A Critique of Hugh Dellar on Celta


I don’t like to get involved in arguments or debates online.  It just happens from time to time.

This is one of those times.

Hugh Dellar, former Celta tutor, teacher, teacher trainer, coursebook writer, and presenter, recently wrote an impassioned and typically strident critique of Celta and courses of its ilk in the aftermath of what many are seeing as a seminal plenary talk by Silvana Richardson at IATEFL 2016 on the topic of native speaker vs. non-native speaker bias in our profession.

Hugh basically puts forward a case against Celta, and calls for its replacement with something better.

I have nothing against the idea of establishing better qualifications and standards for our profession, but this will only succeed when the argumentation is sound, and so I feel compelled to respond to Hugh’s key arguments.

As I make my living working on short introductory training courses like Celta, it isn’t surprising that I may defend them against criticism.  Before you let such ad hominem thoughts discount the rest of what I have to say, let me assure you that I will be careful to present evidence and give ground as I go along.

I hope to show that while they are superficially persuasive, Hugh’s arguments in his latest post frequently lack either logic or evidence or both.

Despite this, I will end up agreeing with some of what he says and present an (arguably) more strident conclusion of my own.

If you haven’t read Hugh’s post yet, stop here and go and do that now.  It’s worth it.  I’ll wait for you.

For the purposes of this critique we can ignore the first two paragraphs of Hugh’s post,  being just an introduction to the issue for unfamiliar readers.  Suffice to say, after a few caveats about how there are some good trainers and good courses out there doing good work, Hugh gets down to the serious business of straw-manning the Celta.

The first real and relevant assertion Hugh makes is this:

“the current status quo (that) positions CELTA courses as the gold-standard entry point into teaching EFL as a global profession”

The problem here is that Hugh, straight off the bat, is not making an argument but is instead begging the question: is Celta in fact the gold-standard entry point into teaching EFL as a global profession?

The short answer, is: no, it isn’t.  And for a rather obvious reason.

As Hugh himself goes on to point out, the actual entry point for teaching EFL as a global profession is not the Celta but rather no qualification at all.

While 10,000-12,000 people take the Celta each year, a considerably higher number will with great likelihood commence work at private sector ELT institutes worldwide without any training whatsoever.

While this cannot be supported by data for obvious reasons, it is a logical conclusion to draw – one would expect people to take the line of least resistance and thus more people would seek to enter the profession without any qualification if they can than work to obtain one.

This being the case, Hugh has made a rhetorically decorous, but factually unfounded assertion about the market leverage of the Celta, and it does not bear up to logical scrutiny, “gold-standard” or otherwise.

This does not invalidate his point completely, especially as I am not presenting hard evidence either yet, so let’s look at it more closely for a moment.

Hugh’s actual point is a bit more subtle (or slippery) than that.  He says that a “current status quo” somehow “positions” Celta as this “gold standard”.  In other words, there are players operating to raise the Celta to the level of gatekeeper qualification.  At this point in his argument we are left to infer who these may be: the qualification body and its shills, presumably, plus some useful idiots.  As they play no further part in Hugh’s argument, we, like he does, will leave them here.

So for now, let’s just bear in mind that Hugh has yet to buttress his claim that the Celta is a “gold-standard” (by which we can only assume he means either necessary or privileged) entry qualification.  Let’s move on.

Hugh then goes on to make several assertions, strung together: He claims that Celta courses:

“…offer an inadequate entry to the profession, perpetuate the devalued status of EFL teachers around the world . . . and, perhaps most pertinently in the context of this post, inordinately favour native-speakers.”

Now, there is a lot here, so let’s move through it in slow-time.

His first point is that Celta offers an inadequate entry to the profession.  This not only begs the question “does it in fact fail to be adequate as an entry to the profession?”, but it also  avoids any description of what basic entry parameters would look like.

Obviously, therefore, to respond to Hugh’s first assertion, we will have to fill in some background for him.

Adequate entry or back-door entry?

If we presume that by “an adequate entry” to the profession, Hugh would mean that, given support in the initial period, a qualification holder would be able to plan and deliver effective lessons, then the Celta certainly claims that it meets this standard.  Indeed, this is a summary of its PASS level grade descriptors.  Note the relationship of these performance standards with “ongoing support in-post” for an “initial period” post-qualification.

Now, the fact that the Celta claims that its holders can operate successfully with support in no way means that such support is in fact provided, and here I grant Hugh and I would find clear common ground in criticising this fact.  

This is, incidentally, a likely push factor out of the profession; estimates place attrition over two years post-celta at upwards of 66%. If that figure shocks you and leads you to think that it might be evidence to criticise the course as being unfit for purpose, hold fire: the figures are similar for traditional state teacher training in the UK. 

While the absence of this support certainly makes a smooth transition from training to occupation a false hope, this would be more a failure of “The Market” (as Hugh would put it) more generally, than the qualification.

Hugh also, we may note in passing, makes the claim that Celta “perpetuates the devalued status of EFL teachers around the world”.  Here there are in fact two claims: 1) EFL teachers around the world are devalued, and 2) Celta perpetuates this state of affairs.

Claim (1), at least within the unregulated private sector, is probably true and uncontroversial; claim (2) however, does not logically follow from it.

The presence of a qualification like Celta establishes some minimum professional standards and is therefore a good thing.  The fact that so few teachers go on to take a higher qualification (less than 1 teacher for every 10 taking the Celta takes the Delta, for example) is a problem, but without the Celta, these beginning teachers would probably simply start with no qualification at all.

Hugh almost supports his own call for a more richly developed type of initial qualification via this point, but we need to bear in mind that for B to be better than A, A need not be bad.  To make such a claim would be elementary false logic, and I suspect Hugh is aware of this but got carried away.   

If Hugh had simply said “can we agree that, in an ideal world, the entry level qualification for our profession would be qualitatively higher than it is now?”, I suspect he would have met with no opposition.

But when Hugh argues later in his post that the solution to such low professional standards in our field is for training centers to start boycotting Celta, he misses the point by a wide margin.  It isn’t qualifications we need to be scapegoating or boycotting, but teachers, schools and institutions who do not invest in their own or staff’s continuing professional development.

However, we will let this all stand for now and move on.  His boldest claim, perhaps getting carried away by the N/NEST debate at IATEFL recently, is that the Celta qualification:

“inordinately favors native speakers.”

It isn’t apparent on which evidence, if any, Hugh is basing this claim.

And, once again, it is important for the reader to recall that this is all it is: a claim.  Not a statement of fact but one of opinion.

Now, this is quite a claim, and it would normally demand some evidence in its support.

Hugh provides none as far as I can see, beyond the anecdotal (he uses his own experience, as a non-L2 learning literature student who got through on charm and good fortune, by his own account – I suspect there was a bit more to it than that, but be that as it may…)

The plural of anecdote is not data, as we all know, so it would be useful to look at some (albeit partial)  data.

“You can prove anything with facts, can’t you?”

But before we can do that, we need to know what “favoring native speakers” would look like in practice.  Does Hugh mean that “native-speakers” find it easier to enroll?  to pass?  To gain employment afterwards?  It appears mainly the last of these, but to be on the safe side, let’s take each of these in turn.

Given that the standard enrollment procedure for a Celta typically requires applicants to pass a language awareness test, an interview, which typically includes language analysis, practical teaching ideas, frequently teamwork tasks and – incidentally – language competence, it is difficult to see where a “native-speaker” would have an advantage over a “non-native-speaker”, assuming the “non-native-speaker” had a level of English adequate to teaching the language anyway.

So Hugh can’t mean this.

If the Celta were easier for “native-speakers” to pass (or pass well) than “non-native-speakers”, then this is what we would expect to see in the grades data.

This isn’t the case, however.

In an analysis conducted a few years back of data drawn from my own centre (150 candidates per year), while we have a 60:40 “native-speaker”:”non-native-speaker” intake, we have a higher percentage year on year of “non-native-speaker” passes, and this increases the higher the level of pass.

Incidentally, it also tends to be higher for women.

So, if you want to have the best chances of doing well at Celta (at least in the North East of Germany), try to be a “non-native-speaker” woman.

ASIDE: In case you are interested, and bearing in mind that everything here is simply a statistical trend, not a quota or suchlike, the next most successful group are North American post-grads (slight bias towards women again).  And if you want to fail the course, you can be more or less anything and come from anywhere – the fail grade, it appears, really isn’t choosy.

So when Hugh says, as he does later in his post, that “I was as qualified as any one of my Polish or Brazilian colleagues who’d done the same course”, he is both correct and incorrect at the same time.

He is correct insofar as both he and his “non-native-speaker” course-mates all obtained the same qualification, which is unremarkable given they all enrolled in the same course leading to the same qualification (it would be more surprising if this were not the case); he is incorrect in his own terms to say that this made him as “qualified” in a broader sense, as he establishes clearly elsewhere in his own post.

Language learning experience, both in terms of early-years bilingualism and formal study, place them at an advantage over him.  Hugh also explains in the comment thread to his post that this is what he means.  “Non-native-speakers” should be at an advantage over “native-speakers”.  The only question is: does The Market see it the same way?

But this begs another question: are all (or even most) “native-speaker” celta enrollees similar to the picture painted by Hugh, that is to say lacking formal or informal language learning experience or success, teaching experience of some kind, etc?

Again, the evidence suggests no.

While our centre eliminates applicants at the pre-interview stage where they have no language learning background and tell them to go away and try to learn a language before trying to teach one, we set the bar pretty low.  Any experience of language lessons, regardless of success, is enough to keep the chance of an interview open.

However, the number of candidates we accept who meet only this minimal standard is vanishingly small, so we aren’t applying a statistically significant filter.

The vast majority of accepted applicants at our centre speak at least one other language fluently, with a small majority speaking 2-4 languages, acquired either in school/at university or learned while staying in the country.  This is fairly evenly distributed across “native-speaker” and “non-native-speaker” applicants, and much the same distribution can be seen when we look at teaching and coaching experience (though here, for both “native-speaker” and “non-native-speaker” applicants, we see much less second language teaching experience and more generally-relevant teaching experience.)

I see no reason to think my centre is radically different in this regard from any other overseas centre, though I grant that the same may not hold in UK/US or similarly located centres.  Nonetheless, on this basis I would dispute Hugh’s claim that “native-speaker” applicants typically display less of a language learning or teaching background than “non-native-speaker” applicants. Life moves on, and so does the Youth of Today, perhaps. 

Teapots and trapdoors

Hugh’s portrait of a typical “native-speaker” applicant – For natives such as myself, however, who arrive on courses often having spectacularly failed to learn a foreign language at school, with little or no ability to articulate and explain how language works and with no teaching experience whatsoever, a CELTA offers a crash course in how to fake it – is so grotesquely caricatured as to bear virtually no relationship to the data I have available.

Perhaps such sorry specimens get accepted on some courses, but such aberrations occur everywhere – including on PGCEs, BAs, MAs and so on.  Much like Bertrand Russell’s cosmic Teapot, I cannot prove such failures of the application process don’t exist; I simply see little evidence for them near me.

So the only point left with a leg to stand on is the argument that weak “native-speaker” cert holders are valued as much or more than even strong pass “non-native-speaker” cert holders.

Here, sadly, I suspect Hugh is on firm ground.

I say “sadly” not because I am loathe to agree with Hugh – far from it – but rather because of all the many great teachers who are potentially mistreated as a result of this claim perhaps being true.

At IATEFL I spoke to just such a “non-native-speaker” teacher who also happened to be a Celta holder.  They had, like Hugh, been impressed by Silvana Richardson’s plenary, and had been left with a sour taste in the mouth because, like Hugh, they saw a basic injustice in The Market relevant to this discussion.

As they put it, in large parts of the world, even if a teacher holds the Celta (the “gold-standard” entry – level qualification, as we recall), they still get no job offers as long as they are not “native-speakers”.

In other words, the qualification does not “level the playing field” for “native-speakers”; it is far more a bait-and-switch trick for “non-native-speakers”.

If this were true, the Celta would be implicated in a much larger injustice against “non-native-speaker” teachers than I think Hugh even himself thought to formulate: they would be encouraged and enticed by The Market to invest time, money and hope in gaining a qualification like Celta, but afterwards, their hopes of leveraging this “gold-standard” would be cruelly dashed by the lack of job offers granted them.

If this were true, it would be a travesty and a disgrace.

This, in itself, is a very large claim, and hence I make it very tentatively and highly provisionally.  I would like to see longitudinal data showing the employment success, by country, for “non-native-speaker” takers of the award in comparison to “native-speaker” takers; this would settle the argument once and for all, and such data would not be impossible to obtain.  Any and all “non-native-speaker” holders of Celta who have successfully started a teaching career, especially outside your own country, please make yourselves known in the comments section below!

But assuming for a moment it is true, what then?  The intellectually honest thing to do, it seems to me, would be for such courses aimed at the unregulated private sector (only a part of the larger ELT education sector, lest we forget) to stop accepting “non-native-speaker” trainees, as the qualification would provide no leverage for them in the target market place.  As a “native-speaker”, I am disbarred from taking a Cambridge English language proficiency exam for similar reasons.

The morally correct and professionally sustainable thing to do, on the other hand, may well be to explore the idea of raising the qualification bar for our profession; but while these are areas to explore and are certainly inspiring, this won’t be achieved simply by boycotting current solutions or bringing in competition alone.

Everyone, from the top brass working within qualifications providers right down to the  client on the street walking into the local language school for a placement test, needs to change their mind on a lot of things, including – but not limited to – the status of “non-native speaker” teachers and what it means to be “qualified” to teach.

This is the real problem that I believe Hugh is trying to crack, and – despite everything I have said here about his argumentation (in the interests of widening the debate, rather than closing it) – it is one on which I support him and against which I am willing to help apply pressure.


  1. Pingback: IATEFL Birmingham 2016: Teacher training and CPD | Sandy Millin
  2. Pingback: A NNEST perspective on teacher development: a complicity of silence? | teflreflections

Join the conversation...

This site uses Akismet to reduce spam. Learn how your comment data is processed.