Studying in a second language, past the certificate

Language tests are built for reliability, which means controlled conditions. The audio is clear, the speakers are deliberate, the topics sit inside a known range, and every task has a visible shape with a defined end. That is what makes a score comparable between candidates. It is also what makes it a poor forecast of a lecture theatre, where two people talk at once, a reference everyone else grew up with goes past unexplained, and the speaker abandons half of their own sentences.

The other missing ingredient is load. In a test, the language is the task. In a seminar, the language is the medium and something else entirely is the task: following an argument, forming a view about it, and saying that view in a way that survives being questioned. Doing all of that at once is a different skill from doing any part of it separately, and it returns more slowly than reading does. The difference is not evidence that your level was overstated.

Separating the strands helps more than treating a single verdict about your language as the whole story. Reading, listening under time pressure, writing to an academic convention and speaking in a group are distinct capacities that improve at different rates. Working out which one is causing trouble on a given day makes it far easier to act on, and it usually turns out that most of them are perfectly serviceable.

Two registers tend to be missing after even a strong language course, and neither is a matter of general fluency. The first is the discourse layer of academic speech: the words that signal what is happening to an argument rather than adding content to it. The second is administrative vocabulary, the language of forms, portals, deadlines and appeals, which is rarely taught anywhere because everyone assumes it gets picked up along the way.

The efficient response is a running list drawn from your own programme rather than a general vocabulary book. Collect the words that actually appear in your lectures, your reading and your institution's messages, along with the phrase each one arrived in. Words met in context and reviewed in context stay put more readily, and a list built from your own materials stays short enough to be usable, which is more than can be said for most vocabulary apps.

Language courses generally teach a standardised variety, spoken clearly at a moderate pace. Universities run on something else: contracted forms, local idiom, regional pronunciation, and the particular speed of somebody who has delivered the same lecture many times. Staff and students are drawn from many countries as well, so the range of accents in a single room is often wider than any course prepares for. No variety is a degraded form of another. They are all the language, and each takes its own period of exposure.

Familiarity does most of the work here, which is encouraging, because familiarity accumulates without effort. The same lecturer becomes easier to follow through repetition alone. Where recordings are provided and their use is permitted, revisiting a difficult passage at a slower speed is one of the few genuinely reliable fixes available. Reading before a session helps for a different reason: once the content is already known, an unfamiliar sound has something to attach itself to.

Groups are harder than individuals, and social settings are harder than lectures, because nobody is projecting and everybody is interrupting. That ordering surprises people who assumed the academic content would be the difficult part. Knowing it in advance is worth something, since it stops a hard evening in a crowded room from reading as a sudden loss of ability.

Humour is usually the last thing to come back, and its absence is felt more sharply than any vocabulary gap. Being funny depends on timing, on shared reference, and on knowing precisely how far a register can be bent, which is exactly the knowledge a new language does not include. People who are good company at home can feel flattened abroad, present in a conversation without being recognisably themselves in it.

Irony compounds the problem, because a statement whose literal meaning is the opposite of its intent stays invisible until you know the local signals for it. Understatement behaves the same way in places where it is common. Asking whether something was meant seriously is a perfectly normal question, and it tends to produce an explanation and a bit of warmth rather than an awkward silence.

Small talk deserves less anxiety than it usually gets, because it is a protocol rather than a conversation. The content carries almost no information. The function is to signal availability and goodwill, and the openers are formulaic in every language, which means they can be learned as a set. That is a fast way to sound at ease some time before you feel it. More on the social side of this sits in the guide to making friends at university.

One pattern shows up repeatedly among students working in a second language: they speak considerably less than they are able to. Output is public and error is audible, while comprehension stays private. The safest move in any given moment is to stay quiet, and each quiet moment makes the next one likelier. Over a term, other people build an impression from the small sample they have heard, and it becomes surprisingly easy to start accepting that impression as accurate.

The cost is real where participation is assessed, and it is real socially, since groups tend to form around whoever speaks first. The way out is not a sudden decision to become confident. It is a series of small, low-stakes commitments: one prepared contribution per seminar, one question asked in an office hour, one comment in a group chat where editing before sending is possible. Confidence arrives after the evidence rather than before it, and the guide on building confidence as a student covers the wider version of the same problem.

Most of the useful adjustments move effort away from the moment of pressure, to before or after it. Understanding in real time is expensive, recognition is cheap, and preparation is what converts one into the other. The moves below are unglamorous and they transfer across most systems, though it is worth checking what your own institution permits, particularly around recording sessions and sharing materials.

For administrative matters specifically, written channels are usually kinder than spoken ones. Email gives you time to compose and produces a record that can be reread when the pressure has passed. Where a conversation has to happen in person, taking a friend along is unremarkable, and asking whether the office can confirm the outcome in writing is a routine request rather than a sign of distrust.

One skill is worth practising deliberately: asking for something to be repeated without disappearing into apology. Precision helps more than politeness here. Naming the exact point where you lost the thread gives the other person something to work with, and asking for a rephrasing rather than a repetition is usually more effective, since the same sentence delivered again at the same speed rarely lands any differently.

Three phrasings cover most situations: asking somebody to say the last part again, asking what a particular word means in this context, and repeating back your understanding for confirmation. First-language speakers use all three constantly, because they mishear things too. The awkwardness people brace for is largely imagined, and the cost of a misunderstanding discovered later is far higher than the cost of a small interruption now.

It is worth being accurate about what all this represents. Following a technical argument in a second language, negotiating with an office in it, and switching between codes on the way from a lecture to a call home is demanding work that people operating only in one language have never had to do. It comes with side effects: close attention to how meaning gets built, tolerance for ambiguity, and real skill at working with people who do not share a first language. The wider settling-in process is covered in international student adaptation. Where the difficulty is about conventions rather than words, adapting to a new academic culture takes it further.