什么叫"学术语体"
很多中国学生把学术写作误解为"用大词、用复杂句"。事实正好相反：学术写作的核心是精确、克制、可验证。它的"正式"不在于词汇华丽，而在于：
- 不用主观情绪词。
- 每一个判断都尽量基于证据或引用。
- 表达限定明确，不用绝对化语言（"all/never/always" 慎用）。
- 读者主体是同行专家，假设读者已经熟悉基本概念。
学术语体的七条规范
1. 不用缩写
非正式
The data doesn't support this hypothesis, and we can't reject the null.
学术
The data do not support this hypothesis, and the null cannot be rejected.
注意 data 在严格学术语境下是复数，所以是 data do not，不是 data does not。理科和社科论文里这个细节很被看重。
2. 不用俚语、口语、非正式连接词
| 非正式 | 学术替换 |
|---|---|
| kids | children |
| a lot of | numerous / considerable |
| so | therefore / thus / consequently |
| also (句首) | moreover / furthermore / in addition |
| but (句首) | however / nevertheless |
| get | obtain / acquire / receive |
| show | demonstrate / indicate / suggest |
| think | argue / contend / hold / maintain |
| good | effective / beneficial / robust |
| bad | detrimental / suboptimal / problematic |
3. 第一人称的边界
这一点常被简化为"学术写作不能用 I/We"，但事实更微妙：
- 人文学科 / 社会科学论证文：可以适度使用
I argue / In this paper, I propose，但限于结构性宣告。 - 自然科学 / 工程实证论文：传统上避免第一人称，用被动或物作主语：
We collected the samples→The samples were collected。但近 20 年许多顶级期刊（Nature、Science）已经允许We measured...。 - 本科论文 / 课程作业：除非教授明确允许，避免
I think / In my opinion，改用It can be argued / The evidence suggests。
第一人称的真正问题不是 "I"，而是 "I think / I believe / I feel" 这种把判断变成主观感受的表达。学术读者关心的是证据，不是你的感受。
4. 减弱语 Hedging（关键技能）
这是中国学生最容易忽略的学术规范。学术写作不喜欢绝对化，几乎所有判断都需要加一个减弱词，留出空间。
| 类型 | 常用 hedging 词 |
|---|---|
| 动词减弱 | suggest, indicate, appear to, tend to, may, might |
| 副词减弱 | likely, possibly, probably, generally, often, in some cases |
| 形容词减弱 | likely, possible, probable, partial, partial-evidence |
| 名词减弱 | tendency, possibility, likelihood, indication |
过强
This proves that social media causes depression in teenagers.
恰当
These findings suggest that intensive social media use may be associated with elevated depression risk in adolescents.
5. 用名物化和被动平衡客观
第二章我们说过避免名物化和被动滥用——那是普通写作的原则。在学术语境里，名物化和被动恰恰是表达客观性的工具。
过于主观
I found that the policy reduced unemployment by 12%.
学术客观
The policy was associated with a 12% reduction in unemployment.
但仍然要避免"层层嵌套的名物化"，比如 The implementation of the regulation of...。学术不是堆砌，是精准。
6. 避免性别化、文化中心、价值判断词
mankind→humanity / humankind / peoplechairman→chair / chairpersonpoliceman→police officerthird-world countries→developing countries / low-income countriesobviously / clearly / of course→ 多数情况下应删除
7. 一致性高于"创意"
同一个概念在全文里要用同一个名词，不要为了避免重复而换词。比如全文都用 participants，不要一会儿 subjects 一会儿 respondents。术语稳定 = 阅读顺畅。
引用与参考文献：APA / MLA / Chicago
学术写作里不引用就是抄袭。哪怕你只是用了别人的观点，没用原话，也必须标注。三种主流引用规范：
| 规范 | 主要领域 | 正文内引用 | 参考文献样式 |
|---|---|---|---|
| APA (7th) | 心理学、教育、社科 | (Smith, 2020) | Smith, J. (2020). Title. Publisher. |
| MLA (9th) | 文学、人文 | (Smith 23) | Smith, John. Title. Publisher, 2020. |
| Chicago | 历史、艺术、出版业 | 脚注或 (Smith 2020, 23) | Smith, John. Title. Chicago: Publisher, 2020. |
APA 7 正文内引用模板
1 个作者: (Smith, 2020)
Smith (2020) found that ...
2 个作者: (Smith & Lee, 2020)
Smith and Lee (2020) argued ...
3+ 个作者: (Smith et al., 2020)
引用页码: (Smith, 2020, p. 45)
(Smith, 2020, pp. 45-47)
无作者: ("Title", 2020)
机构作者: (World Health Organization [WHO], 2020)
后续引用: (WHO, 2020)MLA 9 正文内引用
(Smith 23) ← 注意没有逗号、没有 p.
Smith argues that ... (23).
("Title" 23) ← 无作者用标题简写引用方式三种
- 直接引用 Quotation：原话用引号，配页码。超过 40 词独立成段、缩进、不加引号（block quote）。
- 转述 Paraphrase：把原话用自己的语言重写。仍需引用，但不用引号。这是学术写作里最常用的方式。
- 总结 Summary：把整篇或整章压缩成几句话，标注作者和年份。
转述不是把句子换两个同义词。你必须把原文意思用完全不同的句法重写。把一段原文先盖住，凭记忆重写，再对照——不一样就 OK。否则可能被判抄袭（patchwriting）。
文献综述（Literature Review）的结构
本科论文、硕博论文、研究 proposal 几乎都需要文献综述。它不是"把读过的论文罗列一遍"，而是梳理已有研究并指出空缺。
组织方式：三种主流
- 按主题（Thematic）：把已有研究按子主题分组讨论。最常用。
- 按时间（Chronological）：按研究发表的时间顺序梳理，看一个领域如何演变。
- 按方法（Methodological）：按研究方法分组（实验研究 / 田野调查 / 元分析 / 理论建模）。
文献综述的标准段落模板
(1) Subtopic claim / question
Research on X has clustered around three questions: A, B, and C.
(2) Group authors who agree
Several scholars argue that ... (Smith, 2018; Lee, 2019; Park, 2020).
(3) Group authors who disagree
Others contend that ... (Wang, 2021; Garcia, 2022).
(4) Synthesis
Together, these studies show convergence on Y, but disagreement
persists on Z.
(5) Gap
Despite this volume of work, no study has yet examined how
[your specific angle] interacts with [your variable], leaving
a gap that the present paper aims to address.第五步"指出 gap"是文献综述真正的关键。这一句话直接决定你的研究有没有意义。
学术连接词清单（按论证功能）
| 功能 | 常用学术连接词 |
|---|---|
| 引用前人观点 | According to X (2020), ... / X (2020) argues that... / As X (2020) notes, ... |
| 支持 / 强化 | furthermore, moreover, in addition, likewise, similarly, equally important |
| 反驳 / 转折 | however, nevertheless, nonetheless, yet, conversely, by contrast, on the contrary |
| 因果 | therefore, thus, hence, consequently, accordingly, as a result |
| 举例 | for instance, for example, to illustrate, namely, specifically, in particular |
| 限定 hedging | arguably, presumably, ostensibly, in principle, in part, to some extent |
| 承认 concession | admittedly, granted, while it is true that, though, despite |
| 结论 | in conclusion, in sum, overall, taken together, on balance |
| 研究空缺 | however, less attention has been paid to / remains underexplored / has yet to be examined |
IMRaD 论文结构
实证学科（理工、医学、心理、经济、教育）几乎所有论文都遵循 IMRaD 结构：
| 部分 | 含义 | 核心任务 | 主时态 |
|---|---|---|---|
| Introduction | 引言 | 背景、文献、研究问题、Hypothesis | 现在时为主 |
| Methods | 方法 | 样本、工具、流程，能让人复现 | 过去时 |
| Results | 结果 | 客观陈述数据，不解读 | 过去时 |
| Discussion | 讨论 | 解读、对比文献、局限、意义 | 现在时为主 |
注意 Methods 和 Results 用过去时（你做过的事），Discussion 用现在时为主（一般性结论），这是论文写作里最常被搞错的时态规则。
一段标准学术段落示例
Several studies suggest that workplace flexibility may
improve employee retention, although the magnitude of the
effect varies considerably across industries (Smith, 2019;
Lee & Park, 2021). In a 12-month panel of 4,500 knowledge
workers, Smith (2019) reported a 14% reduction in voluntary
turnover among employees with hybrid schedules, controlling
for tenure and compensation. However, this effect appears
weaker in roles requiring frequent client interaction
(Lee & Park, 2021), indicating that the relationship is
moderated by job characteristics rather than uniform.
The present study extends this literature by examining how
team-level cohesion mediates the link between flexibility
arrangements and individual retention—a mechanism that has
received limited empirical attention.注意这段里出现的所有学术写作要素：(1) 多重引用，(2) hedging 用 may / appears，(3) 数据具体到百分比和样本量，(4) 转折用 However，(5) 最后一句指出 gap 并预告本文贡献。
找一篇你领域里的顶级期刊论文，把它的 Introduction 段逐句拆解：哪一句是 Hook？哪一句开始引用文献？哪一句指出 gap？哪一句给出 Hypothesis？这种"逆向解剖"做 5 篇，你的学术段落写作会发生质变。