Four parts, thirty questions, about forty minutes. In the real test you hear everything twice and nothing more. Here you also have the transcript — but do not open it until you have answered, because reading the answer teaches you nothing about hearing it.
You will hear three different extracts. For each question, choose the answer (A, B or C) which fits best according to what you hear. There are two questions for each extract. You hear each extract twice.
You will hear three different extracts. For questions 1 to 6, choose the answer (A, B or C) which fits best according to what you hear. There are two questions for each extract.
Extract One. You hear two architects discussing a railway station that has recently been rebuilt. Now look at questions 1 and 2.
Daniel: I keep being asked whether I like it, and I keep failing to answer. The roof is a genuine piece of engineering. That much is beyond argument.
Priya: It is. Though nobody standing underneath it is looking up.
Daniel: No. And that's the thing I can't get past. Everything above head height is superb and everything below it was value-engineered in a hurry. You come off the train into a very expensive shed and then you're funnelled through eleven metres of retail before you reach daylight.
Priya: That's the brief, though, not the architect. The retail paid for the roof.
Daniel: I know. I'd just rather we said so. There's a lot of writing about that building at the moment that describes the shops as a civic amenity.
Priya: Ah. Yes. That I won't defend.
Daniel: The odd thing is that the passengers seem perfectly happy.
Priya: They are, and I don't think that's a small point. Ninety thousand people a day are dry, warm and not late. Twenty years ago they were none of those things. I'm suspicious of any judgement that treats that as beneath notice.
Daniel: You think I'm being precious.
Priya: I think you're right about the ceiling height and wrong about what it costs people.
Extract Two. You hear two colleagues talking about a training course they have both attended. Now look at questions 3 and 4.
Ruth: So. Three days. Be honest.
Callum: Honestly? The content was thin and I'd have got most of it from an hour with the handbook.
Ruth: Agreed. But that isn't why I'd go again.
Callum: Go on.
Ruth: There were nineteen of us from eleven departments, and by Wednesday lunchtime I had a straight answer from procurement about something I've been emailing about since February. Not because the course covered it. Because Aisha from procurement happened to be sitting opposite me eating a sandwich.
Callum: So the course is a pretext.
Ruth: The course is a room. Nobody will authorise three days for a room, so it has to be called a course.
Callum: That's either very cynical or very practical.
Ruth: It's the only argument I've ever heard for these things that survives contact with the evidence. The evaluation forms ask whether you learned anything. They should ask who you met.
Callum: I'd still rather they made the content better.
Ruth: Of course. I just wouldn't measure the course by it.
Extract Three. You hear two friends talking about one of them giving up a permanent job to work freelance. Now look at questions 5 and 6.
Owen: Six months. How is it?
Zeynep: Better than I said it would be and worse than I hoped. The money's fine — that's the thing nobody believes.
Owen: I don't believe it.
Zeynep: It's genuinely fine. I earn more per hour and I work fewer of them. What I underestimated was the admin, and everybody warned me about the admin, so I've no excuse there.
Owen: So what's the bit nobody warned you about?
Zeynep: That I'd stop having opinions. In an office you're asked what you think about things you have no stake in, constantly, and it's irritating, and it turns out it was doing something. Six months in, I notice I don't finish thoughts any more. There's nobody to finish them for.
Owen: Could you not just — meet people?
Zeynep: I do meet people. It isn't the same as being argued with by somebody who has to sit next to you afterwards.
Owen: Would you go back?
Zeynep: No. But I've started paying for a desk in a shared office, which is, if you think about it, paying to be irritated.
That is the end of Part One. Now listen again.
How to read this part. Part 1 gives you two questions and about ninety seconds of speech, and the answer to each one sits in a single line. Everything else in the extract is there to make that line hard to find. The two questions are almost never about the same speaker, so read both before the extract starts and know which voice you are waiting for.
Do it in this order.
The traps here are not vocabulary. They are (a) a word from a wrong option said clearly and early — admin, the roof — and (b) a question about one speaker answered with the other speaker's opinion. If your reason for choosing is “I definitely heard that word”, you have been caught.
On each question. The help for the ones you missed is now attached to the question itself — the options that cannot be right are crossed out, and the place to look is marked.
You will hear a woman called Nila Ferreira giving a talk about the year she spent working as a lock keeper on a British canal. Complete the sentences with a word or short phrase. Write the words you actually hear — this part does not accept a paraphrase.
one or two words
one word
one word
one word
two words
one word
one word
one word
You will hear a woman called Nila Ferreira giving a talk about the year she spent working as a lock keeper on a British canal. For questions 7 to 14, complete the sentences with a word or short phrase. You now have forty-five seconds to look at Part Two.
Nila: Thank you. I'm going to talk about the strangest and best year of my working life, which I spent as a lock keeper on the Trent and Mersey Canal.
Nila: I should say at the start that I had no relevant experience whatsoever. For nine years I'd been a software tester, which is a job I was reasonably good at and had stopped thinking about, and I applied for the lock keeper's post on the last day of the advertisement, largely because my sister dared me to.
Nila: The canal I ended up on was cut in the 1770s, and although people assume it was built for coal, it wasn't — it was built for salt. The Cheshire salt fields are the entire reason it exists, and you can still see the subsidence in the ground beside it where the brine was pumped out.
Nila: The first thing that astonished me was the mechanism. I had assumed there were pumps somewhere, or a motor. There is nothing. A lock works entirely by gravity. You open a paddle at one end, water moves downhill, and a boat weighing thirty tonnes rises eight feet because water found a lower place to be. Two hundred and fifty years old, and not one moving part that a child couldn't operate.
Nila: What took me longest to learn was not the machinery. It was the weather. A canal is a very shallow, very long puddle, and everything about it — the level, the flow, whether you can let a boat through at four o'clock — depends on what fell forty miles upstream two days ago. I got that wrong all summer and had it more or less right by the following spring.
Nila: People ask about damage, and the answer is never what visitors expect. It isn't the working boats and it isn't the old craft, which are handled by people who own them. It's the hire boats. A hire boat is steered by somebody on the second afternoon of their holiday, and it meets the lock gates at three miles an hour, which is quite fast enough to break something that was made in 1774.
Nila: The word I'd use for the pace of the whole thing is deliberate. Not slow, exactly — deliberate. Everything takes the time it takes, and no part of it can be hurried by wanting it to go faster, which took me about four months to stop resenting.
Nila: Visitors asked me a great many questions, and by a very long way the object they asked about most was the windlass — the L-shaped handle you use to wind a paddle up. Everybody wants to know what it's called and everybody has forgotten by the next lock.
Nila: I'm not going back to software. From September I'm training as a stonemason, because the walls of these locks are failing and there are fewer than two hundred people in this country who can rebuild them properly. It's the first time in my life I've chosen work because of what would be lost if nobody did it.
That is the end of Part Two. Now listen again.
How to read this part. Part 2 gives you the answer in the speaker's own words, in the order of the sentences on the page, and it never asks you to work anything out. What it does do is say the wrong word first — loudly, clearly, and often twice — and then correct it. The answer is almost always the word after the correction.
Do it in this order.
Two things lose marks here that have nothing to do with listening. Writing a synonym of the word you heard, and writing too much — the whole phrase instead of the two words the gap wants. Say the sentence in your head with your answer in it before you move on.
On each question. The help for the ones you missed is now attached to the question itself — the options that cannot be right are crossed out, and the place to look is marked.
You will hear an interview with Dr Frances Okonjo, a lexicographer, who is talking about how a dictionary decides which new words to include. For each question, choose the answer (A, B, C or D) which fits best according to what you hear.
You will hear an interview with Dr Frances Okonjo, a lexicographer, who is talking about how a dictionary decides which new words to include. For questions 15 to 20, choose the answer (A, B, C or D) which fits best according to what you hear. You now have one minute to look at Part Three.
Tom: Dr Okonjo, when people find out what you do for a living, what do they get wrong?
Frances: They think we decide. There's a persistent idea that somewhere in a building in Oxford a committee sits and votes a word in, and that if we voted differently the word wouldn't be English. It's the reverse. We're not a gate. We're a record. By the time a word reaches us it has already been admitted by several million people who never consulted us about anything.
Tom: But you do decide what goes in the book.
Frances: We decide what has met the evidence. And the evidence is duller than people hope. We want a word to have been used, in print or in something we can point to, by people who don't know each other, across at least five years, in more than one kind of writing. Not five years of newspapers all quoting the same tweet. Five years of independent use. Most candidates fail on that last part — they're everywhere for eighteen months and then they're nowhere, and we would have looked foolish.
Tom: Do you ever get it wrong?
Frances: Constantly, and nearly always in one direction. We're too slow far more often than we're too quick. And there's a lovely trap that I use to keep myself honest, which is the word “cromulent”. It was invented for a joke on a cartoon in 1996 — the joke being that it wasn't a real word. It is now used, entirely unironically, by people who have never seen the programme, to mean “acceptable”. If you'd asked me in 1996 I'd have said it would be gone by Christmas. It's in the dictionary.
Tom: You must get letters.
Frances: Sacks of them. And I read them, and I want to be careful here, because the tone people use about this is often unpleasant and the underlying instinct is not. Somebody who writes in to complain that we have included a word they dislike is telling me that the language matters to them enough to spend a stamp on it. That's not nothing. What I can't accept is the premise — that including a word is approving of it. We include “genocide”. Nobody imagines we're in favour.
Tom: What has changed most in the ten years you've been doing this?
Frances: Not the speed, which is what everyone assumes. What's changed is that we can now see a word being born. Twenty years ago a word arrived in our reading programme already formed and we had to guess where it had come from. Now we can watch it appear in one small community, mean one thing for a while, and then leak sideways into general use and change its meaning on the way across. We've stopped writing “origin unknown” quite so often. It's the most interesting change in the field, and it has nothing to do with volume.
Tom: And taking a word out?
Frances: That's the one that keeps me awake. Adding a word costs a line. Removing one is a claim that nobody needs it any more, and we are always the last people to know that. In a print dictionary you had to remove things because paper runs out. Online you don't, and I've come to think that isn't a small technical detail. It's a change in what a dictionary is. We used to be a selection. We're becoming an archive, and an archive doesn't throw things away.
Tom: Dr Okonjo, thank you.
That is the end of Part Three. Now listen again.
How to read this part. Part 3 is the longest single stretch of speech in the paper and the six questions run in the order she answers them, one per exchange. So the interviewer's questions are your map: when he asks a new question, you have moved on to the next item, whether or not you got the last one.
Do it in this order.
The distractors here are true statements from the wrong part of the interview, and opinions the interviewer puts to her that she then modifies. If an option matches something you heard but you cannot place who said it or when, treat that as a warning rather than as confirmation.
On each question. The help for the ones you missed is now attached to the question itself — the options that cannot be right are crossed out, and the place to look is marked.
You will hear five short extracts in which people talk about changing career. You must complete both tasks as you listen. Task One — questions 21–25: choose from list A–H what first prompted each speaker's decision. Task Two — questions 26–30: choose from a different list A–H what each speaker says they miss about their previous work. Both answers for a speaker come out of the same forty seconds of speech, so the two lists are on the page at the same time throughout.
You will hear five short extracts in which people talk about changing career. Look at Task One. For questions 21 to 25, choose from the list A–H what first prompted each speaker's decision. Now look at Task Two. For questions 26 to 30, choose from the list A–H what each speaker says they miss about their previous work. While you listen you must complete both tasks. You now have forty-five seconds to look at Part Four.
Speaker One: I'd been a structural engineer for nineteen years and I'd never once considered doing anything else. Then I came off my bike and spent eleven weeks on a sofa, and by week four I'd stopped thinking about bridges and started thinking about why I hadn't thought about anything except bridges since 1998. My department wasn't closing, nobody had passed me over for anything, there was no crisis. I simply had the time. What I miss — and I hadn't expected this — is walking into a room where everybody assumed I knew the answer. Nobody assumes that about me now.
Speaker Two: It was Geoff's leaving do that did it, which is embarrassing to admit. He stood up and talked for twenty minutes about forty-one years, and I realised that if I stayed I could write my own version of that speech already, and it wouldn't need updating. I went back on the Monday and started applying. I don't miss the work at all. What I miss, absurdly, is the shape of the day — the same train, the same fifteen minutes at the same window before anybody arrived. Freelance days have no edges.
Speaker Three: My daughter asked me what I actually did, and I gave her the sentence I give at parties, and she said “no, but what do you do”, and I couldn't tell her. I couldn't describe a single thing I had made or mended. That was in the February and I'd handed in my notice by the June, which was reckless, and I would do it again. The money is the thing, though. I'd stopped noticing what a salary was until I didn't have one — knowing exactly what arrives on the 28th of every month. That, I miss.
Speaker Four: I was in a taxi coming in from the airport, and the driver asked me if I liked my job, and I said yes, and he said — not unkindly — “you didn't say that like someone who did”. We drove another twenty minutes and I didn't say anything else at all. I've never seen him again and I think about him weekly. What I miss is that nobody ever asked me to explain it. You said “I'm a radiographer” and that was the end of the conversation. Now I say what I do and there are always four more questions.
Speaker Five: A friend of a friend at somebody's fortieth said, entirely in passing, that her firm was looking for people who could do exactly the thing I did all day and hated. Six weeks later I was doing it somewhere else for a great deal more money, which is not a noble story but it is a true one. I was very happy about the move for about a year. Then I noticed that I'd stopped saying “we”. Twenty-two people I'd have gone to war with, and now I'm perfectly well paid and entirely on my own.
That is the end of Part Four. Now listen again.
How to read this part. Part 4 asks two questions about the same forty seconds, and candidates lose it by trying to answer both at once. Do not. Both lists are on the page throughout, and the two answers for a speaker are almost always in different halves of what they say — Task One near the start, Task Two after the words “what I miss”.
Do it in this order.
Every list of eight contains two or three options that are mentioned and rejected by a speaker — “my department wasn't closing, nobody had passed me over”. Hearing an option said aloud is the single commonest reason for choosing it, and in this part it is very often the reason it is wrong.
On each question. The help for the ones you missed is now attached to the question itself — the options that cannot be right are crossed out, and the place to look is marked.
The marks below are your best unaided attempt on each part — the one you did after the support closed, not the one you did with it open.
If a part only worked once the transcript was open, the problem is not vocabulary; it is that the answer arrives in a different form of words from the question. That is Part 3 and Part 4 in a sentence, and it is fixed by listening again to the same recording, not by finding a new one.