你其实早就这么做过

本文所有配图均由 AI 生成。

韧性是心智的疫苗。想想疫苗是怎么起作用的：身体先暴露在极少量的病毒面前，于是启动免疫反应；等日后真正被病毒感染时，身体早已准备好，能把它迅速清除。

—— S. Nassir Ghaemi，《A First-Rate Madness》

你按住女儿的腿，护士正在准备针头。她才六个月大。你知道接下来要发生什么，她不知道。你知道她会尖叫，而你不会阻止。

她真的尖叫起来。不是抽泣，是真正的尖叫。眼泪顺着两边脸颊往下淌。你的手稳稳地放在那里，没有挪开。

你知道自己在做什么。疫苗之所以有效，是因为免疫系统必须先见到病原体，而且得是在受控剂量下见到，然后才有能力在真正的、全量的病原体来临时挡住它。¹ 没有那一剂，免疫系统对这个威胁就毫无记忆。等真正的病来到时，它手里什么都没有。所以你按住她的腿。她尖叫。这正是关键。

你开车回家，给她吃了 Tylenol。她没事了。

五年后，她哭着从学校回家，因为 Emma 说她画得不怎么样。你一看见她的脸，就感到那股熟悉的拉力涌上来。

你告诉她，她的画很好看。你告诉她，Emma 只是太刻薄。你很快把这些话一股脑说出来，因为看她哭让你受不了，而你只想让哭声停下。

她慢慢平静下来，回了自己的房间。你回到厨房。

你还不知道，刚才你依据的原则，已经和诊室里那一次不一样了。在诊室里，你按住她，是为了让她接受一点不舒服的东西。在厨房里，你立刻冲上去，是为了让她不必承受那一点不舒服。在诊室里，你是在给她一剂量；在厨房里，你是在把这一剂拿走。

同一个家长。两套原则。你没注意到这个差别。也还没人让你看见，这里其实确实有一个差别。

这篇文章要讲的，就是这个。

1. 那次转向原本是有道理的#

你认识的某个人，多半还是你的亲人，心里一直藏着一幕童年场景，却从未完整对你讲出来。某个老师在全班面前羞辱了他。某个教练让他一连几个月都觉得自己渺小。某个父亲把收回亲近当成一种控制手段。如今他们都五十多岁了，老板一给反馈，他们还是会本能地缩一下。他们不会把这种退缩和当年的教室联系起来，他们只会觉得，自己就是不擅长面对反馈。

他们不是不擅长面对反馈。他们是被教会了把批评理解成武器。那个老师，那个教练，那个父亲，确实都在上课，只不过他们上的，不是自己以为的那一课。

这就是那场疗愈式转向想要替换掉的东西。不只是偶尔一句重话，而是一整套系统。饭桌上的沉默。那个说出“你投球像个女孩”而引得全场大笑、却没人阻止的教练。那个从来没说过一次“我为你骄傲”的父亲。那个嘴上说骄傲、接着又把你做错的每件事全数列出来的母亲。羞耻像餐桌上的一只盘子，被一圈圈传递，每个人都得从上面拿一份。

1986 年，加州通过一项法律，设立专门工作组来推动自尊教育。² 不到十年，“孩子需要对自己感觉良好” 这件事，就成了全美学校和家庭客厅里的默认共识。这不是阴谋，而是一场救援。

而且它要把孩子们从某种真实存在的伤害里救出来。

后来有两位研究者把账算清楚了。Gershoff 和 Grogan-Kaylor。七十五项研究，十六万个孩子，所有能测量到的效应都算进去。³ 其中 71% 指向伤害；那些大到足以让人确信其存在的效应里，99% 指向伤害。

那就是疗愈式转向当时要回应的现实。不是一套哲学，而是一份伤亡统计。

但体罚只是这整套东西里最看得见的那一层。另一项研究，也是同类研究中规模最大的一项，调查了 9500 名成年人，追问不同类型的童年困境最终在他们长大后留下了什么。⁴ 研究者统计的不只是挨打。清单里还有情感虐待、情感忽视、看见父母醉酒、看见父母被铐走、以及在一个从来没人温柔地喊过你名字的家里长大。那些有过四种及以上这类经历的孩子，长大后陷入抑郁、药物滥用或自杀念头的概率，是其他人的四到十二倍。这不是小效应。这是发生在普通人家、紧闭房门之后、持续好几代的公共健康灾难。

那些最早推动改变的父母，对自己想替换掉的旧东西，看得完全正确。没有疑问。

2. 然后我们矫枉过正#

你九岁的孩子坐在厨房餐桌前写数学作业，面前是一道她做不出来的题。你看着她，又感觉那股拉力上来了。你知道，因为你读过文章、听过播客、记住过儿科医生的话，所以“正确做法”应该是陪着她的情绪，而“错误做法”则是说一句“你再试试看”。于是你说：“这题很难，觉得沮丧也没关系。”

她哭得更厉害了。你陪着她。一个小时以后，作业还没做完。她回房间去了。你在厨房里收拾东西，不太明白刚才到底发生了什么。你明明把那些文章教你的事都做了。刚才发生的，正是这个。

真正发生的是：你把那一剂拿走了。

关于表扬的研究，具体得近乎刺人。Dweck 和 Mueller 找来 128 名五年级学生，连续做了六组实验。每个孩子先做一道题。做出来以后，其中一半听到的是“你一定很聪明”，另一半听到的是“你一定很努力”。⁵ 接着进入下一轮：另一道题，难一点还是容易一点，由孩子自己选。

被夸“聪明”的孩子，选了更简单的题。他们躲开一切可能检验自己的东西。被夸“努力”的孩子，选了更难的题。他们想试试看自己能不能应付。

一句话在建构信心，另一句话却砌起一堵墙，一堵把孩子和任何可能推翻“你就是这样的人”这种身份叙述隔开的墙。

Dweck 后来专门写过一篇给教育从业者看的文章，标题非常直白：《小心，表扬可能有危险》。⁶ 那是 1999 年。这种警告并没有广泛传播。对表扬的警告，没有表扬本身那么容易流传。父母们从 Dweck 的研究里吸收进去的，是“要夸努力，不要夸结果”；他们实际执行出来的，却更接近“随时随地都要夸”。这两者不是一回事。前者是手术刀，后者是消防水龙头。

问题比表扬还要更深。问题是摩擦，是难度，是那种“我现在还不知道答案”的经验。

研究者后来试了另一件事：先把一道难题摆在学生面前，在教任何东西之前。没有讲解，只有题目。结果他们的表现，反而比先听讲再做题的学生更好。⁷ 五十三项研究，一万两千个孩子，同样的模式一再出现。

机理并不复杂。挣扎会先把心智调动起来。它把学生已经知道的一切都调出来，对准眼前这道题。等讲解真正来到时，它落进的是一个已经预先成形的槽位里。落在没有准备好的地面上，讲解只会漂着；落在准备好的地面上，它才会真正嵌进去。

这种效应在年龄更大的孩子身上最明显，中学以上尤其如此。但它在各个年龄层、各个学科里都成立。

所以，厨房那一个小时里，你陪着她的情绪，作业却毫无进展，那不是善意。那是抽走。真正有作用的，本来就是那份挣扎；你在它来得及发挥作用之前，就把它抹平了。

一棵长在无风温室里的树，会长得又高又细。木质不会变密，根系也不会向外展开。从外面看，它似乎没问题。拿走负荷，就会发生这种事。⁸ 树会朝着错误的形状生长。在温室里它看起来一切正常，到了外面就是另一回事。

骨头也是同样的道理。经常承受负荷的骨头，密度会增加；从不承受负荷的骨头，并不是保持原样，而是会失去密度。⁹ 这不是伤害，而是设计。身体会按照它所承受的要求，不断重塑自己。没有要求，并不是中性的；那也是一种选择，而且会带来后果。

我们以为自己是在替她卸下重量，实际上是在抽走信号。

Taleb 有一句话正好说到这里：“那些试图帮助我们的人，往往也是伤我们最深的人。”¹⁰ 他写的是经济系统，但这句话可以走得很远。

把这个信号放大到群体尺度，你看到的会是这样一幅图景：大约从 2012 年开始，美国青少年中的重度抑郁比例急剧上升。在 12 到 17 岁女孩中，重度抑郁发生率上升了 52%，从 2005 年的 13.1% 升到 2017 年的 19.9%。¹¹ 同一时期，10 到 12 岁女孩的自我中毒事件翻了四倍，10 到 14 岁女孩的自杀事件翻了一番。

这些数字都是真的。正因为它们是真的，我们才需要老老实实地区分：哪些东西我们知道，哪些东西我们还不知道。

你刚读到的内容里，有一部分是有争议的。Candice Odgers 在《Nature》上写过，认为“手机导致了这场危机”这个论证，并没有它最响亮的支持者说得那么扎实。¹² 她措辞很谨慎。她不是说危机不存在，而是说，我们还不知道其中到底有多大一部分该归因于手机。这个争论至今还在继续，研究者们还在吵。

但手机究竟占了 30%、60% 还是 90% 的责任，那是一个关于“该怎么处理手机”的问题。这篇文章说的是另一件事：该怎么做父母。做父母不是在和手机争夺“谁才是原因”。在整幅图景里，它是唯一一根真正握在你手里的杠杆。你今晚没法在学区董事会上把 Instagram 禁掉，但你可以决定，下次餐桌边又出现一次情绪崩溃时，你要怎么回应。

这就是这篇文章只想请你认真想一想的事。

3. 她不知道自己在哪里#

她的画贴在冰箱门上。老师说她状态很好。你每天晚上都告诉她，你为她骄傲得不得了。

但与此同时，她也开始在一些细小的时刻里问你问题，车上、睡前、灯关了以后，像这样：“你真的觉得我擅长那个吗？” 她不是在讨更多表扬。她是在确认，这些表扬到底是不是真的。她不知道自己究竟擅长什么。她手里没有地图。

从她记事起，她就一直被告知自己很棒。她开始怀疑，大人只是在管理她的情绪。

2017 年的一项研究追踪了 120 对亲子，孩子年龄在 7 到 11 岁之间。¹³ 它记录父母给出“夸大型表扬”的频率。所谓夸大型表扬，不是说“这画很好看”，而是说“这不只是好看，这简直特别特别好看”。随着时间推移，那些最常听到这种夸大型表扬的孩子，自尊水平反而更低，而不是更高。论文自己的总结是：“夸大型表扬，可能会促成它原本想防止的那种自我看法。” Po Bronson 很早以前就把这件事说得更直白：“我们对孩子期待很高，却把这些期待藏在持续不断、亮得发白的表扬后面。”¹⁴

这些表扬原本是为了把孩子抬高，结果却把她压低了。机理也不难理解：夸大型表扬设定了一个孩子自己知道根本达不到的标准。她听见“特别特别好看”，心里其实知道，至少在语言尚未成形的那个层面上知道，这幅画并没有“特别特别好看”。当她低头看向纸面时，她看见的东西，和你告诉她的东西对不上。于是表扬并没有沉淀成信心，而是沉淀成一个裂口：真实的自己，和被要求成为的那个自己之间的裂口。

同一批研究者更早之前还有一项研究，结论更难让人忘掉。¹⁵ 父母最容易给出夸大型表扬的，正是那些挣扎得最厉害的孩子。最需要得到准确反馈、最需要知道“自己现在能做到什么、还做不到什么”的孩子，反而得到最不准确的信息。那个对自己的画没把握的孩子，会被告知她画得非同凡响。那个知道自己数学不稳的孩子，会被告知她特别聪明。表扬正好落在最无益、也最伤人的地方。

Alfie Kohn 几十年前就警告过这种机制。他写道，过量的、笼统的正面表扬，会把孩子训练成一种人：无论做什么，都先把“自我”变成问题本身，于是他们会同时更容易走向自大和自我厌弃。¹⁶

而这场运动最初的承诺，即“高自尊会让孩子表现更好、关系更好、生活更好”，后来也没有站住。到了 2000 年代初，那些当年推动自尊运动的研究者自己也遇到了问题：当他们试图寻找“高自尊确实会导致更好的学业表现、更强的人际关系、更好的行为表现”这些承诺的客观证据时，他们找不到。¹⁷ 他们最终能确认的益处，真实但有限：高自尊的人通常主观上更快乐，也更愿意主动出手做事。这当然不是什么都没有，但它远不是当年承诺的全部，甚至连大部分都算不上。

这里说的是大学生，不是九岁的孩子，但底下的机制其实完全一样。把镜头拉到她上大学之后：那些被父母不断代劳的年轻人，父母替他们打电话、替他们争成绩、替他们处理室友冲突，最后长成了更不确定、更焦虑、也更不会处理成人生活的一群人。五十三项研究，四万六千名大学年龄段的年轻人。¹⁸

你要是一直替他们把难题解决掉，他们就永远学不会：原来自己也能把难题解决掉。Jonathan Haidt 把更大的观察压缩成了一句话：“我们在现实世界里，对孩子进行了远远过度、而且根本没有必要的保护。”¹⁹

你之所以伸手，是因为看她挣扎让你受不了。情绪崩溃让你受不了。朋友的一句冷落让你受不了。没做完的作业、没有收到的邀请、一个出乎意料的分数，这些都让你受不了。伸手并不是爱出了问题，恰恰相反，那就是爱在你心里的感觉。但与此同时，它也在教她：她的不适，是你的问题；而你的任务，是把它解决掉。

你九岁的女儿之所以会问“你真的觉得我擅长那个吗”，是因为她手里没有足够真实的信息，无法自己回答这个问题。她拥有的只有表扬，而她已经开始怀疑，表扬和真相不是一回事。

她真正想要的，其实是一个她能靠来导航的信号，一张关于“我是什么、我不是什么；我会什么、我还不会什么”的地图。你没法靠反复告诉她“一切都很好”把这张地图交给她。你只能靠诚实地指出那些并不好的部分，把地图交给她。

而那正是我们拿走的东西。

4. 更清楚，而不是更强硬#

纽黑文有一位母亲，带着自己 12 岁的女儿去耶鲁看专科医生。女孩有惊恐发作，已经连续六个月没有整夜睡好，也不肯去学校。那位专科医生做了一件不同寻常的事：他不急着见女儿，他想先和母亲工作。

耶鲁在 2020 年发表过一项随机试验，测试一种名叫 SPACE 的治疗方法，全称是 Supportive Parenting for Anxious Childhood Emotions。²⁰ 研究对象是 124 名 7 到 14 岁、被诊断为焦虑障碍的孩子。其中一半接受标准治疗，也就是让孩子参加认知行为疗法；另一半则完全不同：治疗师根本不治疗孩子，治疗师治疗的是父母。

父母学着停止替孩子吸收那些回避行为。不要为了避免分离焦虑就睡进孩子房间。不要在晚餐桌上替一个社交焦虑的孩子讲话。不要因为孩子今天很难，就立刻给老师打电话去把路铺平。相反，他们学会说，哪怕不是逐字逐句，也大致是这个意思：我相信你能应付这件事，我会在这里。

结果是，SPACE 组的效果并不比 CBT 组差。两种治疗都显著降低了孩子的焦虑水平。而且 SPACE 组父母在研究者所谓的 “accommodation” 上，也就是父母替孩子吸收痛苦、并把引发痛苦的事物提前移开的那些行为上，下降幅度远远大于 CBT 组。研究识别出的机制是：当父母不再迁就性地替孩子绕开焦虑时，孩子也就不再收到“你的焦虑是正确的，而且危险是真实的”这种信号。他们会收到另一种信号：我相信你能处理它。

这种信号不是更强硬，而是更清楚。

关键的词，不是“更强硬”，而是“更清楚”。

如果你的孩子被诊断为焦虑障碍，这项研究说的就是她。即便你的孩子只是像所有九岁孩子那样，偶尔会焦虑，偶尔崩溃，偶尔某天不想上学，偶尔一道作业题写着写着就哭起来，这条原则也仍然成立。

Alison Gopnik 提供过一个很有用的比喻。有些孩子像蒲公英，在哪儿都能长；有些孩子像兰花，环境好的时候开得格外好，环境差的时候也格外容易受损。²¹ 但无论哪一种，都不适合活在一个完全没有天气的世界里。

无论如何，这背后都有一项临床试验作支撑。那套有效的治疗，教给父母的，正是这篇文章一直在说的事情：别再把路全都抹平，发出那个清楚的信号，让孩子知道自己能处理。它的效果，和把孩子送去每周一次的 CBT 治疗一样好。这是耶鲁做的研究。证据就摆在那里。关键词是“更清楚”，不是“更强硬”。

这篇文章里没有任何一句话是在替严厉辩护。那个拿着针的护士，并不严厉。她只是清楚。她知道剂量是多少，也知道为什么那一点剂量重要，所以她不会退缩。这不是残忍，而是一种极其精准的爱。

就连一向以温和养育著称的 Janet Lansbury 也把这话说得很明白：“没有纪律，不是善良，而是忽视。”²² 她说的不是惩罚，而是那种稳定、清晰、可靠的引导形状。

你的女儿并不需要你变成另一个家长。她也不需要你少爱她一点。她需要的是一个更清楚的信号：她现在的不适不会永远持续，她没有坏掉，她确实能把这件事扛过去。

所以下一次，当她又坐在厨房餐桌前，作业太难，脸一下子涨红的时候，先留意那只准备伸出去的手。

这不是什么全新的育儿哲学，你也不必变成另一个人。下一次，当你的手又想伸出去，想把孩子本来自己能扛住的事抹平时，可能是那次落空的邀请，可能是那道她放弃了的作业题，可能是朋友说的那句话，先注意到这只手。然后，只让一件很小的事保留下来。只一件。不是一套计划，不是一个长期承诺，就是周二的一件小事。这就是全部处方。

那个拿针的护士，并不比你更强硬。她只是更清楚。

5. 门还开着#

她 27 岁了，这个月已经第三次把车停在办公室外面给你打电话。经理给了她一些反馈。你如果只是从旁听来，大概会觉得那反馈算得上合理。但她把它讲成一场不公。你也同意她，因为你一直都同意她。

你挂了电话，站在厨房里。十八年过去了，厨房一点都没变。

然后你安静地意识到，你在她的声音里听见了自己。不是她的具体措辞，她有她自己的语言；是那种反应的形状。那种本能：只要感到不舒服，就一定说明哪里出了错；一定是有人对你做了什么；这份困难不该落在你身上；你本来应该得到更好的。

这是你教给她的。你是出于爱才这样教的。爱是真的，账也还是到了。

这不是在给过去下判词。爱从来不是问题，而过去也不是你现在居住的地方。现在的问题，是下一次。

因为下一次还会来。下一次情绪崩溃。下一次数学题做不出来。下一次朋友说了什么。下一次作业没写完。下一次成绩和预期不一样。这些时刻都不在过去，它们还会准时抵达，差不多总在某个周二晚上六点。

在其中一些时刻里，你还是会伸手。有些事你还是会替她抹平。这没有关系。剂量不需要每次都有。疫苗接种日程从来都不是“每时每刻都来一下”。它是校准过的，和年龄相符的，也是具体的。

回到最开始。你的孩子六个月大。你按着她的腿。针扎下去。她尖叫。你没有阻止。因为你知道，你其实一直都知道：一点点、可控的、她眼下还承受不了的东西，正是她最终成长为能够承受之人的方式。你在诊室里没有阻止这一切，但你后来在厨房里一再阻止了它。

离她学会“针头并不等于危险”的那一课，其实只隔着一间屋子。离她学会“短促而尖锐的东西也可能让你更强壮”的那一课，也只隔着一间屋子。离你原本就懂得这一点，同样只隔着一间屋子。

厨房还在，下一次时刻也还会来。下次当你的手又伸出去的时候，先注意到它。你不必每次都把手缩回来。只要有一次。只要让一件小事留在那里。

门还开着。

参考资料#

延伸阅读#

• Dweck, C.S. Mindset: The New Psychology of Success. Random House, 2006. 关于成长型思维，以及“夸智力”和“夸努力”之差异的核心来源；也是理解本文中心机制最直接的背景材料。
• Haidt, J. The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness. Penguin Press, 2024. 当下最有影响力的青少年心理健康恶化论述之一；它与本文提到的 Odgers 反驳形成直接对照。本文没有在“手机是否是主因”上选边，但理解这场争论时，这本书很重要。
• Kapur, M. "Productive Failure in Mathematical Problem Solving." Instructional Science, 38(6), 523–550, 2010. 本文所引“先挣扎、后讲解”研究路径的奠基论文；如果想深入理解为什么“先卡住”反而会帮助学习，应从这里读起。
• Lebowitz, E.R. Breaking Free of Child Anxiety and OCD: A Scientifically Proven Program for Parents. Oxford University Press, 2021. 对 SPACE 方法面向家长的完整说明；如果你最关心的是“具体该怎么做”，这本书比本文更实操。
• Lythcott-Haims, J. How to Raise an Adult: Break Free of the Overparenting Trap and Prepare Your Kid for Success. Henry Holt, 2015. 关于过度养育及其后果的通俗长篇论述；尤其适合青春期后段到成年初期孩子的父母继续往下看。

You Have Already Done This

All images in this article were generated with AI.

Resilience is the mind's vaccine. Think how vaccines work: exposed to tiny amounts of virus, the body mounts an immune response; when the virus infects one later in life, the body, already prepared, knocks it out.

— S. Nassir Ghaemi, A First-Rate Madness

You hold your daughter's leg while the nurse preps the needle. She is six months old. You know what is about to happen. She doesn't. You know she will scream and you will not stop it.

She screams. Actual screaming. Tears down both cheeks. You keep your hands exactly where they are.

You know what you are doing. The vaccine works because the immune system has to meet the pathogen — at a controlled dose — before the full-strength version arrives.¹ Without that dose, the immune system has no memory of the threat. When the real thing comes, it has nothing to work with. So you hold her leg. She screams. That is the point.

You drive home. You give her Tylenol. She is fine.

Five years later, she comes home from school crying because Emma said her drawing was not very good. You feel the pull the moment you see her face.

You tell her the drawing is beautiful. You tell her Emma was being mean. You say all of it quickly, because watching her cry is unbearable and you want the crying to stop.

She calms down. She goes to her room. You go back to the kitchen.

You do not know, yet, that you have just applied a different principle than the one at the doctor's office. At the doctor's office, you held her still so she could receive something uncomfortable. In the kitchen, you moved in quickly so she would not have to. At the doctor's office, you were giving her a dose. In the kitchen, you removed one.

Same parent. Different principle. You did not notice the difference. Nobody has shown you, yet, that there was one.

That is what this is about.

1. The Turn Made Sense#

Someone you know — probably someone you are related to — carries a moment from their childhood they have never fully told you about. A teacher who embarrassed them in front of a classroom. A coach who made them feel small for months. A father who withheld affection as a control mechanism. They are in their fifties now and they still flinch when a boss gives them feedback. They do not connect the flinch to that classroom. They just think they are bad at feedback.

They are not bad at feedback. They were taught that criticism was a weapon. The teacher, the coach, the father — they were teaching a lesson all right. Just not the one they thought.

This is what the therapeutic turn was replacing. Not merely some occasional harsh word, but a whole system. Silence at the dinner table. The coach who said "you throw like a girl" and everyone laughed and nobody made it stop. The father who never said he was proud of you, not once. The mother who said she was proud and then listed everything you had done wrong. The shame that moved around the dinner table like a plate everybody had to take from.

In 1986, California passed a law creating a task force to promote self-esteem.² Within a decade, the idea that children needed to feel good about themselves had become the default in schools and living rooms across the country. It was not a conspiracy. It was a rescue.

And it was rescuing children from something real.

Two researchers eventually did the math. Gershoff and Grogan-Kaylor. Seventy-five studies, a hundred and sixty thousand children, every effect they could measure.³ Seventy-one percent pointed at harm. Of the effects big enough to be sure of — ninety-nine percent.

That is the regime the therapeutic turn was responding to. Not a philosophy. A body count.

But physical punishment was only the visible edge of it. A separate study — the largest of its kind — looked at nine thousand five hundred adults and asked what different kinds of childhood hardship had done to the grown-ups they became.⁴ The researchers did not only count beatings. The list included emotional abuse. Emotional neglect. Watching a parent drunk or in handcuffs. Growing up in a house where no one ever said your name with kindness. Children who had four or more of those experiences — four or more — were four to twelve times more likely as adults to struggle with depression, substance abuse, or suicidal thoughts. That is not a small effect. That is a public health catastrophe, happening in ordinary houses, behind closed doors, for generations.

The parents who started the movement to change this were right about what they were replacing. Full stop.

2. And Then We Overcorrected#

Your nine-year-old is at the kitchen table with her math homework. A problem she cannot do. You watch her. You feel the pull. You know — because you have read the article, listened to the podcast, absorbed the instruction from your pediatrician — that the right move is to sit with her feelings. The wrong move is to say "just try it." So you say: "This is hard, and it's okay to feel frustrated."

She cries harder. You sit with her. An hour later the homework is undone. She is in her room. You are cleaning the kitchen. You are not sure what just happened. You did everything the articles told you to do. That is what just happened.

What happened is that you removed the dose.

The research on praise is specific enough to sting. Dweck and Mueller ran six studies on a hundred and twenty-eight fifth graders. They gave each child a problem. When the child solved it, half got "you must be smart." The other half got "you must have worked hard."⁵ Then came the next round — another problem, harder or easier, the child's choice.

The "smart" kids chose easier. They avoided what might test them. The "worked hard" kids chose harder. They wanted to test themselves.

One phrase built confidence. The other built a wall — a wall between the child and anything that might disprove the thing she was told she was.

Dweck wrote about this in a practitioner article with the blunt title "Caution — Praise Can Be Dangerous."⁶ This was 1999. The caution was not widely heard. Warnings about praise do not travel the way praise itself does. What parents absorbed from Dweck's work was "praise effort, not results." What they applied was something closer to "praise constantly." Those are not the same thing. One is a scalpel. The other is a firehose.

The problem runs deeper than praise. It is friction. Difficulty. The experience of not knowing the answer yet.

Researchers tried something. Give a student a hard problem before you teach them anything. No instruction. Just the problem. They do better than the student who got taught first.⁷ Fifty-three studies. Twelve thousand kids. Same pattern.

The mechanism is simple. Struggle primes the mind. It pulls up everything the student already knows and points it at the problem. When instruction arrives, it lands in a slot that's already been shaped for it. Instruction on unprepared ground floats. Instruction on prepared ground fits.

Strongest for older kids — middle school and up. But it holds across ages and subjects.

The hour at the kitchen table, sitting with her feelings while the homework stayed undone — that was not a kindness. It was a removal. The struggle was the thing. You smoothed it away before it could do its work.

A tree grown in a greenhouse without wind grows tall and thin. The wood never densifies. The roots never widen. It looks fine from the outside. That is what happens when the load is removed.⁸ The tree grows toward the wrong shape. She looks fine indoors. Outside is a different question.

Bone works the same way. Bone that is regularly loaded increases in density. Bone that is never loaded does not stay the same. It loses density.⁹ This is not an injury. It is a design feature. The body continuously remodels itself in proportion to the demands placed on it. Absence of demand is not neutral. It is a choice with consequences.

We thought we were taking the load off. What we were doing was removing the signal.

Nassim Taleb had a line for this: "those who are trying to help us are often hurting us the most."¹⁰ He was writing about economic systems. But the sentence travels.

Here is what the signal looks like at scale. Since approximately 2012, rates of major depression among American adolescents have risen sharply. Among 12-to-17-year-old girls, the rate of major depression rose fifty-two percent — from 13.1 percent in 2005 to 19.9 percent in 2017.¹¹ Self-poisoning among girls aged 10 to 12 quadrupled over the same period. Suicide among girls aged 10 to 14 doubled.

Those numbers are real. They require an honest statement about what we know and what we don't.

Some of what you just read has been argued about. Candice Odgers, writing in Nature, has said the case that phones caused this crisis is weaker than its loudest advocates claim.¹² She is careful — she does not say the crisis is not real. She says we do not yet know how much of it belongs to the phones. That debate is live. The researchers are still arguing.

But whether phones are thirty percent of the cause or sixty or ninety — that is a question about what to do about the phones. This piece is about a different question: what to do about parenting. Parenting is not competing with the phones for the cause of anything. It is the one lever in that whole picture that is actually within your reach. You cannot ban Instagram at a school board meeting tonight. You can decide how you will respond to the next meltdown at the kitchen table.

That is all this piece is asking you to consider.

3. She Doesn't Know Where She Is#

Her drawings are on the refrigerator. Her teacher says she is thriving. You tell her every night that you are so proud of her.

And also: she has started asking you things, in the small moments — in the car, at bedtime — things like "do you really think I'm good at that?" She is not fishing for more praise. She is trying to find out if the praise is real. She does not know what she is actually good at. She has no map.

She has been told she is wonderful for as long as she can remember. She suspects the adults are managing her.

A 2017 study followed a hundred and twenty parent-child pairs — children aged seven to eleven — over time.¹³ It tracked how often parents gave inflated praise: not just "that's beautiful," but "that's not just beautiful, that's incredibly beautiful." Over time, the children whose parents gave the most inflated praise ended up with lower self-esteem, not higher. The paper's own summary of what it found: "Inflated praise may foster the self-views it seeks to prevent." Po Bronson had said it more plainly years earlier: "We expect so much of them, but we hide our expectations behind constant glowing praise."¹⁴

The praise was designed to raise the child. The praise lowered her. The mechanism: inflated praise sets a standard the child knows she cannot actually meet. She hears "incredibly beautiful" and she knows — at some level below words — that the drawing is not incredibly beautiful. She cannot reconcile what she was told with what she sees when she looks at the page. So the praise doesn't land as confidence. It lands as a gap. A gap between the child she is and the child she is supposed to be.

An earlier study by the same researchers found something harder to forget.¹⁵ Parents give the most inflated praise to the children who struggle the most. The children who most need accurate information about what they can and cannot do yet — they get the least accurate information. The child who is uncertain about her drawing gets told the drawing is extraordinary. The child who knows her math is shaky gets told she is so smart. The praise goes exactly where it will do the least good and the most harm.

Alfie Kohn warned about this mechanism decades ago. Too much global positive praise, he wrote, trains children "to make their selves the issue in whatever they do, and thus to be prone to both grandiosity and self-contempt."¹⁶

And the original promise of the movement — that high self-esteem would cause children to do better, relate better, live better — it did not hold up either. By the early 2000s, the researchers who had built the self-esteem movement had a problem. When they looked for objective evidence that high self-esteem caused the things they had promised it would cause — better school performance, stronger relationships, better behavior — they could not find it.¹⁷ The benefits they could confirm were real but limited: people with high self-esteem report feeling happier, and they take more initiative. That is something. It is not everything that was promised. It is not even most of it.

This is college students, not nine-year-olds — but the mechanism underneath is the same. Look at the college-age version of her. The ones whose parents kept stepping in — made the calls, fought the grades, solved the roommate fights — came out less sure of themselves, more anxious, less able to handle being adults. Fifty-three studies. Forty-six thousand college-age kids.¹⁸

When you keep solving the hard thing, they never learn they can solve the hard thing. Jonathan Haidt put the broader observation in one sentence: "We have vastly and needlessly overprotected our children in the real world."¹⁹

You reach in because watching her struggle is unbearable for you. The meltdown is unbearable for you. The friend's slight is unbearable for you. The undone homework, the missed invitation, the grade you didn't expect — these are unbearable for you. Reaching in is not a failure of love. It is what love feels like from the inside. And also: it teaches her that her discomfort is your problem to solve.

Your nine-year-old is asking "do you really think I'm good at that?" because she does not have enough real information to answer the question herself. She has only the praise. And she suspects the praise is not the same thing as the truth.

What she is looking for — what all children are looking for, underneath the drawings on the refrigerator and the "I'm so proud of you" at bedtime — is a signal she can actually navigate by. A map of what she is and what she is not. What she can do and what she cannot do yet. You cannot give her that map by telling her everything is wonderful. You give it to her by being honest about the parts that are not.

That is the thing we removed.

4. Clearer, Not Tougher#

A mother in New Haven takes her twelve-year-old to a specialist at Yale. The daughter has panic attacks. She has not slept through the night in six months. She will not go to school. The specialist does something unusual. He does not want to see the daughter. He wants to work with the mother.

A randomized trial at Yale, published in 2020, tested a treatment called SPACE — Supportive Parenting for Anxious Childhood Emotions.²⁰ A hundred and twenty-four children, aged seven to fourteen, diagnosed with anxiety disorders. Half got the standard treatment: cognitive behavioral therapy with the child. The other half did something different. The therapist did not treat the child at all. The therapist treated the parents.

The parents learned to stop absorbing the child's avoidance. Stop sleeping in the child's room to prevent separation distress. Stop speaking for a socially anxious child at the dinner table. Stop calling the teacher to smooth out a hard day. Instead, they learned to say — in so many words: I believe you can handle this, and I am here.

The result: the SPACE group did as well as the CBT group. Both treatments reduced the children's anxiety significantly. The SPACE parents reduced what the researchers called "accommodation" — all the ways parents absorb a child's distress and move the distress-causing thing out of the way — far more than the CBT parents did. The mechanism the study identified: when parents stop accommodating, the child stops receiving the signal that their anxiety is correct and dangerous. They receive a different signal. I believe you can handle this.

That signal is not tougher. That signal is clearer.

The word is not "tougher." The word is "clearer."

If your child has diagnosed anxiety, this research is about her. If your child is anxious in the way all nine-year-olds are sometimes anxious — meltdowns, school-refusal days, the homework that ends in tears — the same principle holds.

Alison Gopnik has a useful image for the range. Some children are dandelions — they grow almost anywhere. Others are orchids — they do especially well in rich surroundings and especially badly in poor ones.²¹ Neither is served by a world without weather.

Either way, there is a clinical trial behind this. The treatment that worked by teaching parents exactly what this piece has been describing — stop smoothing the path, send the signal that the child can handle it — was as effective as sending your child to weekly CBT sessions. Yale ran it. The evidence is there. The word is clearer, not tougher.

Nothing in this piece is a defense of harshness. The nurse who holds the needle is not being harsh. She is being clear. She knows what the dose is, and she knows why it matters, and she does not flinch. That is not cruelty. That is the most loving kind of precision.

Even Janet Lansbury, herself a gentle-parenting author, has said it plainly: "Lack of discipline is not kindness, it is neglect."²² She does not mean punishment. She means the shape of reliable guidance.

Your daughter does not need you to become a different parent. She does not need you to stop loving her. She needs a clearer signal that her distress is not permanent and she is not broken and she can, in fact, carry this.

The next time at the kitchen table, when the homework is hard and her face goes red — notice the reach.

This is not a new parenting philosophy. You do not have to become a different person. The next time your hand reaches out to smooth something your child could have carried — the missed invitation, the homework she gave up on, the thing the friend said — notice the reach. Then let one small thing stand. Just one. Not a program. Not a commitment. One small thing, on a Tuesday. That is the whole prescription.

The nurse with the needle is not tougher than you. The nurse is clearer.

5. The Door Is Still Open#

She is twenty-seven and she is calling you from her car outside her office. Three times this month. Her manager gave her feedback — the kind of feedback you would describe, if you heard it secondhand, as reasonable. She is describing it to you as an injustice. You agree with her, because you have always agreed with her.

You hang up. You stand in your kitchen. The kitchen has not changed in eighteen years.

And you realize, quietly, that you are hearing yourself in her voice. Not her words — she has her own. But the shape of the response. The reflex that says: discomfort is a signal that something is wrong. Someone must have done this to you. You did not earn this difficulty. You deserve better.

You taught her that. You did it with love. The love was real. The bill came due anyway.

This is not a verdict on the past. The love was never in question, and the past is not where you live. This is a question about the next time.

Because the next time is still coming. The next meltdown. The next math problem she cannot do. The next friend who said something. The next homework she left undone. The next grade she didn't expect. These moments are not in the past. They are arriving, scheduled, at roughly six p.m. on a Tuesday.

In some of those moments, your hand is going to reach in. Some of them you will smooth. That is fine. The dose does not have to be every time. The vaccination schedule is not "every moment." It is calibrated. Age-appropriate. Specific.

Go back. Your child is six months old. You are holding her leg. The needle goes in. She screams. You do not stop it. Because you know — you have always known — that a small, controlled dose of what she cannot yet handle is how she becomes someone who can handle it. You did not stop it at the doctor's office. You have been stopping it at the kitchen table.

One room over from where she learned that the needle doesn't mean danger. One room over from where she learned that a short, sharp thing can make you stronger. One room over from where you already know this.

The kitchen is still yours. The next moment is still coming. The next time your hand reaches in — notice it. You don't have to pull back every time. Just once. Just one small thing, let it stand.

The door is still open.

References#

Further Reading#

• Dweck, C.S. Mindset: The New Psychology of Success. Random House, 2006. — The primary source on growth mindset and the intelligence-vs-effort praise distinction; essential context for E-04 and E-05. The pop-misreading of Dweck's work is the article's central mechanism; reading her directly corrects it.
• Haidt, J. The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness. Penguin Press, 2024. — The most prominent recent argument about adolescent mental health deterioration; directly adjacent to E-09 and contested by E-17 (Odgers). Not cited because the article takes no position in the phones debate; essential background for understanding why Odgers matters.
• Kapur, M. "Productive Failure in Mathematical Problem Solving." Instructional Science, 38(6), 523–550, 2010. — The foundational paper behind the meta-analysis cited as E-08; provides the mechanism for why struggle before instruction works where instruction first does not.
• Lebowitz, E.R. Breaking Free of Child Anxiety and OCD: A Scientifically Proven Program for Parents. Oxford University Press, 2021. — The SPACE protocol described in E-16, written for parent audiences. The clearest available guide to what "accommodation reduction" looks like in practice; directly answers CA-5 (the mother of the diagnosed-anxious child) in full.
• Lythcott-Haims, J. How to Raise an Adult: Break Free of the Overparenting Trap and Prepare Your Kid for Success. Henry Holt, 2015. — Contextual secondary for E-10; the most accessible book-length treatment of helicopter parenting and its outcomes in college-age adults. Good further reading for the secondary audience segment (parents of 15–22-year-olds).