2016年6月·第1套 英语四级阅读理解（四选一） 第 1 篇 · 46–50 题答案与解析

What does the author say about the threat of robots?是 What...say about 句式，要求复述作者关于某话题的具体表述，对应文章某句的同义改写。

题干What does the author say about the threat of robots?是标准 wh-疑问句，主干『the author say』，about 引出讨论对象『the threat of robots』。结构平铺直叙，无陷阱。

• A)偷换概念

  It may constitute a challenge to computer programmers把作者结论『危险可避免』偷换成『对程序员是挑战』，文中没有这层意思；programmer 出现在They are dangerous only if programmers are careless，与本题作者关于威胁性质的核心判断无关。

• B)无中生有

  It accompanies all machinery involving high technology扩大到『凡是高科技都伴随该威胁』，原文只针对 AI 机器人，未做行业泛化。

• C)✓ 正确
  • if we figure out how to turn human values into a programmable code↔if human values are translated into their language
  • This danger can be avoided↔It can be avoided

  C 把『威胁可避免』加『条件是把价值观翻译成机器语言』这两层信息完整保留，是定位句的标准同义替换。

• D)与原文相反

  It has become an inevitable peril as technology gets more sophisticated说威胁是『不可避免的祸患』，与定位句This danger can be avoided直接相反。

What would we think of a person who invades our personal space according to the author?是 What would we think of... according to the author，问的是作者文中给出的具体判断结论。

题干主干『What would we think of a person』，定语从句『who invades our personal space』修饰 a person，『according to the author』是状语限定信息来源。结构清晰、无否定无比较。

• A)过度推断

  They are aggressive从『侵犯私人空间』联想到『具有攻击性』，但原文只评判『教养』维度，未涉及攻击性。

• B)与原文相反

  They are outgoing意为『外向开朗』，是褒义词，与原文负面评判（不是有教养的人会做）相反。

• C)偷换概念

  They are ignorant意为『无知或不知道』，强调『信息缺失』；原文是『没教养』，强调『礼仪修养缺失』。两者维度不同。

• D)✓ 正确
  • you wouldn't think that's the kind of thing a properly brought-up person would do↔They are ill-bred

  把原文含蓄否定的『不是有教养的人会做的』翻成肯定式『是没教养的人做的』，再用 ill-bred 一词覆盖，结构和语义都对齐。

How do robots learn human values?问的是机器人学习价值观的客观方式，文中必有一句直陈手段，属于事实细节复述。

题干是简单 How 疑问句，主语 robots、谓语 learn、宾语 human values。无定语从句、无比较，结构最简。

• A)无中生有

  By interacting with humans in everyday life situations说的是『日常生活中与人类互动』，原文未提互动方式；原文只说从数据中提取规律。

• B)偷换概念

  By following the daily routines of civilized human beings说『跟随文明人日常生活作息』，与『数据提取』无关；原文从未提『日常作息』这一方式。

• C)✓ 正确
  • drawing patterns from large sets of data on human behavior↔By picking up patterns from massive data on human behavior

  C 完整保留『提取规律』加『大量人类行为数据』两个核心成分，是定位句方式短语的标准同义替换，无信息漂移。

• D)偷换概念

  By imitating the behavior of properly brought-up human beings说『模仿有教养者的行为』，是 imitating（模仿）而非 drawing patterns（提取规律）；前者是行为复刻，后者是规律抽象，机制完全不同。

What will a well-programmed robot do when facing an unusual situation?问『会做什么』，答案是文中给出的具体动作，是典型行为细节题。

主干『What will a well-programmed robot do』是 will 引导的将来疑问句，『when facing an unusual situation』是 when 引导的时间状语，分词短语 facing 省略了『it is』。结构清楚、无陷阱。

• A)张冠李戴

  Keep a distance from possible dangers中的 distance 来自第4段keep a comfortable distance from humans，但那是讨论『人际距离』，不是异常情况的应对动作。

• B)✓ 正确
  • check the correct course of action with a human↔Stop to seek advice from a human being
  • stop, send out beeps, and ask for directions from a human↔Stop to seek advice from a human being

  P8给原则、P9给动作细节，二者都指向『停下来问人』这一核心动作，B 精准覆盖。

• C)偷换概念

  Trigger its built-in alarm system at once说『立即触发警报系统』，但原文send out beeps是『发出嘟嘟声向人请示』，是『咨询信号』而非『警报』；且 at once 也属添加。

• D)张冠李戴

  Do sufficient testing before taking action中的 testing 来自fail to do sufficient testing，但那是『人类未做够测试』导致风险，不是『机器人面对异常时的动作』。

What is most difficult to do when we turn human values into a programmable code?要求复述文中明确指出的最难之处，答案在原文有直陈句。

主干『What is most difficult to do』，不定式『to do』补足语义；when 状语从句『when we turn human values into a programmable code』中 turn...into... 是固定搭配『把…转化为…』。结构稍长但层次清楚。

• A)✓ 正确
  • deciding exactly what we believe is moral, and how to create a set of ethical rules↔Determine what is moral and ethical

  A 把『判定道德标准』和『建立伦理规则』两层信息浓缩成『determine what is moral and ethical』，动词与对象都对齐定位句，是合规的同义替换。

• B)无中生有

  Design some large-scale experiments提到的『设计大规模实验』在原文不存在；原文只提『测试』（testing），且未限定规模。

• C)张冠李戴

  Set rules for man-machine interaction中的『人机交互规则』借if you were talking to another person and they came up close in your personal space与program a robot to check the correct course of action with a human的场景诱骗，但末段说的『最难』是『定义道德』，不是『定人机交互规则』。

• D)偷换概念

  Develop a more sophisticated program说『开发更精密的程序』，借As Artificial Intelligence (AI) becomes increasingly sophisticated的 sophisticated 一词诱导，但末段说的『最难』是『决定何为道德』，不是『开发程序』。