AI in the classroom?

If you happen to follow Philip Kerr’s blog, you will know that he has recently published this post called AI and Teacher Training where he talks about ChatGPT. In short, ChatGPT is a new AI chatbot that is trained using machine-learning algorithms and can understand open-ended queries. Not only does it comprehend orders, but it also understands the language – even Czech (my native language).

Well, it probably doesn’t come as a surprise that I, as an English teacher, find this tool truly fascinating. Not only can ChatGPT answer questions, but it can also write essays. This could come in handy in many situations in the classroom. For example, we can use the tool to train our students to give detailed answers when answering questions during exams. Also, ChatGPT could be used as a research tool for the students to find suitable material for their essays and presentations. And finally, it may be very useful for the teachers themselves if, for instance, they need another source to prepare for their lectures, conferences, webinars, etc.

What I also like about ChatGPT is that it remembers what the user said earlier in the conversation but most importantly, it is trained to decline inappropriate requests, so it should be relatively safe in the classroom.

Plus, it can definitely be fun to use. Alternatively, you can teach students how to diplomatically avoid answering odd questions, how to be polite when answering sensitive questions, etc. Here’s an example.

I’m not saying that this is something you can’t do with a simple Google search, but the fact that you can actually see the bot typing the answers in real time, so to speak, is fascinating indeed. Also, if you type How old am I into the Google search box, you’ll never get a genuine-like answer, such as the one from the bot. Well, I admit it may even appear scary to some, especially if you’re a fan of sci-fi movies related to the topic of AI. 🙂

One way or another, we should be careful and warn our students about ChatGPT’s limitations. Some argue that the biggest problem with ChatGPT is it often fails at accuracy. It sometimes writes plausible-sounding but incorrect or nonsensical answers. Users have discovered that ChatGPT makes up historical and biographical data. Also, it can generate discriminatory results. So, we teachers need to constantly encourage our students to verify their search results. This, in fact, can be a useful skill to learn and practise anyway. Our students should be aware of the fact that these days, they should not trust anything they come across on the internet unless they check different sources and look at the information from various angles.

All in all, if nothing else, ChatGPT can be a great tool for students to practise asking good questions. I’m deliberately not saying ‘correct’ questions because the bot will provide answers to wrongly formed questions as well, which may not detrimental after all, especially if our aim is to focus on the content and some sort of ‘communication’ instead of grammatical accuracy.

Anyway, below are some of the questions I asked when I first played with ChatGPT and the answers I received. You can see for yourselves how useful or ‘accurate’ they are. 😉

What is the meaning of life?

The meaning of life is a question that has puzzled people for centuries, and there are many different answers to this question depending on one’s beliefs, values, and personal experiences. Some people believe that the meaning of life is to seek happiness, fulfilment, or purpose, while others believe that the meaning of life is subjective and can vary from person to person. Others may believe that the meaning of life is to serve a higher power or to contribute to the greater good of society. Ultimately, the meaning of life is a deeply personal and philosophical question, and the answer will likely be different for everyone.

What is love?

Love is a complex mix of emotions, behaviours, and beliefs associated with strong feelings of affection, protectiveness, warmth, and respect for another person. It can also include deep romantic or sexual attraction. Love can be experienced in many different forms, such as the love between family members, friends, and romantic partners. It can also be described as an action or a choice to care for and support another person. Some people believe that love is an important part of a happy and fulfilling life and that it is a fundamental human need. Others may view love as a mysterious or elusive concept. Ultimately, the meaning and experience of love can vary greatly from person to person.

Is there life after death?

The concept of life after death is a belief that has been held by many different cultures and belief systems throughout history. Some people believe that the soul or spirit lives on after the body dies, while others do not believe in an afterlife. There is no scientific evidence that proves or disproves the existence of an afterlife, and it is ultimately a matter of belief and faith. Beliefs about the afterlife can vary greatly from person to person and can be influenced by cultural, religious, and personal experiences. It is important to respect the beliefs of others and to recognize that people have the right to hold their own beliefs and opinions on this topic.

Does technology make learning and teaching easier?

As a teacher, I’m on a constant lookout for ways of helping my students to improve their level of English, especially at the time when they (think they) are plateauing, which is at around CEFR B1/B2 level. However, this sometimes proves to be quite challenging. The thing is … how can you measurably achieve improvement in language learning?

Let’s take writing, for example. Writing is a skill where the measurability of progress is relatively achievable, especially with all the tech tools available these days. For instance, during the online teaching period, I was able to run my students’ electronic versions of their writings through various text analysing tools, which in the pre-covid times was off the table since they would ususally submit the handwritten versions. Anyway, it turned out that my senior students, regardless of my preconceptions of their writing abilities, had invariably reached the B2+ level, according to this GSE text analyser. So although some students’ writings were clearly better than others, they were all labelled as B2+. This was actually great news because, in my teaching context, the students only aim to achieve level B1/B2, i.e. this level is sufficient for them to pass their state exam in English.

I must admit, though, that the uniform results made me a bit suspicious so for the sake of comparison, I decided to run some other texts through the same profiler; I took an excerpt from my own blog and excerpts from two other blog posts written by native speakers and/or professional bloggers. To make my investigation even more thorough, I looked for samples of writing by examinees aspiring to the C2 level. In addition, I used an academic text. Finally, I chose two pieces of texts written by renowned novelists. To my utter surprise, all the samples were labelled as B2+.

To a layperson, it may seem strange that no matter who the author was, the majority of the words used in those texts were the A1 words. For example, my best student’s essay comprised 57 per cent of words at A1. Compared to the renowned novelist, the ratio was not very different (see below).

My student

A famous novelist

One of the obvious conclusions may be that to come off as a decent writer you just need to throw in lots of A1 words plus a sprinkle of C2 expressions (and some in between these two levels, of course). In this regard, my student did really well and as far as the choice of vocabulary is concerned, there is not much he could do to make his writing look better. Obviously, next time, he could look for synonyms to replace some of the lower-level expressions, plus he could add a few C2 words into the mix, but will this make his essays better? And what does ‘better’ even mean? Does it mean more readable, more complex or more concise? And what do we want to achieve in the first place when we are talking about improving our students’ level of proficiency? Is it our job to spoon feed our students with low-frequency expressions in order to move them up the imaginary ladder? Or is it something else?

You know, the problem with the CEFR scale is that it is linear. Plus, we (and our students) naturally desire to constantly move upwards. A C1 language user is deemed attractive in the eyes of an A2 user, not the other way around. But is a C2 essay better than a B2 one? In the same vein, I could ask if being able to speak at the C1 level is better than being able to hold a conversation at the B2 level. The answer is obvious: it depends on the situation. A lower-level student can make do with what they know and they can convey their message just fine. All things considered, I believe it is not necessarily wise to ceaselessly push our students up towards the highest levels of proficiency. They might get the impression that once they reach the C2 level, they will have achieved the ultimate goal and that’s that. But is it really the terminus? I mean, you can feel stuck on the intermediate plateau for years and still be learning tons of new things. There are loads of language items you can work on incorporating into your language toolkit. And apart from vocabulary and grammar, you can keep refining other language areas and skills.

I do admit my theory has one flaw; I’m only discussing productive skills. Obviously, an L2 student’s productive skills will always be at a lower level than their receptive skills. In other words, in any language, even your mother tongue, you usually understand more than you can actually say/write. So I’m not saying it’s not worth constantly investing your (the student’s) time and energy into enlarging their vocabulary because, to put it simply, knowing more high-level words is useful because you understand more, learn more and can consequently produce more. Based on my experience, from the B2 level onwards, it mainly boils down to the range and amount of vocabulary you know. There aren’t many grammatical structures that will puzzle you or impede your understanding of a text at this level. But not knowing more than 2% of the words in a text you are trying to understand can prove tricky.

So what are some of the ways to help our students navigate the journey? As an L2 learner myself, I find the English Vocabulary Profile Online handy. Personally, I only focus on the C1-C2 words, especially phrases, and in my mind’s eye, I sort them out into three categories: the ones I know and use, the ones I know passively but don’t know properly, and the ones I don’t know. This is a great way to revise and/or fill in the gaps in my knowledge. There’s a similar tool – The English Grammar Profile -which I also sometimes use, but mainly as a teacher. And I advise my students to explore it as well, especially if they need to prepare for an exam because this is a more focused way of studying than, say, watching and listening to random stuff.

Apparently, learning and teaching a foreign language are both equally challenging and complex processes. We all know that. Technology is great but it doesn’t always make things easier. The more advances there are, the more questions emerge – for the teacher as well for the learner (for me anyway). Add to that the plethora of research findings about how languages are best taught and you may end up pretty frustrated, right? Well, let’s take one step at a time. By researching, doing, and reflecting on the doing (in the form of this post, for example), I’ve taken that one little step. We’ll see what the next one will be.

Homophones – pain in the neck?

• 

Recently it has come to my attention that I tend to misspell certain words. Such a discovery may not seem particularly groundbreaking since everybody errs. What does bother me a bit though is that these misspellings often pass unnoticed (by me as well as my spellchecker). I’m specifically talking about homophones, i.e. words having the same pronunciation but different meanings. Although for some reason, it’s unlikely that I will confuse mourning with morning, chances are that I will use brake instead of break without realizing that there’s something wrong with my sentence. I mean, I obviously know the difference between the two expressions but I confuse them nevertheless. Other words I tend to ball up are basic words such as heel vs heal, knew vs. new. Believe it or not, I even caught myself using no instead of know once or twice. And yes, once vs one’s can be tricky too. Well, it seems that the more notorious the word is, the higher probability there is that I will mess things up. Also, short words tend to be trickier since generally, you automatically pay more attention when producing more complex language. This implies (to me) that as I write, I actually hear the words in my head. Funnily enough, once I’m using a more complex expression, which I’m not 100% sure how to pronounce, I hear myself spelling it in my mind (the Czech way though).

Anyway, I’ve recently learned that as far as homophones are concerned, a difference in spelling doesn’t always indicate a difference of origin. As a rule of thumb, dictionaries treat homophones as different words simply because they are spelt differently. So a traditional dictionary will not give you a clue as to whether the words are historically of the same origin. But if you dig deep enough, you’ll find out that for example flower and flour have much more in common than you would expect. So, a crazy question occurred to me: is this type of ancient knowledge somehow ingrained in our brains? Well, my hypothesis is a bit flawed, at least in my case, because I’m not a native speaker of English. But still, maybe one of my genes was inherited from someone whose mother tongue was English indeed. Shakespeare maybe? Who knows? One thing is certain, language and brains are amazing entities. At the same time, I think it’s not really surprising that the brain, having to constantly make millions of decisions at every point of our lives, occasionally chooses the wrong option out of the two available in its inventory – and opts for no instead of know. It’s not a tragedy after all; unless this misstep influences our future in some way, everything is fine (apart from the fact that we made idiots of ourselves).

But here’s the thing. While I sometimes err when it comes to homophones, my students don’t as often as one might suppose. They throw around all sorts of other spelling mistakes, particularly typos are their favourites, and they like to coin new words too. But homophones? No, that’s not a big problem. There’s this idea at the back of my mind, I must have heard it somewhere, so correct me if I’m wrong, that native speakers tend to make homophone errors more often than L2 learners do. So my hypothesis is (and maybe somebody out there has already tested this) that the more frequently you are exposed to a language, the better you get at it but at the same time, you become more susceptible to committing a homophone error in writing. It seems that when your level of L2 is not high enough, which is the case of some of my students, your brain really needs to focus on in what is happening and is less prone to making careless mistakes of this sort. I mean, when producing and essay, my students probably think twice before engraving their words in stone (at least in the ideal world scenario), so these slips will not happen as often. They simply want to get things right and thus play it safe. So confusing weak with week is most likely off the table because they are familiar with both words but don’t use them too automatically yet. On the other hand, they might not even ‘consider’ confusing words such as wright vs right, simply because they are NOT familiar with the former. However, when they have enough knowledge, they might do so as a result of trying to use a rule in a context where it does not belong (overgeneralization error).

To conclude on a happier note, homophones are not just a pain in the neck. They can be fun since they are used to create puns, which is a feature I like to use in my lessons.

What about you and homophones?

Fighting my personal biases

The debates about discriminatory hiring practices in ELT are giving me sleepless nights. Although I’m trying hard to be a good girl and stay on the right side of the barricade, I can’t pretend there are no fleeting moments of hesitation and doubt. I mean, I strongly admire all those educators who publicly stand up for the rights of non-native teachers of English. As a NNEST myself, I am happy to see that brave people all over the word are fighting for my rights. Thanks for your bravery; I honestly appreciate it.

My problem is that most of my formal education took place at the time when such a debate was absolutely unthinkable. What? NNESTs can be as good as NESTs? Are you kidding me? We used to look up to them and what they said was taken as the ultimate truth. Ironically, later on at university, right after the fall of communism, when there was a boom of ‘backpackers’ from the west, we students preferred NNESTs – probably because they seemed more organized in what they did and because they could teach us the rules of the language. Lessons with NESTs were generally fun and truly beneficial acquisition-wise, but they were utter and complete chaos (with some exceptions, of course). But still, whenever there was a problem, we were told to go and ask a native speaker.

Anyway, the debate which is going on these days is intriguing. Sometimes, though, to my utter consternation, I catch myself not nodding all along the way. Throughout my career, I’ve taught very young kids, as well as teenagers and adults. I also became a student at the tertiary level again for a while back in 2011. I apologize for my impudent generalization, but judging by what I’ve observed so far, I can’t say that here in the Czech Republic we are ready to claim that qualified NESTs are as good as qualified NNESTs. Based on my random observations, in other countries the situation is slightly different – they have a much longer tradition of learning languages, their mother tongue is somehow related to English and thus they acquire it faster or easier, they have been able to travel more, etc. It’s getting better here, it surely is, and I believe there are loads of teachers who are already exceptions to the rule, but generally speaking, there’s an awful long way for us to go before we’re able to join the crowd of confident NNESTs fighting boldly for their rights in the ELT business. We need to remain humble and work hard rather than ‘join the demonstration’ just because everybody else has. If you are from the Czech Republic and you feel I’m being biased and unfair, I’m sorry but that’s how I see it….

The debate about discriminatory hiring practices in ELT I’m following with great interest constantly makes me ask hypothetical questions: What if there are two equally qualified teachers, with the same amount of experience, applying for the job of an English teacher at a Czech school – one NEST and one NNEST? What criteria come into play in such a situation? I’m sure that the employer will probably have to consider other factors, sometimes equally discriminatory, without explicitly saying so, such as the applicant’s ability to speak the students’ L1, her pretty face, his congenial manners, or the fact that one of the applicants is a young female about to start a family.

Or what if a school really likes to have a mixture of NESTs and NNESTs, which is perfectly justifiable in our teaching environment, and as they already have six NNESTs and no NESTs, they desperately need to hire one NEST. Is this discriminatory?

In one of my previous posts, I had a fruitful discussion with Vedrana Vojkovic, touching on the issue of NESTs vs. NNESTs. When re-reading the comment section, I realized I had sounded pretty biased. Unfortunately, I can’t change my view yet. The debate revolved around teaching English to very little kids, precisely those at kindergartens (which, by the way, wasn’t Vedrana’s original intention, but I stubbornly stuck to the topic anyway). As I see it, the more proficient the teacher, the higher level they usually want to/are asked to teach. So it goes without saying that those who have achieved a native-like proficiency are not likely to end up teaching English in kindergarten, unless they really love small kids. They will become teachers at the tertiary level or do something completely different. Yet, quite a few pre-school institutions offer optional English lessons. These are usually taught by someone, anyone, who can speak some English. These teachers are either fully qualified kindergarten teachers, who are, however, not qualified to teach English, or students who need some pocket money. One way or the other, it seems to be a general consensus that after all, you don’t need to be terribly proficient if you want to teach little kids.

One thing is certain; as far as I know, there are no teacher training programs for language teachers working at a pre-school level. Taking into account L1 acquisition principles, I’m convinced that if you want to teach very young learners, you need special training, very different from the one we normally get as teachers aiming at the primary/secondary/tertiary level. And I’m not only talking about methods but also about one’s language proficiency. This lead me to a conclusion that a chatter with a NEST might be more valuable at this level than a lesson with an unqualified teacher of English, who teaches a few random words a day. Needless to say, little kids acquire these words precisely the way they hear them, i.e. out of context, sometimes with totally wrong or imprecise pronunciations, which later on hinder understanding and communication. I was once told by an owner of a language school: “Just go there and do something. It doesn’t really matter what you’ll do, does it?”

These were some of the random thoughts that are swirling in my head these days. If you happen to have sensed some kind of bias in my voice, I’d like to make it clear that the way I reason stems from my life experience. Also, I realize that some of my convictions may appear as mere generalizations. However, I’m not saying that what I claim here is right or wrong. It’s just the way I see it now.

Some of my nostalgic (linguistic) memories of the Netherlands

I’m finally back home from a short visit to a lovely Dutch town called Valkenswaard. My heart still aches a bit since I’m missing all the friendly people I met there – the students and teachers from six European countries that had got together to work on a music/poetry project. But I know the memories will soon fade and life will return to normal. Well, not quite, I’m afraid…. Things will never be the same.
————————————————

As a Czech tourist, when you land in Eindhoven, you immediately notice a few things. The architecture is slightly different from what you can see in a typical Czech town. The lovely traffic lights that look like children’s toys make you feel you’ve just come to see Legoland. But the main difference can’t be perceived visually – it is when you open your mouth to speak and prick up your ears to listen that you finally realize you are in the Netherlands; everybody speaks English there. Every bus driver, every shop assistant, and every waitress will reply fluently once you start asking them questions in English.

This is something you will hardly experience in the Czech Republic. In an extreme situation, once they are approached by a foreigner, people will even run away or pretend they don’t speak English. The reason is simple – generally, Czechs are not very confident in English.

So while in the Netherlands, I asked myself (and other people too) the same question over and over again: How come Dutch people are so proficient in English? I always got the same reply: we don’t dub English programs and thus we’re exposed to heaps of English from a very early age.

But I think there is another reason behind their high English proficiency. Dutch is a Germanic language and it is closely related to English and German. Dutch shares with German a similar word order, grammatical gender, and largely Germanic vocabulary, which contains the same Germanic core as German and English. Nonetheless, the fact that Russian is a Slavic language closely related to Czech didn’t help me achieve a native-like proficiency in it when I was forced to learn it back during the communist regime. Apparently, one ingredient vital for a successful acquisition of an L2 was missing – motivation.

Now, considering the fact that the Netherlands has a tradition of learning languages and almost 90% of the population can easily converse in English, it’s obvious that the L2 proficiency of their English teachers reflects the situation. I met a Dutch (as well as a German and a Belgian) teacher of English, whose L2 proficiency was absolutely stunning. Had I not known what their nationalities were, I wouldn’t have guessed they were non-native speakers of English. The NNEST vs. NEST dichotomy suddenly seemed useless and redundant. If I had ever doubted that non-native speakers of English can achieve native-like proficiency, this was the final proof that they can.

But I also met a German teacher of geography and a Belgian music teacher whose fluency in spoken English (and several other languages) was equally astounding. I remember a few occasions in the past when my English had been described as flawless but honestly, now I think people were only trying to be nice to me; most of the time in the Netherlands I felt humbled. In spite of this, I’m immensely thankful for this experience.

If only I could spend more time at the school – observe lessons, talk to the teachers, students, and other members of the staff. I would like to get under the surface and find out if their approaches to learning and methods of teaching English are very different from what we do here. I’d like to interview more people in the streets and pubs; I’d love to ask about their motivation and general attitudes to foreign languages….

Stressing out about stress

I can’t remember how many times I’ve told my students that stress – the emphasis placed on the sound or syllable spoken most forcefully in a word or phrase – is a very important aspect of spoken English. I tell them that although this linguistic feature may seem trivial to native speakers of Czech, it can be a matter of communicative survival in English. The trouble is that Czech has a fixed stress, meaning that its position can be predicted by a simple rule, i.e. it almost always comes on the first syllable. It’s not a big issue if you place the stress elsewhere – you will likely be understood, provided you get other aspects of pronunciation right.

My students often struggle with sentence stress – the stress placed on words within sentences – and I wrote about ways of handling it here. They also find it difficult to deal with lexical stress – the stress placed on syllables within words. There are two notorious words I repeatedly correct – hotel and event. It doesn’t matter how many times I model the pronunciation; in most cases my students will get it wrong the next time again. There are obvious reasons for this: as already mentioned above, it’s natural for my students to speak stressing the first syllables in words. Moreover, despite the fact that in written Czech the word for hotel is identical to its English counterpart, we pronounce it slightly differently, i.e. we place the stress on the first syllable.

Now, my students are not the only ones who sometimes struggle with this aspect of spoken English. I remember at least two occasions when my message seemed totally unintelligible to my Australian friend, just because I placed the word stress incorrectly. For example, I remember that my friend looked really puzzled when I told him about the problem with mosquitoes. I pronounced it [ˈmɒskitəʊs] instead of [məˈskiːtəʊs]. I had to repeat the word several times and even describe the insect before my friend got the meaning. I was pretty frustrated because to my Czech ear, the difference is not so dramatic, and if I heard the word pronounced in different ways, I think I would always understand. By the way, this is one of the dangers of monolingual classes taught by a teacher speaking the same L1 – we understand one another and easily ignore things that seem unimportant to us. 

Another communication breakdown happened when I used the word teetotaller. I said [ˈtiːtəʊtlə] instead of [tiːˈtəʊtlə]. Neither repeating the word nor raising a glass of beer helped my friend to get the meaning. I had to spell the word (which got me into even more trouble, as you can imagine)! I know that this isn’t a very frequent word but this situation clearly demonstrates what an important role word stress can play.  

I’m really happy I experienced those two communicative failures since I can share these stories with my students; I can show them what pitfalls there are waiting outside the safe L2 classroom.