mean they’ve received attention ever since they began investigating it, or they’ve received attention because they have been investigating it (which doesn’t imply anything about when the attention began—it could have been only recently). To express the temporal sense, some re-wording is necessary: “(Ever) Since we began investigating”. To express the causative sense, just begin with Because.]

Finally, ambiguity aside, since is often better replaced with because in the middle of sentences because the latter is more clear and direct:

I did poorly on the test since because I hadn’t studied enough.
I couldn’t call her back since because I didn’t get her number.
We decided to investigate the mechanism in cultured primary cells since because these have important differences with the immortalized cell line.

When you’re inclined to write Since at the beginning of a sentence, make sure no ambiguity could result from its dual meanings. And when you’ve written since in the middle of a sentence, especially preceded by a comma, see if because with no comma wouldn’t work better.

Garner’s “superstitions”
Grammar Girl
The Grammar Logs

13. Can And and But be used to start sentences in formal writing?
Every grammar and usage guide I can find says both are fine to start sentences in the most formal academic and legal writing. This article by general prescriptivist Bryan A. Garner is a fine summary of the issue. The (prescriptive) grammarian whom Garner seems to idolize the most is Englishman H.W. Fowler, who called the proscription against sentence-initial And and But a “fetish” and an “ungrammatical piece of nonsense”.

In that article, Garner says he and other usage guide writers prefer to change However to But at the beginning of a sentence and that however only sounds good between commas after the first clause. I definitely disagree with that, and I will note that the flow of the paragraph often makes However (always followed by a comma) sound better than But at the start of some sentences. And I strongly dislike however surrounded by commas in the middle of a sentence because it is more often confusing and awkward than simply starting the sentence with However. This is because sentence-initial However introduces the contrast immediately and leaves the rest of the sentence—the actual contrasting statement—uninterrupted, whereas mid-sentence …, however,… not only waits to introduce the fact that there’s a contrast but interrupts the train of thought conveyed by the things that come before and after it.

Finally, the only thing about sentence-initial And and But that every guide proscribes is putting a comma after the word as if it were an adverb or interjection. Only put a comma after it in rare cases when some other clarifying or introductory phrase immediately follows it.

So don’t listen to any pedantic fetishist who wants to outlaw And and But at the beginning of sentences in any context or in any type of document. But don’t overdo them, either. In formal writing, one every few hundred words is plenty.

Garner’s “superstitions”
Arnold Zwicky and people he cites

14. State-of-being verbs take adjectives, not adverbs
When you use a verb to describe a noun’s state of being, to describe a characteristic of that noun, the verb must be followed by an adjective, not an adverb, because you’re really modifying the noun, not the verb. The most common, though possibly most overlooked, state-of-being verb is (obviously) to be, but they also include seem, look, appear, smell, feel, taste, sound, etc.

It’s pretty annoying when someone says they “feel badly” or something “smells badly”, because those verb–adverb pairs mean something different from their intent. For some reason, this mistake seems to be made most often with the adverb badly.

I feel bad. This means you feel sorrow, guilt, regret, or some other negative emotion. Bad is modifying I, not feel.
I feel badly. This means your sense of touch isn’t very strong or discriminating, and using your hands to feel for something isn’t effective. Badly is modifying feel, not I.
That dog smells bad. This means the dog has a bad odor. Bad is modifying dog, not smells.
That dog smells badly. This means the dog’s sense of smell is poor, so he isn’t good at smelling things. Badly is modifying smells, not dog.
This food tastes good (not well).
He seems honest (not honestly).
The building looks weak (not weakly).
She appears hopeful (not hopefully).

15. What does beg the question mean?
Question begging or begging the question is a logical fallacy that means circular reasoning; assuming your conclusion in your premises. The Nizkor Project gives some good examples of question begging:

  1. Bill: “God must exist.”
    Jill: “How do you know.”
    Bill: “Because the Bible says so.”
    Jill: “Why should I believe the Bible?”
    Bill: “Because the Bible was written by God.”
  2. “If such actions were not illegal, then they would not be prohibited by the law.”
  3. “The belief in God is universal. After all, everyone believes in God.”
  4. Interviewer: “Your resume looks impressive but I need another reference.”
    Bill: “Jill can give me a good reference.”
    Interviewer: “Good. But how do I know that Jill is trustworthy?”
    Bill: “Certainly. I can vouch for her.”

If you mean raise the question, urge the question, or prompt the question, then say one of those. They take the exact same amount of time and syllables and actually convey what you mean.

16. Between is not limited to relationships between two nouns
The word between has the same Old English roots that twain and two have, hence the mistaken assertion that between can only be used to relate two nouns or ideas. True, when you are referring to two things, among is wrong, but that doesn’t mean between can only be used for two nouns. Rather, the difference between their meanings is more subtle and interesting than that.

As noted in Garner’s Modern American Usage, the Oxford English Dictionary best clarified the true distinction between between and among: “In all senses between has been, from its earliest appearance, extended to more than two…. It is still the only word available to express the relation of a thing to many surrounding things severally and individually; among expresses a relation to them collectively and vaguely.” Garner later states it as: “between expresses one-to-one relations of many things, and among expresses collective and undefined relations.” In many uses, among expresses a relationship similar to amid.

“I’ve come into your plans…and between the three of you, as well—but I couldn’t help it.” (Stephen King, Wizard and Glass)
the relationship between children and their parents
He walked between the trees vs. He walked among the trees. [Different meanings.]
the space lying between the three points
correlations between disease stage and age, gender, comorbidities, and medications
My aunts and uncles have 10 children among them.
The three thieves divided the loot evenly among themselves.
a god among men

Garner quotes an interesting example, “the appropriate balance between family, solitude and community,” but I’m not sure I agree with that one. I’m not going to tell Garner or most other accomplished writers, editors, or grammarians they’re wrong, but I don’t see why “the appropriate balance among family, solitude and community” would be sub-optimal in that sentence. Between usually implies a unidirectional or bidirectional relationship between two things, between one thing and several other things individually, or between individual pairs of a group of things. I don’t see the relationship of family, solitude, and community as any type of individual or pair-wise relationship, nor as any series of pair-wise relationships; I see it as a balance of all three things simultaneously, which is exactly what the word among is for. If three thieves divide loot among themselves, then I think a balance would be struck among family, solitude, and community.

Garner’s Modern American Usage
The real difference between between and among
PFS Publishing
Grammar Girl

17. Farther vs. further
The best summary I’ve seen is that farther means “more distant” whereas further means just “more/to a greater extent”. For this reason, when referring to a metaphorical distance or separation, further and farther are interchangeable.

He lives farther away than I thought.
That was the farthest home run he’s ever hit. [Longest would be synonymous]
I just ran farther than I ever have without stopping to rest. [Longest would be synonymous but unwise because long also has a temporal meaning]
How much farther is it until the next rest stop?
Due to time constraints, I won’t speak further on this matter.
Just as the rain abated, lightning caused further delay.
Further studies are warranted.
before we go any further/farther
Some students are falling further/farther behind.

Further can also be a verb, but farther cannot:

to further a cause
to further your own interests

Grammar Girl
Merriam-Webster Dictionary
Garner’s Modern American Usage

18. Disinterested vs. uninterested
This must be the cornerstone issue in the descriptivist vs. prescriptivist debate. Descriptivists will point out that disinterested used to mean both “not intrigued, bored, indifferent” and “impartial, unbiased, having no stake in the matter” and that uninterested actually originally meant “unbiased” and only acquired its “bored, indifferent” sense in the 1700’s.

Well, to me, Grammar Girl, and Bryan Garner it seems imprecise and sloppy to have the words overlap so much and basically be interchangeable for the “bored, indifferent” meaning when we have the option of using them distinctively. Why use disinterested to mean “uninterested” when we already have a word for this purpose that looks and sounds remarkably like uninterested? It can’t be denied that 17th- and 18th-century purists would have been equally right (or wrong) to make the exact opposite argument. And I would have been on their side then as well. I’m on the “traditionalist”, “purist”, or “prescriptivist” side of this issue now because I (along with many others) see the words as meaning different things and I see the benefits of preserving their distinction. It seems a beneficial prescription for more precise usage to use different words to mean different things when you have that option.

Here are some examples of their distinct usage:

In any criminal case, a disinterested judge is highly desirable.
For a more accurate evaluation, we need a disinterested third party to arbitrate.
The students seem uninterested in the subject matter.
She was uninterested in hearing my excuses.

Descriptivists also note that the use of either of these words rarely creates ambiguity in context, so I guess we’re left pointing out that the benefits of distinction come more in the realm of logic and purity than clarity and meaning.

Finally, check out the entry on disinterest vs. uninterest, above disinterested and uninterested, in Garner’s Modern American Usage. It didn’t occur to me that there would be a need for different entries on the two forms of the words, but I think it’s interesting and useful.

19. Like vs. as

In the contexts where like and as can be confused, like is a preposition and as is a conjunction. Therefore, like should come before a noun phrase, and as should come before a clause (i.e., containing a noun and a verb).

As I said [not *Like I said]
As I was saying [not *Like I was saying]
Appearing and disappearing as only Batman can
I studied as I had never studied before. [Note that “as if/as though” would change the meaning, and “like” would be ambiguous because it could imply either of these meanings!]
He’s sleeping like a dog.
No one can play the guitar as Jimi Hendrix could.
No one can play the guitar like Jimi Hendrix.
Albert Pujols got off to a slow start, just as he does most years.
Albert Pujols got off to a slow start, just like previous years.
The knockout mice did not respond as the wild-type mice did.
The knockout mice did not respond like the wild-type mice.
As most people do, I distrust politicians.
Like most people, I distrust politicians.

There are many instances where like needs to be replaced with as if or as though because a clause follows, but I recommend limiting such nit-pickiness to formal writing:

He looks as if/as though he’s really struggling
It looked as if/as though it was going to rain
He acts as if/as though the world were going to end.

Grammar Girl

20. Parallelism with the word both or either

I added this to my grammar guide because a lot of non-native English speakers seem to have trouble with this one. They often put a preposition after both or either when in fact it belongs before. (This actually is a black-and-white issue of grammar, not a matter of style or word choice or language change, in which there is definitely a right and a wrong word order.)

In English, it is important to keep lists parallel. Parallelism means, basically, the items of a list are grammatically interchangeable, meaning you could switch their order and retain a grammatically correct sentence. For example, the following sentence is ungrammatical:

*This was observed both in the control group and the treatment group.

The items that come after the word both must be parallel. This means you could switch the order of the individual items and the sentence would mean the same thing and be grammatically correct. How do you identify these “individual items”? First, they start after the word both or either, i.e., the both or either is not part of the list and is not part of the parallel elements. Second, these individual elements are the phrases separated by commas and conjunctions. In the example above, there are no commas and only one conjunction: and. The first item that comes after both is in the control group, and the second item is the treatment group. Notice that one of these phrases has a preposition, in, and the other does not. That means the two items of this list are not parallel, because observe what happens if you switch their order:

*This was observed both the treatment group and in the control group.

That sentence is not possible in English. It is ungrammatical. To make it an English sentence, we have to switch both and in so that what comes after both is a parallel list (with perfectly interchangeable items):

This was observed in both the control group and the treatment group.
This was observed in both the treatment group and the control group.

In sentences with both and a single preposition, the both is what precedes the start of parallelism. If you want a single preposition to apply to all of the items of the parallel list, the preposition must come before both. Alternatively, you could use the same preposition twice if you wanted, in which case it must come after both to retain parallelism:

This was observed both in the control group and in the treatment group.
This was observed both in the treatment group and in the control group.

Similarly, if you want to use different prepositions for each item of the list, then you cannot put one of those prepositions before both:

He likes to listen to audiobooks both at home and on the road.
He likes to listen to audiobooks both on the road and at home.
This was observed both before and during treatment.
This was observed both during and before treatment.

The exact same is true of either, which takes the conjunction or:

No side effects were observed in either the control group or the treatment group.
No side effects were observed in either the treatment group or the control group.
We analyzed the tissue by either CT or MRI.
We analyzed the tissue by either MRI or CT.

21. If X had happened vs. *If X would have happened
I put an asterisk before “If X would have happened” because it is ungrammatical. I hear it and see it all the time, which doesn’t make it grammatical or acceptable. The speaker’s meaning is still understandable, but it sounds uneducated because that’s not how English verb tenses work.

To express the possibility of something having happened in the past that didn’t actually happen (i.e., a “counterfactual”), English uses the past perfect tense, which requires the auxiliary verb had. Not has or have; only had:

If the Patriots had won, they would have reached their sixth Super Bowl in 12 seasons.
If I had remembered to bring my book, I would have been less bored on the plane.
If I had known it was so important, I would have come earlier.
If he hadn’t made that left turn at Albuquerque, he wouldn’t have gotten lost.
If she hadn’t trained so hard, she would have lost the race.

In summary: it is ungrammatical to present a counterfactual possibility with *If [noun] would have…. The conditional construction would have can only be used to present the “result” or “outcome” of a counterfactual that is presented elsewhere in the sentence (or, I guess a previous sentence). The If [noun] had and If [noun] hadn’t constructions are used to present the hypothetical scenario that something happened in the past that did not in fact happen.

22. Into, in to, onto, and on to in phrasal verbs
Some phrasal verbs that consist of a verb + in or a verb + on cannot be changed to verb + into or verb + onto. I have no rule or even guideline for when the latter is inappropriate except that sometimes it just is (why is any preposition right but another wrong?), but I see it somewhat frequently. Here are a few examples to help you avoid that mistake:

You log in (or log on) to a computer, not *log into (or *log onto).
When you’re done with the paperwork, hand it in to me (not *hand it into me).
Turn your project in to the professor as soon as it’s finished (not *Turn it into the professor).
I’ve been slow to catch on to all the nuances (not *catch onto).
Let’s move on to the next topic (not *move onto).
She was the one who turned me on to classical music (not *turned me onto).

Sometimes using into instead of in can be correct, for instance, with the verb zoom. Zoom can be used in several ways in 21st-century English, mainly: (1) to move quickly or suddenly with a humming or buzzing sound; and (2) with in or out, to increase or decrease the viewing magnification of a document, subject, or scene, simulating getting closer to or farther from it, respectively. In this Sports Illustrated column by Andy Staples, he uses the single-word verb zoom correctly and adds the preposition into to state where Chip Kelly would be zooming:

Kelly can’t zoom into Philadelphia, promise to Win The Day and merrily roll along.

whereas this use of zoom into would be incorrect:

If you *zoom into that part of the picture, we can analyze it better.

23. Eager vs. anxious
For more effective, precise writing, use eager when you mean “looking forward to something with enthusiasm or desire” and anxious when you mean “nervous, worried, or distressed about a potentially negative upcoming event or experience”. This distinction is easy to remember because anxiety means a state of uneasiness or apprehension, such as about future uncertainties. Each pair of sentences below mean different things:

I’m eager to meet this girl Brian wants to introduce me to.
I’m anxious to meet this girl Brian wants to introduce me to.
I’m eager to see what I got on the test.
I’m anxious to see what I got on the test.
I’m eager to meet my students on my first day at my new school.
I’m anxious to meet my students on my first day at my new school.
I’m eager to see my family this Thanksgiving.
I’m anxious to see my family this Thanksgiving.

The reason this is a confusing usage issue is that people often use anxious when they mean eager in casual conversation. To be honest, in contrast to my invented examples above, when anxious is used in its “distressed, worried” sense, it typically follows a different structure than eager, at least in everyday speech: We often say we are anxious about something, whereas we are eager for something or eager to do something. On the other hand, when people unwisely use anxious to mean eager, they often phrase it in the same way they would if they had used eager, with for or an infinitive verb after it. So if we were going to rephrase my anxious examples above according to common usage in speech, but still use the word correctly, they might look like this:

I’m anxious about meeting this girl Brian wants to introduce me to.
I’m anxious about getting this test back.
I’m anxious about meeting my students on my first day at my new school.
I’m anxious about seeing my family this Thanksgiving.

But what about sentences with different structures from these? In those cases, there is no preposition after anxious to provide clues as to which meaning was intended. You have to convey your meaning by, y’know, using the right word:

He boarded the plane eagerly.
She felt anxious as she turned into the driveway.
My anxiety grew as my first day of work approached.
His eagerness had faded after all that had transpired since the announcement.

What’s worrisome is that this poor usage has spilled over into writing all too often, in publications that I would have thought had higher standards than that. For instance, Michael Gerson writes in the Washington Post, “The end of childhood, of course, can be the start of adult relationships between parents and children that are rewarding in their own way. I’m anxious to befriend my grown sons.” And Robert Talbert writes in his blog at the Chronicle of Higher Education, “Second, I’m teaching my first upper-level course since arriving at GVSU, one section of our Modern Algebra course, which I have not taught in a few years and I am anxious to get into it.” We can tell from context that Gerson definitely meant eager, and I’m about 95% sure Talbert meant eager.

Why didn’t they use eager, then? Good question. It’s an unfortunate point of history that anxious has come to mean two completely opposite things to some English speakers, and it’s even more unfortunate that so many smug ultra-descriptivists practically trip over themselves in their rush to proclaim how non-judgmental they are. Well, they should be judgmental. They should actually criticize sloppy, ineffective usage once in a while. They wouldn’t recognize ineffective usage if it jumped off the page and smacked them in the face.

As a final note, I see that both and the American Heritage Dictionary even give as a minor definition of anxiety “eager but tense desire” and “eager, often agitated desire”, respectively. That surprised me because I am 99.99% sure that I have never, ever heard or seen anxiety used in that way. I’m extremely skeptical that many American English speakers would consider that a standard use of anxiety. I wonder if the dictionaries felt compelled to add that as a minor definition because of so many people’s misuse of anxious; I can imagine them thinking, “Uh-oh, sticklers are pointing out that anxious shouldn’t mean eager because anxiety doesn’t mean eagerness; well, we’ll just fix that by adding a new definition for anxiety.” I just can’t accept that that definition is standard without a lot more personal experience to back it up.