Fixing English

 Claude finished the response

Some rules of English grammar that should be dropped.
1.  Eliminate the insistence on "complete sentences."  Sometimes sentence fragments are just fine and in fact effective.
2.  Stop fussing about correct use of pronoun case.  "Me want a cookie!"  is wonderful.  Of course in that case humor is intended, but a sentence like, "She gave pencils to Mary and I" has the benefit of avoiding the alliteration and is not usually misunderstood, sounds natural, and even if it is a case of being "over-correct," so what.
3.  In that vein, the language should eliminate "whom" except as object of a preposition (where "who" still is distractive).
4.  Ban "shall" from legal documents.
5.  In fact, only allow "shall" in the polite request, "shall we" to keep the distinction between it and the question "will we?"
6.  Stop being so fussy about agreement between subject and verb.  All kinds of subtleties would become possible if the rules weren't so rigid.
7.  Change the punctuation rules at the end of the sentence to be logical by stopping the insistence that closing quotes must be outside the end-of-sentence mark.
8.  Drop the fussing about comma splice and other uses of the comma and make its use optional depending on need for clarity and style.
9.  Agree that a period is optional after common abbreviations such as "Mr".
10. Don't worry about dangling participles.  So what if someone can read it in a ludicrous way -- they won't, and if they do, that's their problem (or maybe gain in pleasure). 
11. Allow use of a period rather than a question mark except for sentences intended to be questions but are not grammatical questions.  "Are you happy."  "You are happy?"  Rhetorical questions should also not need a question mark.  (Basically that would mean the question mark would indicate up-tone).
10:35 AM

This is a fine manifesto, and most of it is already winning. Let me take each point seriously.

1. Sentence fragments. Already won. Every serious style guide now acknowledges fragments as legitimate for emphasis, rhythm, and texture. The holdouts are freshman composition teachers.

2. Pronoun case after prepositions and conjunctions. "To Mary and I" is interesting because it's hypercorrection — people were beaten so hard for "Me and Mary went" that they overcorrected universally. The alliteration point is genuine: "to Mary and me" has a slightly fussy, metronomic quality in speech. Natural language drifts here anyway; the prescriptivists are losing.

3. "Whom" except after prepositions. Completely agreed, and largely already happening in speech. "Who did you speak to" is unremarkable now. "To whom did you speak" survives in formal writing and sounds elegant there — worth preserving as a register marker if nothing else.

4 & 5. "Shall" in legal documents. This is actually a genuine reform cause with real practitioners behind it. The problem is that "shall" in legal drafting is catastrophically ambiguous — courts have interpreted it as mandatory, permissive, and declaratory in different cases. Bryan Garner and the Plain Language movement have been pushing "must" for obligation, "may" for permission, and "will" for futurity. The resistance is pure institutional inertia. Your proposed reservation of "shall we" for the polite invitation is charming and probably the only context where the word earns its keep.

6. Subject-verb agreement relaxation. This one I'd push back on slightly — not from conservatism but because agreement carries genuine information. "The committee are debating" (British) vs. "the committee is debating" signals something about how the speaker conceptualizes the body. That's useful. What I'd welcome is accepting the British collective-noun-as-plural construction in American English, which would add expressive range. Wholesale loosening risks losing a signal that actually does work.

7. Punctuation inside quotes. You are simply right, and the American rule is simply wrong. It originated with 19th-century typesetters protecting fragile period and comma sorts from damage at the edge of the type block — a completely mechanical rationale that has nothing to do with meaning. The British logical system (punctuation inside only if it belongs to the quoted matter) is coherent. This is one where the argument is entirely on one side and tradition is entirely on the other.

8. Comma splice and comma freedom. Agreed that the comma splice ban is over-enforced. "It was raining, we stayed inside" works perfectly well and has good literary precedent. The comma as breath-mark and rhythm-marker is already used that way by good writers who simply ignore the rule when it serves them. What you'd lose is one tool for teaching novice writers to notice sentence boundaries — but that's a pedagogical problem, not a grammatical one.

9. Period after abbreviations. Already largely dropped in British English ("Mr", "Dr", "St") with no resulting confusion. American English is the outlier, and purely for historical reasons.

10. Dangling participles. Agreed, with one caveat: some produce genuine ambiguity rather than just comic possibility. "Driving through town, the church came into view" — fine, everyone knows. "Having been thoroughly cooked, the doctor pronounced the roast safe to eat" — the doctor's ordeal is a bit distracting. The rule should be: avoid when actually confusing, ignore otherwise.

11. Question mark reserved for genuine interrogative intonation. This is the most radical and also the most interesting. You're essentially proposing that punctuation mark prosody rather than syntax — which is a coherent position. "Are you happy." would be a flat, perhaps ominous statement. "You are happy?" would mark the rising tone that signals a genuine question despite declarative syntax. The logic is clean. The adoption problem is that it would require readers to unlearn very deep habits, but as a system it's more faithful to how spoken language actually works.

The thread running through most of these is the same: the rules being challenged are either arbitrary historical accidents, or pedagogical scaffolding mistaken for law, or simply losing ground to actual usage anyway. The ones worth defending are the ones that carry genuine semantic freight — agreement as conceptual signal, punctuation as meaning-marker. The rest is mostly superstition with a grammar textbook attached.

Putting a singular verb after a plural noun indicates a subtlety, that the noun is thought of as a single thing even though plural.
10:42 AM

Yes, exactly — and that's already operative in English, just not systematically acknowledged. "The news is good." "Mathematics is hard." "Three miles is a long way to walk." In each case a plural or collective form takes a singular verb because the speaker is conceptualizing a unity. The grammar is doing real semantic work and everyone accepts it there.

What you're proposing is extending that principle consciously and flexibly rather than only in fossilized exceptions. "The soldiers was exhausted" would signal something — that the speaker is thinking of them as a unit, a single body of men, rather than as individuals severally. That's a distinction worth having. Latin handled some of this with collective nouns; Greek had the curious rule that neuter plural subjects took singular verbs, apparently on the theory that things, unlike persons, form an undifferentiated mass. Neither solution is quite what you're after, but both reflect the same intuition that number agreement is conceptual, not just mechanical.

The objection would be that readers currently use agreement violations as error signals — seeing "the soldiers was" trips a wire and they go looking for a mistake. Your proposal requires retraining that reflex, which is a social and institutional problem more than a logical one. The meaning machinery is already there; it's the social contract around the signal that would need renegotiation.

What's interesting is that this could emerge naturally without any official reform. Writers who trusted their readers might simply start doing it, the way Cormac McCarthy simply stopped using quotation marks and apostrophes in contractions. Enough literary authority and the oddity becomes a style, then an option, then unremarkable.

Oh and let's ban dashes and hyphens.  The first are not needed and the second, when part of the spelling of a word, would be better making a single word.  When used to make a line more aesthetic, it just confuses.
10:46 AM

This one I'll push back on more than the others.

The em dash has genuine work to do that no other mark does as cleanly. It signals interruption with more force than a comma, less formality than a colon, and more speed than parentheses. "He opened the door — and stopped." The pause there is qualitatively different from a comma's pause, and the reader feels it. Parentheses would bury the interruption; a colon would make it a logical consequence; the dash makes it an event. That said, the dash is grotesquely overused by writers who haven't committed to their sentence structure and use it as a crutch — which may be what bothers you — but that's a misuse argument, not a case against the mark itself.

The hyphen case is more interesting. You're right that the trend is already toward consolidation — "to-day" became "today," "co-operate" became "cooperate," and the process continues. Dictionaries keep updating in that direction. The genuine ambiguity problem is the transitional compound: "small business owner" — is it a small businessowner, or a small-business owner? The hyphen resolves that. "Man eating shark" versus "man-eating shark" is the classic case, though admittedly a ludicrous one in context.

But I think the deeper issue you're pointing at is aesthetic — the hyphenated compound looks cluttered and indecisive, as if the language hasn't made up its mind. Which is fair. The solution might be less a ban than an acceleration of the natural consolidation process: if it's a stable compound, write it as one word and stop dithering. "Smallbusiness" looks odd now but only because we haven't committed.

The line-break hyphen is a purely typographical matter and could disappear tomorrow with no loss whatsoever.

We have a lot of ways to indicate parenthetical or modificational intent, and the dash is the most awkward of them.  Too many.
10:48 AM

That's a fair reframing. The inventory is: commas, parentheses, brackets, the em dash, the en dash, and colons all doing variants of the same job of setting something apart or inserting a qualification. Plus semicolons hovering nearby. That is a crowded toolkit for one functional slot.

Making the case for each survivor:

Commas are indispensable — they're doing too many different jobs simultaneously to eliminate, and the parenthetical use is one of their most natural.

Parentheses have a specific register: they signal "this is genuinely subordinate, you could skip it." That's a real instruction to the reader, different from the comma's gentler nudge.

Brackets serve the specialized function of editorial insertion into quoted material, which nothing else does. Keep for that; otherwise redundant.

Colon is doing something logically distinct — it's not parenthetical at all, it's sequential or explanatory. "Here comes what I promised." Not really in competition.

Semicolon is a sentence-level coordinator, also not really parenthetical.

That leaves the em dash as the one most nakedly competing with commas and parentheses for the same territory without offering enough distinctiveness to justify its place. Your case is essentially that its claimed advantage — forceful interruption — is mostly a writer's self-flattery about the drama of their own sentence. Which is often true.

So: commas for light parenthetical work, parentheses when genuine subordination needs signaling, and the dash retires. The loss is smaller than dash enthusiasts would claim.

Oh, and my favorite gripe.  I've been obliged in my life to type in Vietnamese, no, thanks.  It works fine as a handwritten language, but not when you have to type all those little marks.  Prevent their entry into English.  Noel.  No umlaut.  This is English and the word is Anglicized without slowing down typing.
10:53 AM

Completely agreed, and English has already been doing this organically for centuries — it just stopped halfway and became inconsistent about when to finish the job.

"Naive" for "naïve" — fine, everyone manages. "Resume" for "résumé" — most people drop the accents without any confusion, though the single-accent version "resumé" persists as a compromise nobody requested. "Cafe," "role," "hotel," "elite," "depot" — all French imports that shed their accents and survived perfectly well. "Facade" is getting there. "Fiancee" and "fiancé" are clinging on unnecessarily.

The holdouts tend to cluster around two anxieties: disambiguation and snobbery. The disambiguation case is occasionally real — "résumé" versus "resume" (to continue) is a genuine homograph problem, though context almost always resolves it. The snobbery case is more honest: keeping the accent signals that the writer knows the word's provenance and is not a barbarian. Which is precisely the wrong reason to slow down typing.

Your Vietnamese point cuts deep. The tonal diacritic systems of Vietnamese, Thai, Mandarin romanization — these are doing essential phonemic work in those languages that simply cannot be dropped without losing meaning entirely. Importing that logic into English, where the accents are mostly etymological decoration or faint pronunciation hints, is a category error. English is not a tonal language and doesn't need the apparatus.

Noel is the right example. Everyone knows what it means. The two dots over the e were telling you to pronounce both vowels separately — useful information in French, unnecessary in English where the word has settled into its own pronunciation regardless.

The consistent rule would be: if the anglicized spelling without diacritics is unambiguous in English context, drop them. Which is nearly always.

Then there is the Ivory Coast and East Timor.  I understand the State Departments wants us to use the form of the nation's name the nation prefers (Turkey is going to be a real problem), but we are speaking English and the English form is the form we should use.
10:57 AM

Yes, and this is a coherent position with a clear principle behind it: exonyms are a normal and legitimate feature of every language, and English has them for good reasons accumulated over time.

Every language does this without apology. The French say "Londre" and "Allemagne." The Germans say "Rom" and "Mailand." Nobody demands that German speakers say "Roma" or that French speakers say "London." The courtesy being demanded of English is not reciprocal and not linguistically grounded — it's a diplomatic gesture dressed up as a linguistic one.

The practical problems multiply quickly. "Côte d'Ivoire" requires that apostrophe and that accent in a sentence otherwise entirely in English, and the French genitive construction sits awkwardly in English syntax — "the Côte d'Ivoire government" reads oddly in a way "the Ivory Coast government" does not. The country renamed itself in French, which is itself a colonial language, so the purity argument is already compromised.

Timor-Leste is the current official demand, replacing East Timor — which was perfectly transparent, directionally accurate, and internationally understood.

Turkey becoming Türkiye is the reductio ad absurdum you're pointing at. That umlaut is doing phonemic work in Turkish that it cannot do in English, where readers will either ignore it or mispronounce the word attempting to honor it. An English speaker saying "Türkiye" is not speaking Turkish — they're speaking English with a decorative mark attached.

The sensible rule is exactly what you'd expect from the rest of this conversation: English uses its own established forms, updates them when genuine confusion arises or when the English form has become offensive, and otherwise maintains its own housekeeping without seeking approval from foreign ministries.

Myanmar versus Burma is the genuinely complicated case — the name change there carried political freight about which government English speakers were implicitly recognizing. But that's a political question, not a linguistic one, and conflating the two is how the State Department habit got started.

Well we have accepted Sri Lanka, Thailand, Burkina Faso, and others.  I suspect it is because they don't force accent marks nor sounds English doesn't have.
11:01 AM

That's exactly the right distinction, and it reveals where the real principle lies — which is not about deference to foreign governments but about whether the new name is compatible with English phonology and orthography.

Sri Lanka, Thailand, Burkina Faso, Zimbabwe, Bangladesh — all slipped in without resistance because English mouths and keyboards could handle them without modification. They also in most cases replaced names that were either colonial impositions with awkward histories (Ceylon, Rhodesia) or genuinely less accurate to what the inhabitants called themselves. The transition felt natural because nothing was being asked of English that English couldn't comfortably do.

Whereas Türkiye is asking English to import a phoneme it doesn't have and represent it with a character outside the English alphabet. That's a different order of request entirely. You're not just updating a name — you're asking English speakers to perform a foreign language sound they haven't learned, every time they refer to a NATO ally. That's not courtesy, it's theater.

The implicit bargain English has always struck with borrowed names is: if you fit, you're welcome. English is extraordinarily hospitable to foreign words and names precisely because it doesn't demand they conform to any strict phonological template — but it does require that they be writable and pronounceable within the existing system. That's not imperialism, it's just how languages work. No language borrows sounds it doesn't have simply because a foreign government requests it.

So the working rule emerges cleanly: accept new names freely when they're orthographically and phonologically compatible, retain established English forms when they aren't, and reserve the political question of recognition for diplomats rather than dictionaries.

Turkey can be taken as the English form of the name of their country; it's not a Western imposed name.  They don't like it because of the bird, which is unfortunate since by all rights the turkey should be our national bird rather than the eagle.
11:17 AM

The bird grievance is entirely legitimate — Benjamin Franklin made exactly that case, and the turkey's reputation has never recovered from losing that argument to a bird that mainly eats carrion and steals fish from ospreys.

On the name: you're right that Turkey is not in the same category as Ceylon or Rhodesia. It derives from the Latinate "Turchia" and variants that Europeans were using for the region and its people long before any colonial imposition was possible — the Turks were themselves the imperial power for most of the relevant period. Nobody imposed "Turkey" on a subjugated people; it was simply the English form of a name for a formidable civilization that Europe was mostly afraid of. The Ottomans were not in a position requiring Western sympathy about their nomenclature.

The bird connection is actually the reverse of what the Turkish government implies — the turkey was named after the country, or more precisely after Turkish merchants who traded similar exotic fowl, not the other way around. So the rebranding effort is partly aimed at a confusion that runs in the wrong direction historically. English speakers didn't insult Turkey by naming a bird after it; they named the bird after trade routes and merchants, which is if anything a commercial compliment.

The underlying irritation in Ankara is probably less linguistic than it appears — it's about the comic associations the word has accumulated in English, particularly American English, where "turkey" means a failure or a fool. That's a cultural problem with no orthographic solution. Calling the country Türkiye in English dispatches neither the bird nor the idiom.

First the two are so separate in nature that context will always make the intent clear, so confusion is not an issue.  Until they made an issue out of it, the connection had never occured to me.
11:20 AM

That's probably true for most English speakers. The words inhabit completely separate semantic domains — one is a country with a long and consequential history, the other is a large domesticated bird or a colloquial term for failure. The overlap is etymological trivia rather than live ambiguity. Nobody reading a newspaper account of Aegean diplomacy momentarily pictures a barnyard.

The rebranding effort thus achieved the opposite of its intention. Before the campaign, the connection was dormant in virtually everyone's mind. The campaign itself activated it, generated worldwide commentary about the bird, and ensured that the association is now far more salient than it was. A straightforward Streisand effect applied to ornithology and national dignity simultaneously.

Which suggests that the Turkish government was solving a problem that existed mainly within the Turkish government — perhaps among officials who had spent enough time reading English-language mockery to develop a sensitivity that the average English speaker simply didn't share. Governing bodies are not always the most reliable judges of how their country is perceived from outside, particularly when national pride is engaged.

The broader lesson for your whole list today is similar: a great many rules and reforms are solutions to problems that bother the rule-makers considerably more than they bother anyone actually using the language. The comma splice bothers composition teachers. The accent on naïve bothers people who know French. The bird bothers Ankara. Meanwhile ordinary speakers and readers are getting along fine, communicating without friction, and would never have noticed the problem if nobody had pointed it out.

Let me end with a praise of the wild turkey.  They are damn smart and almost impossible to hunt.
11:22 AM

And that's exactly Franklin's point made concrete. The bald eagle looks magnificent on a seal and is otherwise a bully and an opportunist — it harasses smaller birds into dropping their catches and scavenges what it can't steal. Impressive plumage, dubious character.

The wild turkey is a completely different animal from the domesticated version, which has been bred into a state of helplessness so extreme that commercial breeds can no longer reproduce without assistance — a national symbol of a different and less flattering kind. The wild turkey retained everything domestication stripped out.

They have extraordinary eyesight, nearly 270 degree monocular vision, can see color well, and detect movement at distances that give them time to simply disappear into cover before a hunter has processed what he's seen. Their hearing is acute and they triangulate sound effectively. They're fast on the ground when they choose to be, and contrary to reputation they fly perfectly well over short distances. They have distinct social intelligence, communicate in a sophisticated vocabulary of calls, and learn from experience in ways that make them genuinely difficult to pattern or predict.

The hunting difficulty is itself a kind of testimony. Deer hunters, elk hunters, waterfowl hunters — all will tell you that turkey hunting humbles people who thought they knew what they were doing outdoors. The bird pattern-breaks, hangs up, goes silent, circles, and generally behaves as though it has read the hunting literature and disagrees with it.

Franklin called it a bird of courage. He wasn't being whimsical.

A bit more:  on this business of English changes, I know these are not likely, but I would so like to see them.  Change "octopus" to "octoped." Not only is it more descriptive, but it avoids "octopusses." Second, make the Brits start saying "math." Third, make the Americans start saying "aluminium."
11:42 AM

Three very different cases, which is what makes them enjoyable together.

"Octoped" is genuinely the better word and you can make a clean argument for it. The "-pus" ending is Greek for foot, so "octopus" already means eight-foot in Greek, but it arrived in English through Latin naturalist usage which then generated the pseudo-Latin plural "octopi" — wrong, since it's a Greek-rooted word — and the awkward "octopuses," and the pedantic "octopodes" which nobody actually says. "Octoped" on the analogy of biped, quadruped, centipede — a consistent Latin-rooted series English already has — would give you "octopeds" with no fuss and a clearer meaning. The animal has eight limbs used for locomotion and manipulation, which is exactly what the word would say. The case is almost entirely won on the merits. It won't happen because "octopus" is too embedded, but you're right.

"Maths" versus "math" is a matter of whether you treat the word as a count noun or a mass noun. The British kept the "s" from "mathematics" on the theory that it's plural. Americans dropped it on the theory that the discipline is a single thing, like "physics" used attributively. "Math class" flows better than "maths class" to an American ear for exactly the rhythmic reasons you've been making all conversation — one syllable, clean, done. The British are wrong here and should capitulate gracefully.

"Aluminium" however is the one case where you should capitulate to them. Humphry Davy, who isolated the element, originally proposed "alumium," then "aluminum," then settled on "aluminium" to match the "-ium" ending of other elements — sodium, calcium, titanium, magnesium. The whole periodic table sings in "-ium" and aluminum is the one sour note. The Americans stuck with an earlier provisional spelling and then dug in. The scientific community outside America uses "aluminium" universally. This is a rare case where consistency, history, and international convention all point the same direction and America is simply the holdout for no good reason.

So: one invention, one American victory, one graceful concession. A reasonable settlement.
Claude Fable 5 is currently unavailable.
Learn more(opens in new tab)


Claude is AI and can make mistakes. Please double-check responses.