The “conservatism” of JK Rowling

I’m always interested to read people’s (reasoned) thoughts on the Harry Potter books and was intrigued by this article posted today on a blog I keep tabs on:

I think Matthew Hosier’s idea that Rowling’s writing is basically conservative (if that’s a fair way of putting it; maybe it isn’t! – but you can read the article for yourself and get a better idea of what his basic claim is) is probably not the right one. Now, I don’t think Rowling is actually super-progressive: her politics in general seem fairly centre-left; she is a vocal detractor of Jeremy Corbyn, apparently basically because he’s too left-wing for her; more progressive/socialist types have a tendency to get annoyed at her (see for example this post and its follow-up(s) on another, very different blog I follow). But I think it would be a mistake to describe her as conservative, and any apparently conservative elements in her writing I would suggest probably have another source than repressed conservatism, an actual hankering for a return to an (idealised) 1950s, etc. Rather I think what’s going on is a sort of historical romanticism (or romanticisation of the past).

This is ultimately not about political viewpoints but aesthetic effects. (Obviously the two bleed into one another and authors need to be careful with this.) The Harry Potter books are, by and large, going for a variant on the aesthetic of classic children’s literature (the sort of books, probably, that Rowling herself grew up with). And this is why the heavily-present boarding school element, the lack of divorce etc. etc. come in. Major elements of the traditional aesthetic are imported in wholesale. Elements which are strongly at odds with it tend to be left out.

But this is still basically an aesthetic thing. I don’t think Rowling is pro-boarding school any more than CS Lewis or JRR Tolkien were (at a guess) pro-absolute monarchy. But just as a mediaeval aesthetic gives you kings and queens, so a classic children’s literature aesthetic gives you boarding schools and traditional marriage. It needn’t say anything about the author’s own politics, conscious or otherwise.

Some thoughts on language teaching in schools

This was originally posted on the “essays” section of my website, but it might as well go here as well.

Recently, there has been quite a lot of bother – brought to my attention by one particular Twitter account which I follow closely – about the way in which the UK government wants English to be taught in schools, and as language and writing are both topics close to my heart I thought I might presume to weigh in with some remarks of my own.

Before I go any further, a few comments on who I am – and who I aren’t. I am someone who has studied language at a high level, having two degrees in linguistics and working toward a third (a PhD, with a substantial concentration on the linguistics of English as well as other languages). I am someone who greatly enjoys reading and writing fiction, and who reads and writes a lot. (More specifically on the writing front, I have written several novels that are complete in substance even if they still require some polishing, as well as some poetry and a number of plays. None of this, of course, means I am necessarily any good at writing.) I am not someone who has ever taught in school (though I have taught a little in other contexts), nor someone who has ever studied the theory of education in any depth at all. I do not have children of my own.

And some background on the topic of present discussion: the current situation, as far as I can glean, is that the government is pushing heavily a particular approach to the teaching of English in schools with a strong focus on spelling, punctuation and grammar (“SPaG”). Specificially, this takes the form of a quite strongly traditional approach to language teaching involving requiring primary-age children to be able to identify parts of speech (and not just the basic ones like “noun” and “verb”), “fronted adverbials” and so forth, mixed in with a heavy dose of prescriptivism about where commas should go and so forth. The aim is to help children to “write better”. This approach has been the cause of great consternation amongst, apparently, most people with some kind of interest in children’s education who aren’t in the government.

Firstly, let me say that I am not opposed to the teaching of grammar in schools. Indeed, I would be very disappointed if grammar were not taught in schools. Some of the responses to the present situation seem broadly to imply that formal grammar shouldn’t be taught in schools at all, which I think is wrong-headed. Now, to an extent I’m biased because I study grammar full-time and greatly enjoy doing so, so maybe you should just ignore me on this. But it seems to me that grammar is one of those things – like maths, science, history, music and so forth – which definitely should be taught in schools, as part of a basic curriculum. Any practical worth is by-the-by here: we don’t teach children who Henry VIII is because knowing it is likely to be terribly helpful in their future lives, and most people, dare I say it, will live their lives quite happily without ever using a quadratic equation or thinking about the chemical properties of sulphuric acid after they’ve finished their GCSEs. But these things should still be taught – because they’re part of our basic cultural heritage, and because they give people who might want or need to take these things further a starting point from which to do so.

But let me also say that any teaching needs to be appropriate, and by that I’m thinking primarily of age-appropriateness. It’s no good trying to teach young children to get to grips with concepts that are too complicated for them. We teach primary school children arithmetic, but not differential calculus, for a reason. For a similar reason, we need to be careful about what we try to teach them about language.

Let me make clear, as well, that I’m not opposed to the teaching of Standard English. The sad reality is that the way people write and speak affects their chances in life – that things like an accent, or a particular way of forming the past tense, or non-comformity to arbitrary spelling rules are used to cast aspersions on people’s intelligence, to permit or deny them jobs and university places. Perhaps this is something we should be working to change – but it’s not something that’s going to change quickly, if it can be changed at all, and in the meantime we ought to be equipping children to get round these challenges as best they can.

But there is Standard English, and there is over-the-top prescriptivism. There are rules you need to follow to get on in life, and there are rules which nobody cares about except annoying pedants who nobody likes and are mostly interested in asserting their own superiority – and who probably don’t follow their own rules consistently anyway. We need to make a distinction between the two, and not end up teaching the second in place of the first.

Grammar is the theory of language – and like any theory of anything, it’s incomplete. There are things we don’t know at all, and things that are unclear: things that people disagree on. This is arguably much more true of grammar than it is of, say, physics or maths. And, given this, it makes no sense to try and teach grammar as if everything were clear-cut — as if there’s always a “right answer”. The present approach seems to be trying to do this, but that’s not fair, because even within the domain of traditional grammar there’s quite a lot of uncertainty and fuzziness. (And quite often, what we might think about language will turn out to be just plain wrong.) As an example of uncertainty: is the word running a noun or a verb, or both, or neither? It’s not obvious — not even to me, and certainly not to your average six-year-old. There are limitations to the terminology, and often further confusion created by having multiple different terms for what may be a single concept (e.g. “conjunctions” and “connectives”). Trying to teach grammar by assuming that everything’s set in stone just leads to problems, and throwing in what amounts to little better than arbitrary prescriptive judgements just makes things worse.

We can also look at grammar as the science of language. And like any science, you can approach it as rote-learning of facts, but probably a better way is to allow students to discover it for themselves. Why not ask pupils to see what sorts of different words they can identify, rather than trying to force imperfect labels on them from above? An advantage of this is that it leaves more room for the fuzziness and lack of clarity that is part of the study of language as done by professionals. (Of course, you don’t want to take this too far, and risk the children reaching utterly the wrong idea and being stuck with it forevermore because you won’t let them know what other people think.) And a great thing about studying language, as opposed to (say) the behaviour of electrons or the movements of the planets, is that vast amounts of data are available to all of us – there’s no need for expensive equipment or anything like that.

Finally, on the topic of writing better – I don’t think studying grammar helps very much at all. And I say this as someone who, as I have hopefully made clear, thinks the study of grammar is great thing. But quite frankly, whether you can identify a fronted adverbial or not makes very little difference to how good a writer you are. Shakespeare and Chaucer would have had at best very primitive educations in English grammar, and it didn’t hurt their writing one bit. The best way to get good at writing, as anything else, is to practise it. I think everyone – children and teachers and parents – would probably be much happier (and, in the children’s case, much better equipped for life) if they spent a little bit less time trying to force words into boxes (“the noun box here, the verb box here, the subordinating conjunction box here”) and a bit more time actually practising using language.

I see red

This post is a slight revision of a couple of previous posts that I had on another blog. The intended purpose of that blog was to comment occasionally about infrastructure issues affecting pedestrians in the UK, as I believe that pedestrians are very often unfairly discriminated against in favour of motor transport in the way our roads are designed. However, I wasn’t able to devote enough time to the project to make it worthwhile, and it seems to make more sense to just have one blog where I might talk about pedestrian issues from time to time alongside other things of personal interest (e.g. linguistics).

The starting point was an article in the Cambridge News, now sadly removed from the online archive as has happened to most older articles on the side. The article was about a UKIP campaigner complaining about a particular type of pedestrian crossings called “puffin crossings”, a matter which happens to be close to my own heart. The most salient feature of these is that the “pedestrian signals” – the red and green man – are on little yellow boxes on stands on the pavement, designed to be viewed from the side you are crossing from. This contrasts with the traditional “pelican crossing” design where the red and green man are displayed on the opposite side of the road at the same height as the traffic lights, facing you as you walk toward them. (Puffin and pelican crossings differ in other ways which I’ll get back to later. Apparently some puffin crossings do have farside signals, but these aren’t very common.)

Now I don’t often concur with the things UKIP people like to get incited about, but on this issue I am fully in agreement. I find nearside signals extremely annoying, because they require me to turn myself to face in a direction I don’t want to go, which is far from ideal. Sometimes I do just turn my head, but a lot of the time I find that can be quite literally a pain in the neck. And some of the time, as this UKIP guy points out, you can’t see them at all, because there are people in the way. I usually just end up watching the traffic lights side-on waiting for them to go red.

(I am aware of at least one set of signals so badly placed that you have to face away from the road entirely to see them change. This may come up in a future post, if I ever get around to writing it.)

Anyway, the county council’s response to all this, according to the News, is that “Puffin crossings have been proven to be around 20 per cent safer than pelican crossings“.

Unfortunately it is difficult to know where the “proof” that puffin crossings are 20% safer has come from. Looking online I can’t find a great deal, but one study that gets referred to repeatedly is the imaginatively named Puffin pedestrian crossing accident study, by Maxwell et al. (2011) (you’ll have to sign up to read it but that seems to be quite easy). So, what does this study actually find?

Well, it does report that when pelican crossings are replaced by puffin crossings the newer type show fewer “accidents”. The overall reduction in accidents across both types of crossing studied is 19%. Accidents involving pedestrians are reduced by 24%, which is “statistically significant at the 10% level” (i.e. not really all that significant, although the authors don’t say this explicitly). Accidents involving pedestrians were reduced by 39% at junctions and 22% elsewhere, which the authors do explicitly acknowledge aren’t statistically significant results. This is not, therefore, the most convincing set of data you will ever encounter.

Even if we do accept that puffin crossings are safer for pedestrians, why is this? It may be that, by placing the red and green men to encourage people to look towards oncoming traffic, you make them less likely to step out in front of an oncoming vehicle: but even the authors of this study admit that this is “indicated but not fully proven”. (This logic seems somewhat flawed to me, in any case. You might be more likely to notice cars on your side of the road, but at the same time you are surely less likely to notice them coming the other way, so it’s not at all clear that you are necessarily less likely to get hit on account of the signal position.)

It may well be that puffin crossings are safer for reasons entirely separate from the presence of the nearside pedestrian signals. Puffin crossings also differ from pelican crossings in that only the older style have a “flashing green man” phase, which according to the study do seem to be an important contributory factor in accident rates. And puffin crossings are fitted with detectors to allow the red traffic light to stay red until pedestrians have finished crossing, which you would also expect might contribute to a lower accident rate.

So it may be entirely wrong to claim that nearside signals make a crossing safer. There is a strong possibility that it is the other advantages of the puffin design that are responsible for the reduced rate of accidents – assuming the rate of accidents really is reduced, which as we have seen is not as clear as we might like it to be.

It seems to me that the county council’s response is a flawed one based on misunderstanding or over-interpretation of the data. There is no clear reason to suppose that nearside signals are better than farside ones.

To compound matters further, consider this article (like the last one, you’ll have to sign up with the website to read it, but that doesn’t seem to be too tricky). Its title is Trials of farside pedestrian signals at a Puffin crossing, which is fairly self-explanatory. The authors find (p. 69):

Overall, the effects shown in this study of the introduction of a farside signal at a Puffin crossing were neutral or negative, except for the pedestrians’ stated [p]reference for the modified arrangement.

I would have thought that if an arrangement is preferred by the very people it is designed for this is a very strong argument in its favour! However, the authors don’t make a big thing of this.

The negative effects in question seem to be (p. 61):

  • people perceived a longer delay before the green man phase;
    (Nb. this is not to say there actually was a longer delay.)
  • increased uncertainty about when you should start to cross, and reduced understanding of the signals;
    (This seems odd to me, as surely the signals are pretty self-explanatory; possibly it relates primarily to confusion during the “blackout period” when neither a red nor a green man is displayed – I wonder if something could be done about this,)
  • fewer people “strongly agreeing” that it is easy to see the red/green men while starting to cross.
    (Once you factor in people “agreeing” as well as “strongly agreeing”, though, the modified arrangement comes out slightly on top.)

Also pedestrians were less likely to look left and right before and during crossing (p. 62). This might be thought to make things more dangerous, though note that this has not – as far as I am aware – been conclusively demonstrated. In any case, a slight increase in danger may be a reasonable trade-off for a large increase in convenience: and perhaps our main focus should be on the criminal drivers putting people’s lives at risk by jumping red lights! Ideally, if the green man is displayed, there should be no need for a pedestrian to look at all (except perhaps for a police car or ambulance with its sirens on, which most people would notice even without looking) – but of course we know drivers do not always follow the rules.

One major advantage of nearside signals which I had not previously thought of is that they may be easier for people with various types of disabilities. However, this doesn’t mean that farside signals shouldn’t be present as well, as they were here. (I suppose a disadvantage of having both would be that it is more expensive.)

So ultimately there may be some slight disadvantages of having farside signals, but I am still not convinced there is a strong argument against them, and it does seem I’m not alone in preferring them.

The names of the gods

Inspired by this thread, I’ve been looking a bit into the relative frequency of Greek and Roman god names, using Google Ngrams as a tool.

The main finding seems to be that, for whatever reason, most Greek god names were not in particularly wide usage in English before the mid-nineteenth century, when many of them saw a notable uptick in frequency that has continued through to the present day.

My approach has been to stick various god names into Ngrams (generally with the timescale 1800–2008; 2008 being the most recent one can go) and see what comes out. The big caveat here is that we can’t always be sure if the words are actually being used to refer to the gods themselves: often they might be applied to other things. For example, Mercury, Venus, Mars etc. are the names of planets; Hermes is a popular name for postal services; Hades is sometimes used to mean “Hell”, etc.

Sometimes the Greek form has been around for quite a while. This is the case with Hermes for example:

You can see a not dissimilar graph for yourself if you compare Hades and Pluto. Note, though, that the Roman form is still generally the more frequent.

In many other cases, though, the Greek form only really starts to be used towards the middle of the 1800s, with a gradual overall increase in frequency since. This is the case with Poseidon, for example, which only really started to be used in about 1830, with another big jump in the 1860s:

There may be a correspondence between increased use of Greek Poseidon and decreased use of Roman Neptune, suggesting the first is being used in place of the second.

We see a similar pattern with Greek Zeus and Roman Jupiter/Jove – here, the increase in the Greek form at the expense of the Roman across the nineteenth century is quite marked:

The Roman forms remain more popular than the Greek, however.

Greek Aphrodite and Artemis behave similarly (opposed to Roman Venus and Diana), though the Greek forms in these cases has never been very popular. Again, you can see this if you put the names into Ngrams yourself. Greek Dionysus has gone from being very rare to competing with Roman Bacchus in frequency:

Greek Eros shows a similar pattern as opposed to Roman Cupid. Greek Persephone, barely used 200 years ago, is now maybe three times more frequent than the Roman Proserpine.

The Greek Hera, Athena (= Roman Juno, Minerva) were less relatively unfrequent to begin with, but there’s still a notable jump at a similar point in time, and nowadays the Greek and Roman forms are about as common as one another, as the graph for Athena and Minerva shows:

Some Greek forms have never been popular. This is the case for example with Cronos (poor Cronos):

Ares (=Mars) and Hephaistos (=Vulcan) have similarly remained unpopular. However, with all three if we leave the Roman forms off the graph we can still see an increase in frequency in a similar time period to before:

The general pattern then, to reiterate, is that many Greek forms only really take off in English in the mid-1800s, though there are a few exceptions. In some cases, the Roman forms remain more popular; in others both variants are now about equally frequent. In at least one, it’s the Greek form that is now dominant.

I don’t know what exactly the explanation for these patterns is, but perhaps someone else does.