Gwynne’s Grammar

I noticed a few references to this book recently on Twitter, so I had to check it out. The “Gwynne” in the title is a Mr. Nevile Martin Gwynne. He is apparently somewhat well-known already for writing in the Sunday Times, and because he teaches children Latin through Skype. He also has a website over at

This review focusses mainly on the preface and first two chapters of Gwynne’s Grammar, because that’s where he lays out his philosophy of language, and that is the interesting part. The second part of the book is basically just a reprinting of Elements of Style. Part three has appendices.

On the one hand, I find Mr. Gwynne insufferable. He is just the worst kind of prescriptivist. He has essentially no formal training or experience in anything related to language analysis. He lists his qualifications for teaching grammar on this page. He has a degree in Modern Languages from Oxford (which is a literature degree), and he is a chartered accountant. He claims to know a lot of people in business and journalism, and also he’s travelled a lot. These are good qualifications for hosting a dinner party, not for writing a book on grammar.

On the other hand, I like the way that he writes. I find his arguments clear and easy to follow, it’s just that they are full of holes and lacking in substance. Mr. Gwynne believes what thousands of other misled grammarians believe, but I rarely do I see it put so plainly, and for that I do have to congratulate him.

One thing that makes Mr. Gwynne stand out from other prescriptivists is his insistence that grammar is a science, and we need to use evidence in our arguments. Despite this, he is completely unwilling to actually look at any evidence. This is a contradiction that runs throughout his philosophy. This is a revealing passage:

“I hope it can be seen to be reasonable that, provided I have exercised due care in arriving at the facts, I believe prescriptiveness to be more often justified than most modern grammarians do. I believe that I have sufficiently shown, too, that, when assessing what correct grammar is or is not, we should be influenced neither by prevailing fashion nor by present-day majority vote nor by the pronouncements of acknowledged experts – and not even if those experts are unanimous – only by adequate evidence”

If he is not willing to examine how people are actually using language, and he is not willing to listen to experts in the field, where does he go searching for this “adequate” evidence? How do you gather any evidence about language at all, if not through observation of language users?

One thing he does is to appeal to authoritative sources that he likes, in contradiction of his statement that wouldn’t shouldn’t rely in acknowledged experts. For example, in a later part of the preface, he has to explain why he chose to use British spelling when publishing the American edition of the book. He writes:

“…my keeping of the British spelling does not require complete defiance on my part, because, for instance, my 1928 edition of Webster’s Collegiate Dictionary gives ‘practice’ and ‘practise’ as alternative spellings for the verb”.

So it’s OK for him to do this, because an authority said so. Notice also that he uses a 1928 edition of the dictionary. What’s up with that? Gwynne’s Grammar was published in 2013. The reason is probably that new dictionaries won’t support his point, either by not having that spelling, or by explicitly marking it as “British”. For instance, if you type “practise” into Merriam-Webster’s online dictionary, it takes you to the entry for “practice”, and there is only one mention on the page of “practise” being an alternative.

This use of older grammar sources is also part of a larger pattern, because Mr. Gwynne’s other major strategy is to appeal to history. He regularly cites works from before 1950 or even before 1900. In his view, language change is a bad thing that we ought to avoid. In fact it’s so bad, Mr. Gwynn is willing to just pretend it doesn’t happen. Newer forms of language are simply wrong, and we should always prefer the older form of a language. He lays this out pretty clearly in this paragraph:

“I maintain, on the contrary, that, when there is a good enough reason, the traditional rule should be stated all the more uncompromisingly the more it is fading away under the pressure of prevailing fashion-perhaps even stated as one to be defended for all time, and yes, even after the battle seems irretrievably lost.”

Mr. Gwynne is not the first person I have encountered who thinks this way. Still, this position never ceases to amaze me. How far back you gonna go Gwynne? Should we revive all case marking in English? Should we go back to having grammatical genders? Umlaut plurals? A different pronoun system? Velar fricatives?

Actually, that’s all Old English stuff, which is really just a corruption of Proto-Germanic. And when it comes down to it, Proto-Germanic is nothing but badly-spoken Proto-Indo-European. Grimm’s Law has led to serious decay in our language, and I’m sure Mr. Gwynne will agree that the rational thing to do is re-instating Proto-Indo-European pronunciation rules instead of this modern English stuff. Who cares that it happened 10,000 years ago? As Mr. Gwynne says we should carry on fighting, even after the battle seems lost.

This whole line of argumentation is ridiculous. For most people, the logical, rational thing to do is to understand language as existing in a context. It has a time and a place. If you want to provide grammatical advice to modern speakers, you need to be advising them about modern language. Telling people to abandon common sense and go with centuries-old language is ludicrous.

Mr. Gwynne is of course not really arguing that we should speak Proto-Indo-European (or Proto-World or whatever). He realizes the absurd consequences of his argument. He is going to have impose some arbitrary cut-off time of his own liking, and rules that go back any further are rules he’ll ignore. That’s why he put in that little subordinate clause “when there is good enough reason”. These are weasel words for “when it suits my own purposes”.

“As opponents of the teaching of formal grammar delight to point out, English has not remained exactly the same during the last several hundred years….What the same opponents almost always omit to mention, however, is that the changes during this period have been remarkably small.”

This is not limited to “opponents of teaching formal grammar”. Ask anyone with any training in a relevant area: linguists, philologists, lexicographers, historians, anthropologists… They will all tell you that English isn’t the same as it was in the past because it is a documented fact. It is not a conspiracy to prevent formal grammar instruction. Linguists would love to see more formal instruction, not less! This is all a fantasy of Mr. Gwynne’s. But he won’t give up. See if you can swallow this one:

“For instance, Shakespeare can be followed nearly as easily as if the plays and sonnets were written today.”

This claim hinges on some questionable semantics for “nearly as easily”. I can guess that Shakespeare is quite readable to Mr. Gwynne, but that’s because he has a graduate education in literature and he speaks a particular dialect of UK English.

While I grant that Shakespeare is not totally impenetrable to a speaker of Modern English, it is nonetheless extremely common to print translations and explanations of older terms in the margins because most people run into difficulty. Just look at this page for example, which is a student aid that gives side-by-side of Macbeth in Middle and Modern English. It’s certainly not impossible to understand the older variety, but it’s quite a stretch to say I can read it “nearly as easily” as modern English. I have to read more slowly and carefully and think about it a lot more.

This is, of course, why people say it’s good to study Shakespeare: because you have to read it more carefully and think about it. It’s good mental exercise. It’s just silly to say that Shakespearean English is roughly the same as Modern English.

“Words such as ‘thou’ and ‘unto’ have slipped away, and the original meaning of ‘nice’ and been largely lost, and words such as ‘X-Ray’ have been introduced, and ‘mouse’ has acquired a new meaning; but such changes are far too few to make English a different language, as it undoubtedly is compared to the original Anglo-Saxon.”

There is an amazing lack of self-awareness here. Everything he mentions here is fantastic evidence against his own position. Languages do change, and they change a lot, and in a lot of different ways. Yet he can’t actually bring himself to admit that. He glosses over the magnitude of these changes, pretending that what he presents in this paragraph is a totally exhaustive list of every change ever. This is like a creationist arguing that birds didn’t really evolve into different species because they all still have wings and feathers, so modern and ancient birds are basically the same thing.

He is completely wrapped up in his own little world. All that matters is his dialect of English, and how close he thinks it is to Shakespeare. Is he seriously unaware of the diversity in Modern English? Does he consider Australian, Canadian, Texan, Jamaican and South African English to be basically the same thing as Middle English spoken in England 300+ years ago? Come on.

Remember earlier how Mr. Gwynne declared that we should always fight to preserve older grammar rules? I predicted he was going to have an arbitrary cut-off point for when he would no longer care to fight. Now we’ve reached it. As he indicates in the quote above, Anglo-Saxon is as far back as he’s willing to go. That’s the point where Mr. Gwynne thinks language stopped changing to any significant degree, and we can make appeals to history up to that point.  Prior to that, things were too different for us to make comparisons or something and we should ignore those changes because reasons.

To be fair, Mr. Gwynne does attempt offer up some evidence in support of his crazy claim that English hasn’t changed much. This evidence comes in the form of a grammar book published in 1894. I nearly fell off my chair when I read that. In general, he recommends nothing but old books. In the “further reading” section at the back, the vast majority are from before 1950, with at least one from before 1900. There are a few modern books, from the 2000s, but mostly he recommends them because they reject modern usages.

So exactly how little has English changed? Mr. Gwynne says that his 1894 reference book has a complete list of every word that has changed in meaning between 1485 and 1894. A complete list of 400 years of lexical change. Can you believe it? I sure can’t. How large was this book? Mr. Gwynne explains:

“I have counted them, and they total one hundred and twenty-seven. A few other words have been completely lost. The total loss of any kind, in both grammar and vocabulary, is minuscule in the context of the English vocabulary as a whole.”

That’s incredible. He really, truly believes that exactly 127 words changed over four centuries, and nothing else. English is otherwise basically identical. This is obviously false. An afternoon with an etymological dictionary (here’s a really good one for example) will set you straight. I’m not sure that anyone has truly attempted a full count of all the words that have changed, which would probably be infeasible in any case, but there’s just no way the rate of change was that slow.

Mr. Gwynne also tries to sneak in an extra proposition here, by claiming there’s been little change in grammar as well. But actually he presented no evidence of this at all. He only discussed vocabulary, then threw in grammar at the very end. Nice try, but I see what you did there. The reason he has to pull this rhetorical trick is because grammar actually has changed an enormous amount. Word order, case marking, pluralization rules, past-tense formation rules, the pronoun system, and so on. Let’s not forget do-insertion, which affected both question formation and negation of verbs.

Along with this nonsense about language change, we have some self-centered claptrap about how totally awesome English is. (The word ‘claptrap’, by the way, has changed in meaning twice since the 1700s. I wonder if it is on that famed List Of 127).

“Those who speak English today have the prodigious good fortune of having inherited from out ancestors a language which has two really spectacular features: one is that is it the most widely spoken language there has every been. The other is that during the last four centuries, it has been, together with classical Greek and Latin, one of three great vehicles of thought, communication, science and culture of all time.”

English has only become widely spoken recently. It makes no sense to group Greek, Latin, and English together in this way – they were important languages at different periods in time. Further, Latin and Greek were never “world” languages like English is. For example, they have never had any major hold in North America, South America, Asia, Sub-Saharan Africa or Polynesia.

As for them being the most important languages, this is just cheerleading. There are no facts to back this up. Mr. Gwynne likes these languages, he grew up in a culture influenced by them, so they’re the best. If Mr. Gwynne were an American, this paragraph could just as well have been “U-S-A! U-S-A! U-S-A!”.

What about Sanskrit, the language of the Vedas and Hinduism? There is an enormous body of poetry, science, philosophy, medicine, drama and literature in Sanskrit, and it continues to be used as a religious language by hundreds of millions of people. How could a “cultured” British man like Mr. Gwynne never have heard of Sanskrit? How about Chinese, and the huge body of technical, scientific, and literary works written in that language? And why restrict ourselves to written languages? Consider Quechua, the language of the Incan Empire, spoken by millions of people, and still in common use today.

At this point, let me remind you that Mr. Gwynne listed among his qualifications for teaching grammar that he has travelled widely and has met a lot of people. He utterly fails to demonstrate it here. How can you travel widely, but remain stuck in such an ignorant Euro-centric mindset?

Let me finish this by discussing Mr. Gwynne’s most curious achievement: a step by step proof that happiness depends on partly on grammar. This list is my paraphrasing of the proof, except the bit in quotes. Click that link to see an original version. Are you ready for this?

1) For genuine thinking, we need words
2) If we don’t use words correctly, we can’t think correctly. Grammar is the way to use words correctly.
3) If we don’t think correctly, we can’t make good decisions
4) If we can’t make good decision, “we shall make a mess of our lives and also of other people’s lives”
5) If we mess up our lives, we shall be unhappy

Conclusion: Without good grammar, you will be unhappy

In Step 1, Mr. Gwynne uses the term “genuine” thinking to distinguish conscious, complex thoughts about specific things, from automatic thoughts like “I’m hungry”. This is a reasonable distinction to make. But do you really need language to think? He is dead certain that you do. This is actually a major question in science today, in fields like psychology, neurology, linguistics, and artificial intelligence (others too I’m sure).

This isn’t an area I know a lot about myself, so I won’t put my foot in my mouth here, but if you google the terms “thinking without language” you will find some interesting articles right away. The short of it is that there’s no simple yes/no answer. Some thoughts can happen without language. Many thinkers and artists describe their deepest, most profound thoughts as happening without words. Some thoughts clearly are influenced by language, like the thoughts I had to think to write this blog post. I was definitely thinking clearly and on purpose with a language. I’m not going to say Mr. Gwynne is wrong here, but it’s not as obvious or self-evident as he makes it out here.

Let’s even grant that thought requires words, and move on to Step 2. It doesn’t follow from the first premise, because words do not have any inherent meaning. There is no absolute ‘correct’ use or pronunciation of any word. In linguistics, this concept is called ‘arbitrariness (of the sign)’. The fact that different languages refer to the same object with different words is strong evidence that words don’t have necessary logical connections to meanings. Additionally, the fact that the both the semantics and pronunciation of words change over the time, shows that meanings and words are not necessarily connected in any way. There is no absolute ‘right’ way to use a word. It’s all contextual. Mr. Gwynne doesn’t really care about any of this though, he just wants us to use words his preferred way, and he’s totally unaware of the real linguistic diversity in the world.

I guess I’m OK with Steps 3-6, but the first half of this argument is on such shaky grounds I think we can safely reject the whole thing. In his book, Mr. Gwynne follows this short argument with a longer more detailed version, but it has the same fundamental problems. This second version, however, gets hilariously alarmist, and by the last step:

“Step nine. Would that the harmful effects of bad grammar stopped there. They do not. Civilization itself exists only in the various societies that make it up. If enough societies in the world crumble as a result of bad decisions taken because of bad thinking, yes, the whole of world civilisation faces collapse, with consequences for each individual that are literally incalculable.”

I have, on some occasions, made fun of prescriptivists by saying things like “Oh no, if we keep saying “him and I” then the English-speaking society will collapse”. That was always hyperbole. I never thought I would see someone seriously advance this argument, let alone as part of a detailed 9-step process.

Despite the lofty promises of being evidence-based, Gwynne’s Grammar provides nothing of use to any modern reader. He might as well be offering charms and relics to ward of evil grammar demons for all the science we can find in here. I cannot recommend enough that you avoid this book.


Filed under Book Review, Linguistics

5 responses to “Gwynne’s Grammar

  1. Maybe a better way to put it is that thinking about ideas requires words. They are certainly necessary to communicate them to others. (And, really, what value do they have if not communicated and discussed?)

    This is exactly why I have concerns about the “decay” of modern usage. The apparent loss of richness, precision, and nuance, directly impacts our ability to effectively communicate ideas. I suspect it may even impact our ability to think about ideas.


    • Like I said, it is worth googing the terms “thinking without language”. You don’t need to just have suspicions – why not look for some evidence to confirm or deny them?

      There are also some other consequences to your position that you should consider. If you can’t think with words, then animals cannot think since they have no words. Pre-linguistic babies cannot think. Depending on your definition of “word”, sign language users could be excluded from thinking individuals.

      You are also committed to a strong version of the Sapir-Whorf hypothesis that language influences thought, and that speakers of different language think in fundamentally different ways. Your position also implies that bilingual individuals can change how they think by changing language (would be difficult to explain the phenomenon of “code switching” in this case). There is limited scientific evidence to support these ideas, and much of the work in this area is controversial.

      On the question of language decay, I would pose to you the same question as I did Gwynne: where is your cut-off point? Let’s take the loss of case-marking. In modern English, only some pronouns show case. In Old English, all nouns showed case. Presumably, you don’t view this loss as decay or something that has seriously hindered English. However, had you lived during the time that cases were disappearing, I get the feeling you would have complained loudly about the precision or nuance being lost. I’m certain that there really were people just like you back then, making exactly this complaint.

      Today, however, everything is fine and communication is not inhibited in any way by this language change. The stricter word order rules that replaced case marking have allowed English speakers to carry on communicating just well as they used to.

      I guess what I’m saying is that you should take a look at the bigger picture. Whatever change you think is “bad” right now will eventually look just as meaningless as the loss of case-marking, and English won’t be worse for it.


      • Don’t assume I didn’t look at look at some of those links or that I don’t have some past experience with regard to these matters. Both assumptions would be incorrect. Also, please note that I used the word “ideas” twice; that was not casual but crucial.

        Here are some quotes from linguist Arika Okrent’s article — the first that appears when I Google [thinking without language]:

        “Can you experience sensations, impressions, feelings without language? Yes, and very few would argue otherwise. But there is a difference between being able to experience, say, pain or light, and possessing the concepts “pain” and “light.” Most would say true thought entails having the concepts.”

        She goes on to talk about how artist Temple Grandin says that her concept of “dog” is tied to all dogs she’s ever know (so it mine, for that matter). But she (and I) still have the concept “dog” and that concept has a label.

        The terms “thinking” and “words” need to be carefully defined to have a discussion like this. That’s why I carefully associated the word “ideas” with “thinking.” The realm of thought is a spectrum. At one extreme is raw experience: pain, pleasure, hunger, anger. These immediate experiences don’t require words (although communicating them does). At another extreme is intellectual thought, and this — almost by definition — is word-based.

        Or rather, idea-based. And words are handles on ideas. They’re labels applied to classes of objects. All apples are, in reality, distinct objects, no two exactly alike. But the word, “apple”, applies to a class of objects to which all apples belong. (The English word is local; other languages, and other times, have other words, obviously.)

        These class labels don’t even have to be language words. Consider the international symbol for “No!” for example. Or the use of some kind of recognizable “male” and “female” symbols for bathroom doors.

        Arika concludes:

        “However, while it appears that we can indeed think without language, it is also the case that there are certain kinds of thinking that are made possible by language. Language gives us symbols we can use to fix ideas, reflect on them and hold them up for observation. It allows for a level of abstract reasoning we wouldn’t have otherwise. […] We may be able to think without language, but language lets us know that we are thinking.”

        Which is exactly what I said.

        In Matthew Ankeny’s blog article he concludes:

        “Put simply: the words you know and use are going to affect the way you act. […] So the better words you have, the better you can describe and (ostensibly) understand and experience the world. […] Even so, for those of us with language (that’s you, dear reader), the more language we acquire the more nuanced and complex the world becomes.”

        Which, again, is what I said.

        I’m not talking about strong Whorfianism, although I think there is some substance to the weak version. An extreme example of the strong version suggests that New World natives would fail to see a large European ship in their waters because they had no terminology for it. I think that’s utter nonsense.

        I do think poor command of language limits thinking. It clearly limits precision and expressiveness. A very good, and humorous, example is found here:

        I’m not interested, at all, in preserving some specific version of language. I delight in the dynamism of the English language. My complaint is about those who learn only a tiny portion of it and, in doing so, limit their ability to think about, and express, ideas. This isn’t about what English is, but how it’s used.


  2. Dr Terry Denman

    Gwynne is mad. He thinks the sun orbits the earth and that the atom does not exist. He thinks there was no Holocaust and that Hitler was an agent of the Jews. He was (is?) an anti-Semite. He has been charged with offences against children in Australia. His “grammar” is of this type of nonsense.

    Liked by 1 person

  3. I prefer Humpty Dumpty’s logic when he told Alice that,

    “Words mean what I want them to mean, nothing more, nothing less.”

    Has Gwynne read Alice?


Tell the world what you think!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s