10 min read

C2 Proficiency Speaking Part 2: Your Reaction, Not the Picture

At C2 the pictures move into the collaborative task and the question asks for your reaction to them. Describing what you see answers the wrong question.

At B2 First and C1 Advanced, the pictures are the long turn. At C2 Proficiency they move into the collaborative task, and the question changes with them.

Cambridge says the opening question "focuses on your reaction to aspects of one or more pictures".

Reaction, not description. That single word is the difference between answering the task and answering a task that was not set.

SpeakShark is free for practising that kind of response, three AI conversation sessions a day with no card, each capped at five minutes and four turns.

In this guide: what Part 2 actually is · two stages, four minutes · reaction, not description · how many pictures · sustaining an interaction · the decision task · what we could not verify · how we researched this · FAQ

Key takeaways

  • Part 2 is four minutes, with a one minute reaction question first.
  • The question asks for your reaction to aspects of the pictures, not a description.
  • Cambridge says one or more pictures, without fixing the number.
  • The skills list opens with sustaining an interaction, a phrase absent lower down.
  • The whole test is 16 minutes per pair, or 24 per group of three.

What Part 2 actually is

Cambridge's description:

The interlocutor gives you some spoken instructions and one or more pictures to look at. First, you have to answer a question which focuses on your reaction to aspects of one or more pictures (1 minute). The second part is a decision-making task which you have to do with the other candidate.

Two stages inside one part, and they ask for opposite things. The first is solo and reflective. The second is shared and convergent.

Two stages, four minutes

Stage What it asks Time Who with
Reaction question Your response to aspects of the pictures 1 minute On your own
Decision-making task Reach a decision together The remainder The other candidate

Set that against the whole test:

Part What it is Time
1, Interview Questions addressed to each candidate in turn 2 minutes
2, Collaborative task A reaction question on pictures, then a decision task 4 minutes
3, Long turn and discussion Card, two minute solo turn, then discussion 10 minutes

16 minutes per pair, or 24 minutes per group of three. Part 2 is the middle of the test in every sense: a quarter of the clock, and the bridge between talking alone and talking at length with somebody else.

Reaction, not description

This is the sentence to hold onto: the question "focuses on your reaction to aspects of one or more pictures".

Three words matter.

Reaction. What the images make you think, not what they contain. A view, a response, an argument they provoke.

Aspects. Not the whole picture. Something in it, selected by the question.

Your. The content is meant to come from you. The pictures are a stimulus, not a subject.

The failure mode is inherited from lower levels, where description is legitimately part of the task. At B2 First, Cambridge lists describing as one of three skills for the long turn. At C1 Advanced, it lists describing as one of four. At C2, in this part, describing is not on the list at all.

Level Picture task Is describing on the skills list
B2 First Long turn, a pair of photographs Yes
C1 Advanced Long turn, three pictures, talk about two Yes
C2 Proficiency Collaborative task, reaction question No

That is a real change, not a nuance, and it is invisible unless the three lists are read side by side. The B2 First long turn and the C1 Advanced long turn show what the earlier versions ask instead.

A minute of reaction has a shape that fills it reliably:

  1. Answer the question in one sentence. A view, stated.
  2. Say why, tying the reason to something in the images.
  3. Take it somewhere. A consequence, a comparison, a case where it would not hold.
  4. Close the loop. Return to the view, adjusted.

None of those steps require naming what is visible, which is exactly the point.

How many pictures, Cambridge does not say

At B2 First it is a pair of photographs. At C1 Advanced it is three pictures, of which you talk about two. At C2 Proficiency Cambridge writes one or more pictures, twice, and leaves it there.

That openness is worth preparing for rather than resolving. A task written to work with one picture or with several is a task where the count is not the point, and a candidate whose strategy depends on comparing two images has built something that may not apply on the day.

The reaction question works regardless of how many images appear, which is presumably why it is written that way.

Sustaining an interaction

Cambridge's skills list for this part:

Sustaining an interaction: exchanging ideas, expressing and justifying opinions, agreeing and / or disagreeing, suggesting, speculating, evaluating, reaching a decision through negotiation, etc.

The list after the colon is the same one Cambridge gives for the collaborative tasks at B2 First and C1 Advanced. What is new is the heading: sustaining an interaction.

It frames the part as endurance rather than exchange. Anyone can produce one good contribution. Keeping a conversation productive, without it collapsing into agreement or drying up, is a different demand, and it is the one the phrase names.

The Council of Europe's C2 descriptor points the same way: a C2 user "can express him/herself spontaneously, very fluently and precisely, differentiating finer shades of meaning even in more complex situations". Finer shades of meaning is what a sustained interaction is made of, and it is what "I agree" removes. CEFR levels explained for speaking walks the rest of the ladder.

The decision task

The second stage is a decision-making task done with the other candidate.

The same principle applies as at every level below: reaching a decision quickly is not the skill on display. Cambridge's list ends with "reaching a decision through negotiation", and negotiation is the assessable part. A pair who agree in fifteen seconds have completed the task and demonstrated almost none of the list.

What helps, and none of it can be memorised:

  • Do not accept the first suggestion outright, even when you agree. Say why, then say what the alternative offers.
  • Disagree when you disagree. It is on Cambridge's list.
  • Invite the other candidate in if they have gone quiet, and come back in politely if they have not.
  • Close deliberately when the time is nearly up, rather than being closed by the examiner.

Practising against something that responds is the only realistic rehearsal. Role play scenarios for speaking practice, finding an online speaking partner and a free daily session between them cover it. For the underlying habit, twelve daily speaking drills from B1 to C1, a 30 day plan to improve speaking and speaking exercises for adult learners all fit around work. If clarity slips under concentration, how to improve English pronunciation helps, and if the block is nerves, why your English speaking is not improving and online methods to build confidence in English address it.

Next in the test is Part 3, which takes ten of the sixteen minutes. Coming up from below, B2 First versus C1 Advanced speaking and the C1 Advanced format cover the earlier rungs.

Two details about the room, both from Cambridge: there are two examiners, one who talks to you and one who listens, and one of them could be examining remotely. Examiners may also enter marks using a mobile phone app.

What we could not verify

We have not reproduced the assessment criteria. Cambridge publishes its own scales.

We could not find a published split of the four minutes beyond the one minute reaction question. How long the decision task runs is not stated.

We could not find the number of pictures. Cambridge says one or more and we have not gone further.

We have not published sample tasks. Cambridge publishes a sample test, and that is the right source.

We are not stating pass rates or average scores by part. We found no attributable figure.

How we researched this guide

The Part 2 description, the one minute reaction question, the four minute total, the skills list, the three part structure, the 16 and 24 minute totals, the two examiner setup, the remote examiner note and the marking app detail come from Cambridge English's C2 Proficiency exam format page, read directly in a browser.

The comparison of skills lists comes from reading the C1 Advanced and B2 First format pages the same way. Checking them field by field is how the absence of describing at C2 surfaced, and an absence is precisely the kind of thing a paraphrase cannot preserve: every summary of a picture task says "talk about the pictures", which is true at all three levels and useless at this one.

The C2 descriptor comes from the Council of Europe's global scale, Table 1, read on the page rather than quoted from memory.

Practise speaking, from SpeakShark

SpeakShark is an AI English speaking practice app, and it is our pick for Part 2 because both halves of it need a responder. A reaction question needs something to react to, and a decision task needs somebody who might not agree with you. You talk, the AI answers what you actually said, and you get speaking feedback while the conversation is still going. The free tier gives basic feedback; the detailed pronunciation and grammar breakdown is on Premium.

Being straight about the limits: a free session runs five minutes with four turns, which suits daily practice of the component skills rather than simulating a full 16 minute paired exam.

Start free with three sessions a day and no card. Paid sessions run ten minutes with unlimited turns. Limits are on the pricing page, and how it works walks through a full session.

We are a speaking improvement tool. We are not an exam preparation provider, and we are not affiliated with Cambridge English, the IELTS partners, Pearson or any exam board. For test format, booking and official practice material, go to the exam body directly. Use SpeakShark to make your spoken English stronger, and use official material to learn the test.

Sources

FAQ

What is C2 Proficiency Speaking Part 2?
It is the collaborative task. Cambridge says the interlocutor gives you spoken instructions and one or more pictures to look at. First you answer a question that focuses on your reaction to aspects of one or more of the pictures, which takes one minute. The second stage is a decision-making task that you carry out with the other candidate.
How long is C2 Proficiency Speaking Part 2?
Four minutes in total, of which the opening reaction question takes one minute. The whole Speaking test runs 16 minutes per pair of candidates or 24 minutes per group of three, so Part 2 is a quarter of it. Part 3 is much larger at ten minutes and Part 1 is two minutes.
Do I describe the pictures in C2 Proficiency Part 2?
No, and this is the trap. Cambridge says the question focuses on your reaction to aspects of one or more pictures. Reaction is not description. The pictures are the stimulus for a view rather than the subject of one, so a candidate who spends the minute listing what is visible has answered a question that was not asked.
How many pictures do I get in C2 Proficiency Part 2?
Cambridge says one or more, which is deliberately open. That is different from the levels below, where the count is fixed: a pair of photographs at B2 First, and three pictures with two to talk about at C1 Advanced. At C2 the number is not published in advance, so the task is written to work whatever appears.
What does Cambridge say to practise for Part 2?
Sustaining an interaction, then a familiar list: exchanging ideas, expressing and justifying opinions, agreeing and or disagreeing, suggesting, speculating, evaluating and reaching a decision through negotiation. Sustaining an interaction is the phrase that appears here and not at the lower levels, and it frames the whole part as endurance rather than exchange.
How do I practise the reaction question?
Take any image and answer a question about what it makes you think rather than what it contains. Give a view in the first sentence, then a reason, then a consequence, and keep going for a full minute without naming what is visible. That constraint is artificial but it trains the exact reflex the task rewards.

Keep reading