Series ID: EDKS-ADV-ZH-0010 · How to Learn Advanced English (Chinese Edition) · Lesson No.010 · C1 → C2

Adapt Complex English for Different Audiences | 把复杂英语准确讲给不同受众

The advanced skill is not making a difficult idea sound simple. It is making the idea easier to receive without changing what is true.

真正高级的能力，不是把困难内容“说简单”。而是在降低理解难度的同时，不改变事实。

A researcher, manager, teacher, parent, student, client and general reader may all need the same underlying information. They do not need the same version of the language.

研究者、经理、教师、家长、学生、客户和普通读者，可能需要理解同一个核心信息，但他们不需要同一种表达版本。

One audience may need technical definitions. Another needs the practical consequence first. One needs a full confidence interval. Another needs to know that the estimate is uncertain and what that uncertainty changes. One needs the mechanism. Another needs the action. One needs the original terminology because precision matters. Another needs the term explained before it can be useful.

一个受众需要技术定义；另一个受众需要先知道实际影响。一个人需要完整统计区间；另一个人只需要知道估计仍不确定，以及这个不确定性会怎样改变决定。一个人需要机制；另一个人需要行动。一个人必须保留原术语；另一个人必须先得到解释，术语才有意义。

So the task is not:

complex English → easy English.

The task is:

source meaning → audience need → accurate new form.

原始意义 → 受众需要 → 准确的新表达。

---

Before we begin: this lesson owns mediation across audiences | 开始之前：本课负责“跨受众 mediation”

This lesson does not replace the existing owners for register, school-stage situational writing, paraphrase, vocabulary or professional reports.

本课不取代现有的语域、学校 situational writing、paraphrase、词汇或专业报告主页面。

• How Register Changes English for Audience, Purpose and Relationship — general register mechanism.
• Secondary 3 English Learning Guide | Register and Style Transformation — school-stage transformation owner.
• How Paraphrasing Preserves Meaning Without Copying — paraphrase mechanism.
• Lesson 002 | Master Collocation, Register and Precision — register/precision inside this series.
• Lesson 004 | Read Deeply, Think Clearly, Write Precisely — meaning/evidence reconstruction.
• Lesson 009 | Write Evidence-Based Reports and Recommendations — evidence-to-action reporting.
• Advanced Vocabulary Lesson 017 | Paraphrase and Meaning-Preserving Vocabulary Choices — lexical reformulation owner.

Lesson 010 owns the integration problem: how to preserve the intellectual object while changing the communicative form for a different person.

第 010 课负责的是整合问题：同一个思想换给不同的人时，怎样改变表达形式，却不改变那个思想本身。

1. CEFR mediation gives us the right frame | CEFR mediation 提供正确框架

The Council of Europe describes mediation as a mode of communication in which a learner helps create meaning, convey ideas and information, or enable understanding across linguistic or cultural barriers. Mediation can happen across languages, but it can also happen within the same language.

欧洲委员会把 mediation 视为一种沟通模式：学习者帮助建立意义、传递信息、促进理解。它可以跨语言，也可以在同一种语言内部发生。

Council of Europe | Mediation

2. At C levels, adaptation is not only “simplify words” | C 级 mediation 不只是换简单词

CEFR mediation strategies include adapting language, breaking down complicated information, amplifying dense text and streamlining text. At advanced levels, the learner can change genre or register to fit audience and purpose while handling technical or complex concepts.

CEFR 的 mediation strategy 包括 adapting language、breaking down complicated information、amplifying a dense text 和 streamlining a text。高级阶段还要能根据 audience 和 purpose 改变 genre 或 register，同时处理复杂或技术性概念。

CEFR Companion Volume | Mediation strategies

3. Audience adaptation is a meaning-preservation problem | 受众改写，本质是意义保真问题

Suppose the source says:

The pilot result is promising, but the small self-selected sample and absence of a comparison group limit causal interpretation.

You can change the language. You cannot change these core states:

• result = promising;
• sample = small;
• participants = self-selected;
• comparison group = absent;
• causal conclusion = limited.

如果这些状态在新版本里消失，adaptation 就变成 distortion。

4. The conservation rule | “守恒定律”

When adapting, preserve:

1. Claim — 主张是什么？
2. Scope — 对谁、什么情况成立？
3. Certainty — 有多确定？
4. Condition — 需要什么条件？
5. Number — 数字关系是什么？
6. Time — 哪段时间？
7. Agency — 谁做了什么？
8. Authority — 谁能建议、批准、执行？
9. Risk — 风险大小和类型？
10. Unknown — 还有什么不知道？

你可以改变句型、词汇、顺序、例子和细节密度，但这些关键状态不能随便改变。

5. Simplification is not deletion | 简化不是删除

Technical:

The association remained statistically detectable after adjustment, although residual confounding cannot be excluded.

Accessible:

The relationship was still visible after the researchers accounted for several other factors, but they cannot rule out the possibility that unmeasured factors influenced the result.

Bad simplification:

The study proved the relationship is real.

前两个版本复杂度不同，核心意义仍相近。第三个版本把 uncertainty 删除了。

6. Do not translate difficulty into certainty | 不要把“难懂”翻译成“更确定”

Many explanations become misleading because the writer removes modal language:

• may becomes does;
• is associated with becomes causes;
• some participants becomes people;
• in this sample disappears;
• subject to approval disappears.

简化最容易丢掉的，恰恰是高级文本里最重要的边界。

7. Do not translate difficulty into vagueness | 也不要把复杂内容说成空话

Source:

System latency increased during peak demand because the database queue exceeded the configured worker capacity.

Too vague:

The system had some problems when it was busy.

Better for a non-technical manager:

When many users arrived at the same time, requests waited in a queue because the system could process only a limited number at once. This increased response time during peak periods.

你可以移除术语，但必须保留机制。

8. Start with audience action | 先问“受众要做什么”

Before adapting, complete:

After receiving this information, the audience needs to ______.

• understand;
• choose;
• approve;
• learn;
• avoid a risk;
• perform a task;
• ask a better question;
• explain it to someone else.

同一份 source 给不同目的，输出会不同。

9. Then define what they already know | 再判断受众已经知道什么

Audience knowledge changes what must be explained.

Audience	Likely prior knowledge	Adaptation need
specialist	technical concepts	precision, methods, exceptions
manager	system context	decision effect, cost, risk
new colleague	partial context	terms + workflow + ownership
general reader	limited domain knowledge	concept first, term second
child/young learner	limited abstract background	concrete model + carefully bounded analogy

10. Expertise changes compression | 专业程度改变可压缩程度

For an expert, selection bias may be enough.

For a general reader:

People who chose to join may differ from those who did not, so the result may not represent everyone.

第二句更长，但认知入口更简单。

11. Shorter is not always simpler | 更短不等于更简单

Residual confounding remains possible.

Only four words, but difficult for many readers.

Other factors that the study did not measure may still have affected the result.

Longer, but more accessible.

12. Plain language can remain intellectually precise | Plain language 可以很精准

The CDC Clear Communication Index emphasises identifying the primary audience, putting the main message early, using familiar language, organising information in chunks and explaining what is known and not known. Although this tool was developed for public-health communication, the underlying audience-control principles are useful far beyond health communication.

CDC Clear Communication Index 强调：确定 primary audience、把 main message 放前面、使用受众熟悉的语言、分块组织信息，并说明已知和未知。它原本服务公共健康沟通，但这些 audience-control 原则也可用于其他专业领域。

CDC Clear Communication Index | Core items

13. Put the main message where the audience can find it | 主信息要放在受众找得到的地方

Expert paper:

Background may come first because readers expect the argument to be built through a conventional structure.

Operational notice:

The deadline has moved to Friday. should normally appear before the history of why.

内容顺序也属于 adaptation。

14. “What is the point?” is an audience question | “重点是什么”是受众问题

One source can yield different main messages:

• Researcher: The design does not isolate causation.
• Manager: Evidence is not strong enough for full rollout.
• User: The service will remain in pilot for another month.

这些句子不同，但可以同时忠实于同一证据状态。

15. Lead with consequence when consequence drives action | 需要行动时，先给影响

Technical first:

The authentication service is experiencing elevated error rates.

User first:

You may not be able to sign in right now. We are working on the authentication service and will update this notice when access is stable.

受众不同，information order 不同。

16. Lead with mechanism when learning is the goal | 教学目标不同，机制可以先出现

For a learner, the purpose may be understanding:

A queue forms when requests arrive faster than the system can process them. That is why response time increases during peak demand.

这里不是先说 action，而是先建立 mental model。

17. Define technical terms at first useful contact | 技术词在第一次真正需要时解释

Latency means the delay between a request and the system’s response.

Then use the term:

Latency was highest between 8 and 9 a.m.

这比先列十个 glossary 定义更符合阅读路径。

18. Keep the technical term when the reader will need it later | 后续会用到，就保留术语

Do not permanently replace hypertension with high blood pressure if the audience will see the technical term in appointments, reports or instructions. Teach the equivalence, then use both strategically.

真正好的简化，不会让受众离开材料后反而无法识别真实世界术语。

19. Remove jargon that performs no unique job | 没有独特任务的 jargon 可以删

Leverage cross-functional synergies to operationalise strategic alignment.

可能只是：

Ask the teams to coordinate their work around the same priorities.

高级写作不是用 jargon 建立地位。

20. Acronyms create hidden barriers | Acronym 会制造隐形门槛

If a reader sees five unfamiliar abbreviations in one paragraph, working memory is now partly doing acronym recovery rather than understanding the idea.

首次使用时写全称，必要时再给缩写。若后面根本不重复，可能不需要缩写。

21. Translate concept before terminology | 先建立概念，再给术语

For a novice:

People who volunteer for a study may be different from people who do not. Researchers call this self-selection bias.

This gives the learner a concept to attach the term to.

22. Analogy is a bridge, not the destination | 类比是桥，不是答案

Analogy:

Working memory is a little like a small desk: if too many items arrive at once, there is less room to work with each one.

Then boundary:

But memory is not literally a desk, and the analogy does not describe all memory systems.

类比必须带边界。

23. Bad analogies preserve familiarity but destroy mechanism | 坏类比熟悉，却破坏机制

If an analogy changes causation, scale or agency, it is not helping.

Test:

• What exactly corresponds?
• What does not correspond?
• Could the audience infer something false?

24. Concrete examples should reveal the mechanism | 例子要显示机制

Abstract:

Register depends on relationship and purpose.

Concrete:

A message to a close friend may say “Can you send it over?” A formal request to an external organisation may need “Could you please send the revised document by Friday?”

例子不是装饰，而是让抽象规则变可见。

25. One example is not the rule | 一个例子不是整个规则

After an example, return to abstraction:

The difference is not simply casual versus formal vocabulary; the relationship, obligation and consequence also change the appropriate form.

26. Numbers must be adapted without being distorted | 数字也要适配，但不能失真

Technical:

17.8% of 1,243 cases.

General:

About 18 in every 100 cases, based on 1,243 cases reviewed.

Both preserve denominator and approximation.

27. Do not make small risks sound dramatic | 不要把小风险说得吓人

Risk doubled can sound large.

If it changed from 1 in 10,000 to 2 in 10,000, the absolute risk is also important.

受众适配必须保护相对风险和绝对风险的关系。

28. Do not make large risks disappear into technical notation | 也不要用技术数字把大风险藏起来

Numbers should support the audience’s decision, not demonstrate the writer’s statistical vocabulary.

29. Preserve uncertainty around estimates | 估计值要保留不确定性

Technical:

The estimate was 42% (95% CI 35–49%).

General reader:

The study estimated the proportion at about 42%, but the true value could plausibly be several percentage points lower or higher.

根据任务需要，可以保留原区间，也可以先解释区间意义。

30. Risk communication requires “what we know / what we do not know” | 风险沟通要分已知与未知

Useful structure:

• Known: what evidence establishes.
• Unknown: what is not established.
• Action: what is reasonable despite uncertainty.

这比假装 uncertainty 不存在更容易建立信任。

31. Adapt evidence density | 调整证据密度

Expert:

May need methods, effect sizes, subgroup analysis and sensitivity checks.

Executive:

May need three decision-relevant findings, one uncertainty and one action.

General reader:

May need one main message, one example and a route to more detail.

删细节不等于删证据逻辑。

32. Use progressive disclosure | 用 progressive disclosure 分层给信息

Layer 1: main message.

Layer 2: explanation.

Layer 3: detailed evidence.

Layer 4: technical source/method.

不同读者可以停在不同深度。

33. A headline must not be stronger than the body | 标题不能比正文更确定

Body:

The pilot suggests a possible reduction in waiting time.

Bad headline:

NEW SYSTEM CUTS WAITING TIME

标题也是 claim。

34. Social-media compression is especially dangerous | 社交媒体压缩最容易失真

When turning a report into one post, preserve at least:

• main finding;
• population/context;
• critical uncertainty;
• source link.

不要把 nuance 全部留在“点击后才能看到”的地方。

35. Executive summaries and public summaries are not the same | Executive summary 与 public summary 不同

An executive summary assumes organisational context. A public summary may need to explain terms, actors and why the issue matters.

同一份报告可以有两个 summary，因为受众 mental model 不同。

36. Academic abstract and press explanation are not the same | Abstract 与公众解释不同

Academic abstract:

compresses methods/results for readers who understand the field.

Public explanation:

often expands key concepts before compressing implications.

有时 adaptation 会让文本变长，而不是变短。

37. Technical support and user support are not the same | 技术支持与用户说明不同

Engineer:

Invalidate the cached credential and refresh the token.

User:

Sign out, close the app, then sign in again. This creates a fresh login session.

同一个修复动作，需要不同语言入口。

38. Teacher explanation and parent explanation are not the same | 教师说明与家长说明不同

Teacher:

The learner has adequate literal comprehension but weak inferential evidence selection.

Parent:

Your child usually understands what the passage says directly, but loses marks when the answer requires choosing the right clues and explaining what they imply.

第二句不是“低水平版”，而是把 mechanism 展开。

39. Parent-friendly does not mean childish | 面向家长不等于幼稚化

Respect the audience’s intelligence. Remove unnecessary specialist code, not intellectual substance.

不要用 baby talk。清楚和幼稚是两件事。

40. Child-friendly does not mean inaccurate | 面向孩子也不能说错

For a younger learner, use:

• shorter idea units;
• concrete examples;
• one new term at a time;
• visible cause-and-effect;
• retrieval questions.

不要为了简单而制造假规则。

41. “Always” and “never” are dangerous teaching words | 教学里的 always / never 要小心

Weak simplification:

Never use passive voice.

Better:

Use active voice when the actor matters and directness helps. Passive voice is useful when the action/result matters more than the actor or when the actor is unknown.

42. Beginner rules should contain escape hatches | 初级规则要给“出口”

Teach:

As a starting rule…

In most ordinary cases…

Later, you will meet exceptions where…

这样 learner 不会把 scaffold 误认为 universal law。

43. Expert-to-novice adaptation should preserve question structure | 专家转新手，要保留问题结构

Expert question:

Does the observed association remain after controlling for baseline differences?

Novice version:

If we compare people who started at different levels, does the relationship still remain?

核心判断任务没有变。

44. Novice-to-expert adaptation may require compression | 新手转专家，需要压缩

Novice:

People who volunteer may already be more motivated, so maybe that is why the result looks better.

Expert:

The result may be affected by self-selection.

高级 mediation 是双向的。

45. Chinese can help identify the concept before English adaptation | 中文可以帮助先锁定概念

If your Chinese understanding is strong, use it to answer:

• What exactly is the claim?
• What is the relation?
• What must not change?

Then rebuild for the English audience.

不要逐字翻译；先锁定 conceptual invariant。

46. Translate function, not surface grammar | 翻译交际功能，不复制表面结构

Chinese:

这个结论还需要进一步观察。

Possible English depending on function:

• Further evidence is needed before drawing a firm conclusion.
• We should monitor the result for another term before deciding.
• The current evidence is encouraging but not yet conclusive.

三句功能不同：research、decision、summary。

47. Do not import Chinese information order automatically | 不要自动照搬中文信息顺序

A Chinese explanation may build context before the request. In some English workplace settings, the reader may need the request first, then context.

But this is not a universal East/West rule. Audience, organisation and relationship matter more than stereotypes.

48. Do not turn cultural adaptation into caricature | 不要把文化差异写成刻板印象

Avoid:

Chinese people are indirect; English speakers are direct.

Better:

Different settings create different expectations about directness, hierarchy, disagreement and explanation. Observe the actual community you are communicating with.

49. Relationship changes what “clear” sounds like | 关系会改变“清楚”的形式

Friend:

Can you send me the file tonight?

Colleague:

Could you send the revised file by 6 p.m. so I can complete the review?

External client:

Could you please send the revised file by 6 p.m. today? This will allow us to complete the scheduled review.

核心请求相同，relationship 和 consequence 改变 packaging。

50. Power distance changes responsibility for explicitness | 权限差异会影响显性程度

When communicating upward, a learner may hedge so strongly that the problem disappears.

Too weak:

There might perhaps be a small issue with the deadline.

Clear and respectful:

The current deadline cannot be met with the available data. I recommend moving submission to Friday or reducing the requested scope.

51. Politeness should not hide operational facts | 礼貌不能隐藏行动事实

You can be courteous and explicit at the same time.

52. Bad news needs structure | 坏消息需要结构

Useful sequence:

1. state the important fact;
2. explain cause if useful;
3. state impact;
4. state available action;
5. state what remains uncertain.

53. Do not bury safety-critical information | 关键风险不能埋在最后

If the reader must act to avoid harm, lead with the action and condition.

文章风格必须服从 consequence。

54. Adapt instructions by user state | 指令要按用户状态适配

New user:

needs every critical step.

Experienced user:

may need only the changed step.

Troubleshooting user:

needs decision branches.

55. Procedures need verbs | 流程说明需要动作动词

Weak:

Completion of verification is required prior to submission.

Clear:

Verify the record before you submit it.

除非名词化承担分析任务，否则动作最好可见。

56. Instruction order should match action order | 指令顺序尽量匹配实际动作顺序

Do not make the user mentally reorder six steps.

57. Conditions should appear before irreversible actions | 不可逆动作前先给条件

Before deleting the original file, confirm that the backup opens correctly.

比删除后再解释 backup 更安全。

58. Explanations need different depth levels | 解释需要多层深度

Try four layers:

1. One sentence.
2. One paragraph.
3. One worked example.
4. Technical detail.

如果四层都能说清，说明你真正掌握了概念。

59. The one-sentence test | 一句话测试

Explain the core claim in one sentence without removing its critical limitation.

This is harder than simply shortening.

60. The 50-word test | 50 词测试

Write a 50-word version for a busy non-specialist. Preserve:

• main claim;
• scope;
• one critical uncertainty;
• action if relevant.

61. The 200-word test | 200 词测试

Add:

• mechanism;
• evidence;
• example;
• alternative explanation.

62. The expert test | 专家版本测试

Restore:

• technical terminology;
• measurement detail;
• methods;
• exceptions;
• source traceability.

如果不同版本互相矛盾，adaptation 失败。

63. The reverse-reconstruction test | 反向重建测试

Give the simplified version to someone. Ask:

What do you think the original claim was?

If their reconstructed claim is much stronger, broader or different, your version lost fidelity.

64. The question-preservation test | 问题保留测试

Does the new version invite the same decision?

Source asks:

Is the evidence sufficient for full adoption?

Adapted version should not silently become:

Why should we adopt this?

65. The certainty-preservation test | 确定程度保留测试

Mark every:

• may;
• might;
• likely;
• suggests;
• appears;
• cannot rule out;
• uncertain.

Check whether the adapted version keeps the same epistemic state.

66. The condition-preservation test | 条件保留测试

Source:

Proceed if support is confirmed.

Adapted:

Proceed.

One missing clause can change an operational decision.

67. The agency-preservation test | 责任主体保留测试

Source:

The review team recommends the change.

Bad adaptation:

The organisation has decided to change.

Recommendation ≠ decision.

68. The time-preservation test | 时间状态保留测试

was planned, is planned, has started and has finished are different project states.

摘要时不能把 timeline 压平。

69. The denominator-preservation test | 分母保留测试

80% of respondents is not automatically 80% of users.

70. The population-preservation test | 人群范围保留测试

Adults in one clinic is not people.

71. The exception-preservation test | 例外保留测试

If the exception materially changes the decision, it must survive simplification.

72. The trade-off-preservation test | 取舍保留测试

If Option B is faster but less reliable, do not adapt it to Option B is better.

73. Use audience personas carefully | Audience persona 要具体但不要幻想

Useful:

A department manager who knows the project but not the statistical method and must decide whether to approve a small pilot.

Weak:

Busy people.

Persona 要能改变写作选择。

74. Ask four audience questions | 四个 audience question

1. What do they already know?
2. What do they need to do?
3. What could they misunderstand?
4. What detail must remain even if they dislike complexity?

75. Use comprehension risk, not writer preference | 按误解风险决定解释深度

Explain more where misunderstanding has consequence.

Explain less where detail does not change interpretation or action.

76. Familiar language is not always informal | 熟悉语言不等于不正式

use may be clearer than utilise.

before may be clearer than prior to.

because may be clearer than due to the fact that.

清楚不是“低级英语”。

77. Advanced English includes the ability to choose simple words | 能选择简单词，本身就是高级能力

C1–C2 control means you can use complexity when it earns its place and remove it when it does not.

78. Sentence length should follow information structure | 句长服从信息结构

Split a sentence when the audience must process:

• two separate conditions;
• a claim and a warning;
• a result and an exception;
• a decision and an action.

79. But excessive short sentences can destroy relationships | 全部短句也会破坏关系

Weak:

The pilot improved speed. The groups were different. We cannot prove cause. We recommend extension.

Better:

The pilot improved speed, but the groups were different, so the result does not establish causation. We therefore recommend an extension rather than full adoption.

逻辑关系必须可见。

80. Chunk dense information | 把密集信息分块

Use:

• headings;
• bullets;
• tables;
• question-answer structure;
• progressive disclosure.

CDC’s communication guidance similarly emphasises headings and chunks for helping audiences find important information.

81. Headings should answer audience questions | 标题最好回应受众问题

Weak:

Background

Stronger for a public reader:

Why the service is changing

Stronger for a manager:

Why the current process cannot meet peak demand

82. Tone should preserve relationship without hiding content | Tone 保护关系，但不隐藏内容

Too blunt:

Your interpretation is wrong.

Too vague:

There may perhaps be another way of looking at it.

Useful:

I think the evidence supports a narrower conclusion because the sample includes only volunteers.

83. Different audiences need different evidence verbs | 不同受众需要不同 evidence language

Expert:

The model estimates…

General:

The analysis gives an estimate of…

Child:

The researchers used the information they had to make their best estimate.

84. Preserve source distance | 保留信息来源距离

The company says… is not It is true that…

The witness reported… is not The event happened exactly this way.

Adaptation must preserve attribution.

85. Preserve quotation status | 保留引语状态

Direct quote, paraphrase and writer conclusion are not interchangeable.

86. Preserve contested status | 有争议的内容必须保留争议状态

If experts disagree, do not simplify to one side unless your task explicitly evaluates and justifies that choice.

87. Preserve date/version | 版本与日期不能丢

Current guidance as of September 2026 is different from timeless truth.

任何容易更新的政策、软件、价格或流程，adaptation 都要保护版本时间。

88. Use examples from the audience’s world | 例子尽量来自受众世界

A student may understand probability through exam questions.

A manager may understand it through forecasting.

A parent may understand it through everyday risk.

概念不变，入口改变。

89. But do not force local examples when the concept is global | 不要为“贴近”硬塞本地例子

Use examples because they clarify mechanism, not because every paragraph needs geographic decoration.

90. Adapt questions as well as answers | 问题也要适配

Expert:

What alternative mechanisms could explain the observed association?

General:

What else could have caused the same pattern?

Same reasoning task; different linguistic load.

91. Adapt feedback | Feedback 也需要 audience adaptation

Technical feedback:

Your paragraph overstates causality and loses the population boundary.

Learner feedback:

You changed “may have caused” into “caused”, and you changed “these 30 participants” into “people”. Repair those two meanings first.

第二句更可操作。

92. Adapt definitions by purpose | 定义也有不同用途

Dictionary-style:

Bias: a systematic distortion in measurement or inference.

Teaching:

Bias means the process pushes the result in a particular direction, not merely that a person has an opinion.

Operational:

Check whether the way we selected users could systematically favour one outcome.

93. Adapt presentations for listening, not reading | 演讲要为耳朵写，不是为眼睛写

Listeners cannot easily reread the previous sentence.

Use:

• clear signposting;
• shorter idea units;
• strategic repetition;
• explicit contrasts;
• spoken definitions.

94. Adapt slides for the room | 幻灯片服务现场，不复制报告

A slide should not carry the full report paragraph if the speaker is explaining it aloud.

Keep traceability through notes/source links when needed.

95. Adapt email for action | Email 首先要帮助行动

Useful order:

1. purpose/request;
2. deadline/condition;
3. minimum context;
4. detail/link.

除非 relationship requires a different opening.

96. Adapt chat for immediacy | Chat 要快，但不能丢关键状态

Approved, recommended, under review and blocked must remain distinct even in one-line messages.

97. Adapt reports into decisions | 把 report 转成 decision brief

Keep:

• decision required;
• top evidence;
• largest uncertainty;
• options;
• recommendation;
• condition that would change it.

98. Adapt discussion into handover | 把讨论转成交接

Lesson 008 taught discussion control. Lesson 009 taught reports. Here, mediation asks:

What does the next person need in order to continue without reconstructing the whole conversation?

99. A handover is a compressed state machine | 交接是压缩后的状态机

• done;
• decided;
• recommended;
• open;
• blocked;
• next.

状态词比漂亮叙事更重要。

100. Worked case: one source, five audiences | 完整案例：同一来源，五种受众

Fictional source statement:

A six-week pilot involved 48 volunteer users. Median task completion time fell from 11 minutes to 8 minutes. Error rate did not materially change. Users received additional onboarding during the pilot, and there was no randomly assigned comparison group. The team therefore recommends a larger controlled trial rather than full deployment.

虚构来源：六周试点包含 48 名志愿用户。任务完成时间中位数从 11 分钟下降到 8 分钟。错误率没有明显变化。试点用户接受了额外 onboarding，而且没有随机对照组。因此团队建议进行更大的控制性试验，而不是全面部署。

101. Audience 1: technical analyst | 受众 1：技术分析人员

In the six-week volunteer pilot (n=48), median completion time decreased from 11 to 8 minutes, while the observed error rate remained broadly stable. Because the pilot combined process change with additional onboarding and lacked random allocation to a comparison group, the observed time reduction cannot be attributed to the new process alone. A larger controlled evaluation is recommended before deployment.

保留：sample、时间、结果、confound、comparison limitation、recommendation。

102. Audience 2: senior manager | 受众 2：管理层

The pilot was faster, but it did not show fewer errors, and we cannot yet tell how much of the speed improvement came from the new process rather than extra onboarding. The team therefore recommends a larger controlled trial before approving full deployment.

减少方法术语，保留 decision logic。

103. Audience 3: participating user | 受众 3：用户

People in the pilot completed the task faster on average, but the team is still checking whether the new process itself caused the improvement. For now, the system will remain in testing rather than being rolled out to everyone.

保留状态：faster、uncertain cause、still testing、not full rollout。

104. Audience 4: new employee | 受众 4：新员工

The first trial looked promising: users finished about three minutes faster, but they also received extra training. Because the trial did not include a comparable group using the old process under the same conditions, we need another test before we can say the new process is responsible for the improvement.

把 comparison group 的概念展开。

105. Audience 5: younger learner | 受众 5：较年轻学习者

The new system looked faster in the first test. But the people using it also got extra help, so we do not yet know whether the system or the extra help made the difference. The next test should compare the two more fairly.

语言简单，但没有把 uncertainty 删除。

106. What remained unchanged across all five? | 五个版本里什么没变？

• pilot looked faster;
• errors did not clearly improve;
• extra onboarding complicates interpretation;
• comparison design is insufficient for causation;
• full deployment is not yet recommended.

这就是 conceptual invariant。

107. What changed? | 什么改变了？

• technical terminology;
• sentence density;
• number presentation;
• method explanation depth;
• decision emphasis;
• example/analogy level.

108. Practice A: preserve certainty | 练习 A：保留确定程度

Source:

The evidence suggests that the intervention may reduce waiting time.

Write:

1. a general-reader version;
2. a manager version;
3. a younger learner version.

Do not upgrade suggests/may.

109. Practice B: preserve condition | 练习 B：保留条件

Source:

The trial can proceed if data protection approval is obtained.

Rewrite for:

• project chat;
• formal email;
• public FAQ.

110. Practice C: explain terminology | 练习 C：解释术语

Explain selection bias to:

• a Secondary student;
• a university student;
• a non-technical manager.

111. Practice D: adapt a number | 练习 D：数字适配

Source:

17.8% of 1,243 cases.

Create:

• technical version;
• general version;
• spoken presentation version.

Preserve denominator.

112. Practice E: preserve authority | 练习 E：保留权限

Source:

The committee recommends extending the pilot; final approval rests with the director.

Adapt for one-line chat without turning recommendation into approval.

113. Practice F: explain a mechanism | 练习 F：解释机制

Source:

Response time increases when request arrival exceeds processing capacity.

Explain to:

• software engineer;
• manager;
• 12-year-old.

114. Practice G: adapt bad news | 练习 G：改写坏消息

Source:

The project cannot launch Monday because the required security review is incomplete.

Write:

• internal executive update;
• team message;
• external client notice.

Do not hide the launch impact.

115. Practice H: simplify without fake rules | 练习 H：简化但不制造假规则

Explain why passive voice is sometimes useful to an intermediate learner without saying “never use passive”.

116. Practice I: adapt a research finding | 练习 I：研究结果适配

Source:

Participants in the intervention group improved more than controls at post-test, but the difference was not maintained at delayed follow-up.

Write versions for:

• researcher;
• teacher;
• parent.

117. Practice J: translate function, not Chinese syntax | 练习 J：翻功能，不翻表面

Chinese intent:

目前来看这个方案可行，但还不能过早下结论。

Write:

• professional meeting version;
• academic writing version;
• public explanation version.

118. Model answers | 示范答案

A general reader: The evidence points to a possible reduction in waiting time, but the result is not yet certain.

A manager: Waiting time may be lower under the intervention, but the current evidence is not strong enough to treat the reduction as established.

A younger learner: The new method might make people wait less, but we need more evidence before we can be sure.

B chat: Trial can start once data-protection approval is confirmed.

B email: We are ready to proceed with the trial, subject to final data-protection approval.

B FAQ: The trial has not started yet. It will begin only after the required data-protection review is approved.

C Secondary: Selection bias can happen when the people who join are different from the people who do not, so the group may not represent everyone.

C university: Selection bias occurs when the selection process creates systematic differences between the observed sample and the target population.

C manager: Because participation was voluntary, the users who joined may have been unusually motivated; the result may therefore overstate what would happen across all users.

D technical: 221 of 1,243 cases (17.8%).

D general: About 18 in every 100 cases, based on 1,243 cases reviewed.

D spoken: Roughly one case in six—221 out of 1,243—met the criterion.

E: Committee recommends extending the pilot; director approval is still required.

F manager: When requests arrive faster than the system can handle them, they queue up and response time rises.

F child: Imagine one checkout counter with many people arriving at once: the line grows because the counter cannot serve everyone immediately.

G executive: Monday launch is not feasible because the mandatory security review is incomplete. The earliest revised date depends on review completion.

H: Active voice is often clearer when the actor matters. Passive voice is useful when the result matters more, when the actor is unknown or when the sentence needs to connect smoothly to earlier information.

I parent: The students improved more immediately after the programme, but the advantage was no longer clear at the later check. That means the programme may have helped short-term performance without yet showing a lasting effect.

J meeting: The option looks workable at this stage, but I would avoid treating the early result as conclusive.

J academic: The preliminary evidence supports feasibility, although a firm conclusion would be premature.

J public: The plan looks possible, but more evidence is needed before a final decision.

119. The five-version drill | 五版本训练

Take one difficult paragraph and write it for:

1. expert;
2. manager;
3. general adult;
4. teenager;
5. younger learner.

Then highlight the invariants in all five.

120. The format-switch drill | 换格式训练

Take one source and transform it into:

• report paragraph;
• executive summary;
• email;
• chat update;
• FAQ;
• spoken explanation.

形式变，事实状态不变。

121. The terminology ladder | 术语阶梯

For one concept, create:

• technical term;
• plain-English definition;
• concrete example;
• analogy with boundary;
• one-question comprehension check.

122. The uncertainty ladder | 不确定性阶梯

Practise:

• possible;
• plausible;
• likely;
• strongly supported;
• established within the current evidence.

不要把这些词当固定百分比。重点是保持 relative commitment。

123. The “what can I delete?” audit | “什么可以删”审计

Delete if the detail:

• does not change understanding;
• does not change decision;
• does not protect accuracy;
• is available through a deeper link.

Keep if deletion changes scope, certainty, condition, risk or responsibility.

124. The “what must I add?” audit | “什么必须加”审计

Add explanation when the audience lacks:

• term knowledge;
• background assumption;
• mechanism;
• reference point;
• action implication.

125. The “what could they misunderstand?” audit | “最可能误解什么”审计

Before publishing, predict one wrong inference.

Then repair the wording before the reader makes it.

126. Use real audience testing when consequence matters | 重要材料要做真实受众测试

The CDC Clear Communication Index recommends testing materials with the primary audience to confirm that people understand and can use the material as intended.

这不是所有课堂练习的必要步骤，但在高影响公众材料、用户指令或风险沟通中，非常有价值。

CDC | How to Use the Clear Communication Index

127. Do not use readability scores as truth | 不要把 readability score 当成真相

A short word can express a difficult concept. A long technical word can be necessary and well explained. Readability formulas cannot determine whether your causal claim, limitation or authority boundary is correct.

读起来“简单”不等于意义正确。

128. Do not remove necessary complexity | 必要复杂度不能删除

Some ideas are genuinely complex.

Your job is to sequence the complexity, not pretend it does not exist.

129. Build from known → new | 从已知到未知

CEFR mediation strategies include linking to previous knowledge. Use a familiar concept to anchor a new one, then explicitly show the new distinction.

130. Amplify a dense text when compression makes it harder | 太密时，要“展开”而不是继续压缩

Dense source:

The policy is regressive.

Novice explanation:

The policy takes a larger share of income from lower-income households than from higher-income households. Economists describe that pattern as regressive.

更长，但更容易理解。

131. Streamline repetitive text when detail blocks the main relation | 太重复时，再 streamline

If three paragraphs repeat the same limitation, combine them once where the reader needs it.

132. Adapt genre, not just sentence vocabulary | 改 genre，不只是换词

A scientific finding may become:

• abstract;
• policy brief;
• teacher explanation;
• parent note;
• FAQ;
• presentation.

每个 genre 有自己的 reader path。

133. Cambridge C1 writing reinforces target-reader control | Cambridge C1 也强调 target reader

In Cambridge C1 Advanced Writing Part 2, the task provides context, topic, purpose and target reader for text types such as reports, proposals, emails/letters and reviews. This is an examination example of a wider professional principle: good writing changes with its communicative job.

Cambridge English | C1 Advanced format

134. A report version and proposal version should differ | Report 与 proposal 版本不同

Report:

The trial reduced median waiting time by three minutes, although staffing cost increased.

Proposal:

Run a second trial focused on whether the time saving can be retained with lower staffing cost.

同一信息，genre 任务不同。

135. A review version changes again | Review 又不同

A review evaluates suitability for a particular audience or purpose. It may ask:

Who would benefit from this tool, and under what conditions?

136. The audience is not always one person | 受众可能是混合群体

A board paper may be read by finance, operations, legal and technical staff.

Use:

• shared executive layer;
• specialist sections;
• appendices.

不要试图把所有细节都塞进第一层。

137. Mixed audiences need terminology bridges | 混合受众需要术语桥

Rollback—the process of returning to the previous stable version—must be tested before launch.

专家不会受伤，新手也能跟上。

138. Audience adaptation can be spoken in real time | 受众适配也可以实时发生

If a listener looks confused:

• rephrase;
• give an example;
• define a term;
• reduce sentence density;
• check understanding.

这连接 Lesson 006 与 008。

139. Check comprehension without humiliating the listener | 检查理解，不要让人尴尬

Instead of:

Do you understand?

Try:

What would this change in practice for your team?

or:

Which part should I unpack further?

140. Ask the audience to reconstruct | 让受众重建

If they can explain the relationship in their own words, you get better evidence of understanding than a polite “yes”.

141. Mediation is not showing that you know more | Mediation 不是展示你懂得更多

The successful mediator becomes less visible as understanding becomes more stable.

142. The strongest adaptation may preserve a difficult word | 最好的改写有时会保留难词

If the term is essential, keep it and explain it.

Do not delete precision just because the word is advanced.

143. The strongest adaptation may replace an easy word | 有时简单词反而太模糊

thing, problem, good, bad may need more precise alternatives if the decision depends on the distinction.

144. Audience adaptation is an optimisation problem | Audience adaptation 是优化问题

You are balancing:

• accuracy;
• accessibility;
• brevity;
• actionability;
• tone;
• traceability.

没有一个指标可以永远压倒其他指标。

145. The first weak-link table | 跨受众表达的第一个薄弱环节

Symptom	Likely weak link	Repair
I simplify but become inaccurate.	conceptual invariant unclear	mark claim/scope/certainty first
I keep all details and overwhelm readers.	audience action unclear	define decision/use first
I use simple words but remain hard to understand.	mechanism hidden	add causal relation/example
I sound childish.	over-simplification	respect adult reasoning; explain terms
I sound too technical.	terminology before concept	concept → term
my summaries sound more certain	epistemic drift	certainty-preservation audit
actions become unclear	authority/state drift	preserve recommend/approve/do
Chinese version and English version feel different	surface translation	translate function and invariant

146. A 20-minute daily mediation routine | 每天 20 分钟 mediation

Time	Task
4 min	read one dense paragraph
4 min	mark invariants
4 min	write general-reader version
4 min	write expert/manager version
4 min	compare for meaning drift

147. A 45-minute growth session | 45 分钟增长模式

Time	Task
8 min	source reconstruction
8 min	audience model
12 min	three audience versions
8 min	spoken adaptation
9 min	fidelity audit + repair

148. A 90-minute deep session | 90 分钟深度训练

Time	Task
15 min	multi-source technical reading
15 min	claim/scope/certainty map
20 min	expert, executive and public versions
15 min	spoken teaching version
10 min	format switch: email/FAQ/chat
15 min	reverse reconstruction + audit

149. Seven-day adaptation cycle | 七天跨受众训练

1. Day 1: scope and certainty.
2. Day 2: terminology and definitions.
3. Day 3: numbers and risk.
4. Day 4: expert ↔ general reader.
5. Day 5: report ↔ email ↔ chat.
6. Day 6: English ↔ Mandarin conceptual mediation.
7. Day 7: unfamiliar full benchmark.

150. Twelve-week C1–C2 progression | 12 周 C1–C2 mediation 路线

Weeks	Focus	Evidence
1–2	meaning invariants	claim/scope maps
3–4	terminology, examples, analogies	multi-level explanations
5–6	numbers, uncertainty, risk	public/technical pairs
7–8	genre and register transformation	report/email/FAQ sets
9–10	real-time spoken adaptation	recorded explanations
11–12	multi-source bilingual mediation	final portfolio

151. Monthly benchmark | 每月基准任务

Choose one unfamiliar 600–1,000 word source containing:

• one important number;
• one qualification;
• one technical term;
• one uncertainty;
• one recommendation or implication.

Create:

1. 100-word expert summary;
2. 100-word general-reader summary;
3. 60-word executive version;
4. spoken two-minute teaching explanation;
5. one bilingual Chinese-English concept map.

Then audit the five invariants.

152. Measure more than vocabulary | 不要只评价“用词高级不高级”

Track:

• meaning fidelity;
• scope preservation;
• certainty preservation;
• audience fit;
• terminology control;
• information order;
• example quality;
• action clarity;
• authority accuracy;
• ability to explain misunderstanding.

153. What not to do | 不要这样做跨受众改写

• Do not replace every technical word with an everyday word.
• Do not delete uncertainty to make the message cleaner.
• Do not hide denominator, scope or timeframe.
• Do not treat adults like children.
• Do not give children false universal rules.
• Do not use analogy without stating where it breaks.
• Do not assume Chinese and English audiences are culturally uniform.
• Do not turn a recommendation into a decision.
• Do not turn a correlation into a cause.
• Do not equate shorter with clearer.

154. Reference floor | 参考基础

• Council of Europe | Mediation — creating meaning, conveying information and enabling communication.
• Council of Europe | CEFR Descriptors — updated descriptor system including mediation.
• CEFR Companion Volume | Mediation Strategies — adapting language, explaining concepts, streamlining and amplifying text.
• Cambridge C1 Advanced Format — context, purpose and target reader in advanced writing.
• CDC Clear Communication Index — audience, main message, familiar language, information design and state of knowledge.
• CDC | How to Use the Index — audience definition, drafting and audience testing.

These sources provide a quality floor, not one universal writing formula. The exact adaptation depends on purpose, audience, consequence and domain.

155. Canonical eduKate routes | eduKate 主页面路由

• How Register Changes English for Audience, Purpose and Relationship
• How Paraphrasing Preserves Meaning Without Copying
• Secondary 3 Register and Style Transformation
• Secondary 4 Situational Writing: Audience, Context, Information Transformation and Tone
• Lesson 009 | Write Evidence-Based Reports and Recommendations
• Advanced Vocabulary Lesson 017 | Paraphrase, Reformulation and Meaning-Preserving Vocabulary Choices

156. SEO language map | 本课自然覆盖的搜索意图

This lesson naturally serves learners searching for advanced English mediation, C1 mediation, C2 English communication, explain complex ideas simply, adapt English for audience, English for different audiences, technical English to plain English, professional English for Chinese speakers, English communication for Mandarin speakers, paraphrase complex ideas, audience adaptation, plain English without losing meaning, explain technical terms, academic to public communication, 高级英语表达, 英语受众意识, 英语改写, 英语简化, 中文母语高级英语, 怎样把复杂英语说简单, 怎样解释复杂概念, 英语 mediation, 英语 audience adaptation and 跨受众表达.

157. The one-page audience-adaptation operating system | 一页跨受众表达系统

1. Reconstruct the source meaning. 先重建原义。
2. Mark invariants. 主张、范围、确定程度、条件、数字、责任、风险。
3. Name the audience action. 受众要理解、决定还是执行？
4. Estimate prior knowledge. 已经知道什么？
5. Predict misunderstanding. 最可能误解什么？
6. Choose the main message. 哪一点必须先出现？
7. Adapt terminology. 保留、定义或替换。
8. Adapt information density. 分层深度。
9. Adapt examples and format. 改入口，不改事实。
10. Audit certainty and scope. 防止意义漂移。
11. Reverse-reconstruct. 看别人会重建出什么。
12. Test with the audience when consequence is high. 高影响时做真实测试。

158. Final assignment | 最终作业

Choose one difficult source from science, education, finance, technology, public policy, history or your own professional field. Use a non-confidential source that you are allowed to work with.

选择一份科学、教育、金融、技术、公共政策、历史或自己专业领域的困难材料。使用非保密、允许使用的来源。

Complete this full chain:

1. Write the source claim in one precise sentence.
2. Mark scope, certainty, conditions, numbers, authority and unknowns.
3. Create an expert version.
4. Create a manager version.
5. Create a general-reader version.
6. Create a teenager/younger learner version.
7. Create a 60-second spoken explanation.
8. Create a one-line chat update.
9. Translate the core concept into Chinese, then rebuild it into English without looking at your earlier wording.
10. Compare every version against the invariant map.
11. Ask another person to reconstruct the original claim from your general-reader version.
12. Repair any change in scope, certainty, condition or authority.

Then ask:

Did I make the idea easier to receive—or did I quietly change the idea so that it became easier to say?

我到底是让这个思想更容易被接收，还是偷偷改变了思想，只为了让它更容易说？

That question separates advanced mediation from ordinary simplification.

这就是高级 mediation 与普通“说简单一点”的分界线。

---

Continue | 继续

The next lesson will move from adapting one idea to integrating several sources: how to synthesise evidence, preserve disagreement and build one coherent position without turning multiple texts into a list of summaries.

下一课将从“一个思想跨受众适配”推进到“多个来源综合”：怎样综合证据、保留分歧，并形成一个统一的判断，而不是把多篇材料写成摘要清单。

Next: EDKS-ADV-ZH-0011 · Lesson No.011 · Synthesize Multiple Sources Without Losing Their Differences · 综合多来源而不抹掉分歧

Back to Advanced English Chinese Edition Hub · 返回高级英语中文版主页