卷:卷二《长安月下·关中六十夜》|主角:Marco(无古人之夜)|地点:洛川·县城出租屋|典:《论语·卫灵公》“过而不改,是谓过矣”、《左传·宣公二年》“人谁无过?过而能改,善莫大焉”|主导感官:听(电话里的呼吸、键盘、暖气片的水声)|过所进度:二字·「客」「侣」|金融-国学对应:Naming the Counterparty(把“别人”还原成一个有名有姓的人,是唯一不能外包的那道工序)↔ 过而不改,是谓过矣
段一·现身
那个号码,他找了十一天。
第八十九夜。
这一夜没有古人来。
马可后来想,这大概也是应该的。
因为这一夜要做的事,没有一个古人替得了他。
十一天前,也就是马援在茂陵原上问他那三个问题的第二天,马可开始查。
他先从最容易的那一头查起。
二〇一四年那笔债。
那不难。境外债是公开的。他在彭博的历史数据里翻了半天,他还有一个老同事帮他调了权限。
那笔债,三亿美金,五年期,二〇一九年到期。
没有违约。
按期兑付了。
马可看着那行记录,坐在那间八百块的出租屋里,忽然有一种很复杂的感觉。
松了一口气。
然后他为自己那口气,感到很难受。
他继续往下查。
那家公司现在还在。规模比十年前大了三倍。
它二〇一九年兑付那笔债的钱,是从哪儿来的?
又发了一笔新债。
马可在那儿坐了很久。
这不是他当年那个结构崩了。
这是它一直在往下滚。
他忽然想起一件事:那家公司的控股股东,二〇二一年被查了。
他搜了一下。
是的。
涉嫌违规担保和资金占用。
而那家保理公司,他当年建议把烂账卖过去的那一家,在起诉书里出现了。
马可关上电脑,在那间屋里坐到天黑。
第二笔更难。
二〇一五年那个结构化产品。一千二百多个客户。
他没有名单。
他离职十年了。
那个产品是他们组设计的,可销售是私行部做的,不同的法人主体,不同的系统。
马可给三个还在业内的老同事发了消息。
两个没回。
一个回了两个字:别问。
第四天,他换了一个思路。
他去搜诉讼。
那年那个产品爆了以后,有没有人告?
有。
他找到了七个案子。
裁判文书网上。公开的。
七份判决书。原告是七个自然人。被告是那家银行的私行部。
每一份判决书上,都有原告的名字,虽然中间两个字被隐去了。
每一份判决书上,都写着他们买了多少。
马可一份一份地看。
第一份:某某某,买了三百万。判决驳回。
第二份:某某某,买了一百五十万。判决驳回。
第三份,
他看到第五份的时候,停住了。
那一份的原告,六十七岁。
判决书里有一句话,是原告的陈述:“该笔资金系原告及配偶全部退休积蓄,用于养老。”
判决:驳回。
理由:原告已在风险揭示书上签字确认。
马可在那间出租屋里,把那句话读了很多遍。
风险揭示书。
他知道那份东西。那是他们组写的模板。
四十七页合同的最后,附一份两页的风险揭示书,客户要抄一句话:
“本人已阅读并理解上述风险,自愿承担投资损失。”
要手抄。不能打印。
为什么要手抄?
因为手抄了,将来打官司,那就是铁证。
马可坐在那儿,忽然觉得胃里翻上来一股东西。
那句话是他们组的一个律师加进去的。
加的时候,所有人都觉得很聪明。
马可记得那次会。
在三十七层的一个小会议室。那个律师姓陈,是个很年轻的人,从一家红圈所跳过来的。
他讲那个手抄条款的时候,用了一个词。
他说:“这样举证责任就完全转移了。”
桌上有个人说了句“妙”。
马可当时也点了点头。
他坐在那间八百块的出租屋里,忽然发现自己连那个点头的动作都记得。
那个动作只有半秒钟。
十年以后,它落在一份判决书上,变成八个字:
原告已签字确认。
段二·古文镜像
“过而不改,是谓过矣。”
English Mirror: To err and not correct it — that is the error.
第八天,他找到了一个电话。
不是那七个原告的。
是另一个人的。
二〇一七年,他调研过一家做速冻水饺的老厂。
创始人姓周。
那个人坚持用当天的鲜肉,不用冻肉,这一条让他们的毛利率比同行低六个点。
马可在报告里写:供应链管理粗放,存在明显优化空间。
基金进去,管理层换了,冻肉用上了,毛利率漂亮了,股价涨了四成。
他退出得很漂亮。
两年后那个牌子塌了。
他找到了那个人。
不难找,那个人后来又开了一家小厂,在山东,做手工水饺,只在本地卖。
有一个抖音号,两千多个粉丝。
最新一条视频是三个月前,拍的是包饺子。
评论区里有一条留言,留了一个座机号,说是订货电话。
马可把那个号码抄在本子上。
然后他放了三天没打。
段三·事件主体
第八十九夜,晚上七点。
马可坐在那间出租屋的桌子前。
桌上有三样东西:
那本错题簿。
那张抄着号码的纸。
一杯已经凉了的水。
暖气片在响,那种老暖气片里水流的声音,咕噜咕噜的。
马可看着那个号码,看了四十分钟。
他一直在想一件事:我打过去要说什么?
说对不起?
那个人会说什么?
他大概会问:你是谁?
然后马可要解释:二〇一七年,有一家基金进了你的公司,我是做尽调的那个人,我写了一份报告,报告里建议改鲜肉为冻肉。
那个人会怎么样?
他可能会挂电话。他可能会骂人。他也可能,什么反应都没有,因为他压根不记得有一份报告,也不知道那份报告是谁写的。
而这最后一种,是最可能的。
他忽然想起马援那句话。
说自己的钱脏,是这世上最便宜的忏悔。
那这通电话呢?
他坐在那儿,把这件事往下想了一层。
要是那个人骂他一顿,他会难受,可他心里会有一样东西落地:我被骂了,我付出了代价。
要是那个人原谅他,他会更好受。
要是那个人什么反应都没有,那才是最糟的,因为那意味着他连一个能付的代价都没有。
他要的是一个代价。
而代价这个东西,是要那个人给他的。
马可坐在那间屋里,忽然觉得后背发凉。
他准备打这个电话,是去向一个七十一岁的老人,讨一样东西。
马可坐在那儿,忽然明白了一件事。
他这四十分钟里想的所有场面,都是关于他自己的。
他在演那个人的反应,而他真正在意的是,自己听完那个反应会好受一点还是难受一点。
他站起来,倒了一杯热水。
然后他坐回去,做了一件事。
他把那本错题簿翻开,写了一行:
第九错(正在犯):我准备打这个电话,是为了让自己好受。
怎么发现的:我在脑子里演了他四十分钟的反应,一次也没想过他现在过得怎么样。
经手人:马可。
写完,他盯着这一行看了很久。
然后他在下面又写了一行:
那怎么办?
他想了很久,写下三个字:
不说对不起。
八点十分,他拨了那个号码。
响了六声。
接了。
那边是一个中年女人的声音,背景很吵。
“喂,订货吗?”
马可怔了一下,他没想到是女的。
“您好,我想找周先生。”
“找我爸?”那边说,“他在包饺子。您哪位?”
马可深吸了一口气。
“我姓马。”他说,“我是二〇一七年,去过你们老厂的一个人。”
那边安静了一下。
“老厂?”
“对。”
那边好像捂住了话筒,跟旁边说了句什么。
然后是一阵脚步声。
“喂?”
一个很老的声音。
有点沙。听得出来是一个上了岁数的人。
“周先生。”
“你是哪位?”
马可握着电话,站起来了,他自己也不知道为什么要站起来。
“周先生,我叫马可。二〇一七年,有一家基金投了您的公司。我是做尽职调查的那一个。”
那边沉默了。
很长的沉默。
久到马可以为他挂了。
然后那个老人说了一句话。
“哦。”
就一个字。
马可站在那间屋里,一时不知道该说什么。
“周先生,”
“我记得你。”那个老人说。
马可怔住了。
“您记得?”
“记得。”那个老人说,“你那时候年轻,戴眼镜,问我肉是几点进的。”
马可站在那儿,忽然眼眶发热。
他确实问过。
他当年在那个车间里,问那个人:您这肉几点进厂?
那个人说:凌晨四点。屠宰场三点半出的。
马可当时在本子上记了一行:鲜肉采购,成本高于冻肉23%。
“周先生,”马可说,“我今天打这个电话,”
“你要说啥?”那个老人打断他。
马可停住了。
他准备了很多话。
可他忽然发现,那些话一句也说不出口。
他站在那间八百块的出租屋里,握着电话,很久没有说话。
然后他说了一句他自己都没想到的话。
“周先生,我想问您一件事。”
“问。”
马可深吸了一口气:
“您现在,还用鲜肉吗?”
电话那头安静了很久。
久到马可能听见那边包饺子的声音,案板,擀面杖,还有人在说话。
然后那个老人说:
“用。”
马可闭上了眼睛。
“每天四点,我自己去屠宰场。”
“您自己去?”
“我自己去。”那个老人说,“我今年七十一。”
马可握着电话,一句话也说不出来。
那边又说:
“我这个厂小。一天就出八百斤。”
“可这八百斤,是鲜的。”
“周先生,”马可艰难地开口,“那老厂,”
“没了。”那个老人说得很平常。
“我知道。”
“你知道?”
“我看到消息了。”马可说。
那边沉默了一会儿。
“那是我儿子接手以后的事。”老人说,“我那时候已经不管了。”
马可怔住了。
“您儿子接手?”
“嗯。”
“那,基金进来以后,”
老人在电话那头笑了一下。
那个笑很轻,一点也不苦。
“年轻人,你是想问我,我恨不恨你们。”
马可站在那儿,握着电话,手在抖。
“是。”
那个老人沉默了很久。
“不恨。”他说。
“为什么?”
“因为那笔钱,我拿了。”
马可怔住了。
“您拿了?”
“拿了。”老人说得很干脆,“基金进来,我套现了一部分。”
“多少?”
“一千二百万。”
马可站在那儿,脑子里“嗡”了一声。
“周先生,我不知道这个。”
“你当然不知道。”老人说,“你们做尽调的,做完就走了。”
他停了一下。
“那笔钱,我给我儿子在城里买了两套房。”
“然后呢?”
“然后我就退了。”老人说,“他要用冻肉,我说了两次,他不听。”
“您没拦?”
那边又是很长的沉默。
“年轻人,我拿了那一千二百万,我拦不动了。”
马可站在那间屋里,忽然觉得腿有点软。
他扶着桌子坐下了。
“周先生,”
“年轻人,”那个老人说,“我跟你说个实话。”
“您说。”
“你们那份报告,我看过。”
马可猛地抬起头。
“您看过?”
“看过。”老人说,“基金给我看的。他们说,这是专业机构的意见。”
“那您怎么想?”
老人在电话那头想了很久。
“我觉得写得挺有道理。”
马可闭上了眼睛。
“成本高百分之二十三,这个数是对的。”老人说,“我自己也算过。”
“那您为什么还坚持鲜肉?”
那个老人说了一句话。
那句话让马可后来在无数个夜里想起:
“因为我娘教我的。”
“怎么讲?”
“我娘一九六几年在食堂做过饭。”老人说,“她跟我说过一句话。”
“什么话?”
“她说:给人吃的东西,你得敢自己吃。”
马可握着电话,眼泪掉了下来。
“我那个厂,从开始到我退,我们家自己吃的饺子,跟卖的是一样的。”
“用冻肉那几年,”他停了一下,“我儿子在家吃的,是我另做的。”
电话两头都安静了很久。
那边包饺子的声音还在。
“周先生,”马可终于说,“我今天打电话,本来是想跟您说对不起的。”
“别说。”老人打断他。
“为什么?”
那个老人的声音很平静。
“因为你那份报告,写的是真的。”
“那百分之二十三,是真的。”
“你没说谎。”
他停了一下。
“你只是没写另一件事。”
马可屏住呼吸。
“哪一件?”
那个老人说:
“你没写:那百分之二十三,买的是什么。”
马可坐在那间出租屋里,握着电话,一动不动。
“周先生。”
“嗯。”
“我能不能问您最后一个问题?”
“问。”
马可深吸了一口气:
“要是当年那份报告里,我写了那一条,您觉得,那家基金会改主意吗?”
那个老人沉默了很久。
然后他说了一句极其实在的话:
“不会。”
马可闭上了眼睛。
“可我会看见。”老人说。
马可猛地睁开眼。
“您会看见什么?”
“我会看见,写那份报告的人,知道我在干什么。”
他停了一下。
“年轻人,一个人做一件事做了三十年,别人不一定要帮他。”
“可他很想知道,有没有一个人,看懂了。”
马可握着电话,很久没有说话。
那边包饺子的声音还在。
“周先生。”
“嗯。”
“我现在在陕西。”
“干啥?”
马可想了想,很实在地说:
“我在一个村里,弄了一个存苹果的库。”
“多大?”
“五百吨。今年存了九百八十四斤。”
那个老人在电话那头笑了。
“九百八十四斤?”
“九百八十四斤。”
那边安静了一会儿。
然后那个老人说了一句话。
那句话马可后来把它抄在了那本错题簿的扉页上。
“年轻人,”那个老人说,“我一天出八百斤。”
“你别嫌少。”
“少的东西,你才看得见每一斤是谁的。”
段四·钩子结尾
那通电话打了四十七分钟。
挂断的时候,是晚上九点。
马可坐在那间八百块的出租屋里,一动不动地坐了很久。
暖气片还在响。
桌上那杯水,又凉了。
他打开电脑,翻开那本错题簿。
他在第九条底下,写了很长一段:
补记(当夜):
我打了那个电话。我没有说对不起,他不让我说。
我原以为这一关是“承认我做过什么”。我错了。
真正的那一关是:那份报告里的每一个数,都是真的。百分之二十三是真的,账期是真的,同业对比是真的。
我没有说谎。
我只是把一样东西,从那份报告里漏掉了。
那一样东西是:那百分之二十三,买的是什么。
我当年不是不知道。我在他的车间里站过,我问过他肉几点进厂,他说凌晨四点。
我把那句话记成了一行成本。
我没有把它记成:一个人每天凌晨三点起床。
他打到这儿,停了很久。
然后他写了最后一行:
这不是一次道德失误。这是一次记账失误。
而我这十几年学的所有本事,都是记账。
他写完这一行,又停了很久。
然后他往上翻,翻到那本错题簿的最前面。
第一条:三十七页材料,答了十一个问题。
第二条:一直在找能拍板的人。
第三条:在陕西一个自己的人都没有。
第四条:以为该说服强东林。
第五条:模型里没有换工那一格。
第六条:刻了铁牌,没跟强磊谈条件。
第七条:把人的名字刻在铁上之前,没给他一个数。
第八条:以为“我什么也不要”能澄清误会。
第九条:打电话是为了自己好受。
他一条一条看下去。
九条。
没有一条是他自己先发现的。
第一条是上官婉儿。第二条是他自己在城墙上,可那是杜甫把他逼上去的。第五条是强东林。第八条是卜式。
他忽然明白了一件事,在那本簿子的最后一页写了下来:
这九条里,有八条是别人替我发现的。
而那八个人里,只有两个是活人。
这说明一件事:我这半辈子,身边一直缺一个肯当面告诉我我错了的人。
不是没人看见。是没人肯说。
而没人肯说的原因,多半跟我有关。
他关上电脑,走到窗前。
洛川县城的夜。
十二月了。天很冷。楼下有一家还开着的小超市,门口挂着一串塑料的彩灯。
马可站在那儿,忽然掏出手机。
他给强磊发了一条微信。
强磊,还没睡吧。
没。刚从库里回来。
今天多少度?
零下一度。稳的。
马可打字:
我问你一件事。
你说。
那本台账上,你记的是什么?
那边有点莫名其妙。
记的是斤数、价钱、日子啊。
马可站在窗前,慢慢地打:
从明天起,加一行。
加啥?
马可看着楼下那串塑料彩灯,一个字一个字地打:
每一批果,记一句话:
这一批是谁家的,那家人今年遇到了什么事。
那边过了很久。
马哥,这个跟存果有啥关系?
马可打字:
没关系。
那记它干啥?
马可站在冬天的窗前,打了很长一段:
强磊,我今天给一个人打了电话。
十年前我写过一份关于他的报告。
那份报告里所有的数都是对的。
可那份报告里没有他。
他在里头是一行叫"原材料采购成本"的东西。
那边隔了很久。
懂了。
然后又来了一条:
马哥,那我今天先记一条。
记啥?
记:本年首存,强东林,九百八十四斤。
这一批是他卖完全年之后特意留下的,挑的是最好的。
他前年存过一次,赔了一万九,三年没敢再存。
存这一批的前一天夜里,他在院子里蹲到十二点,抽了大半包烟,把十六筐苹果重新摆了三遍。
马可站在那扇窗前,看着这条消息,眼泪掉了下来。
他打了三个字:
就这么记。
他刚要收起手机,那边又来了一条。
马哥。
你说。
还有一件事。
马可等着。
今天下午王主任又来了。
马可的心一下子提起来。
他来干啥?
那边打了很久。
他没进库。
他在门口那块铁牌前面,站了大概二十分钟。
看那六条?
不是。
马可皱起眉。
那他看什么?
那条消息终于来了。
他在看底下那块木板。
马可怔住了。
那块粗木板上,是他用朱笔写的三行歪字。
本年冬,首存:强东林,九百八十四斤。
那边又来了一条:
马哥,他看完以后,问了我一个问题。
他问什么?
马可站在那间出租屋里,握着手机。
他问:那块木板上,明年冬天写谁?
马可站在那间出租屋里,看着这条消息,很久没有动。
这是一个很好的问题。
好到他自己都没想过。
他打字:
你怎么答的?
那边很快回了:
我说:谁先存写谁。
他说什么?
那边隔了一会儿。
他笑了一下,走了。
欲知后事如何,请看下回,《过所三刻一个本》
附录一·Dongpo's Note
(今夜无人现身。此页录马可当日所记。)
Nobody came tonight, which is right. Nobody could have done this for me.
Eleven days of searching, on Ma Yuan's instruction, and here is what I actually found.
The 2014 bond did not default. It paid in full in 2019 — out of the proceeds of a new issue. The company is three times larger now. Its controlling shareholder was investigated in 2021 for improper guarantees and misappropriation, and the factoring company I recommended appears in the filing. So the structure I helped assemble did not collapse. It kept rolling, and then it did what it was always going to do, later, to other people.
I could not obtain the 2015 client list. Two former colleagues did not reply; the third sent two characters: don't ask. What I found instead was seven public judgments. Seven individuals who sued the private bank. All dismissed. The fifth plaintiff was sixty-seven and stated that the money was the entire retirement savings of himself and his wife. Dismissed on the grounds that he had signed the risk disclosure.
I know that document. Our team wrote the template. Two pages at the back of a forty-seven-page contract, and the client copies out one sentence in his own handwriting — I have read and understood the above risks and voluntarily bear investment losses. Handwritten, not printed, because a handwritten sentence is decisive in court. A lawyer added it and we all thought it was clever.
Then the call I had been avoiding for three days.
Zhou. Seventy-one. Founder of the dumpling plant I diligenced in 2017, the one who insisted on same-day fresh pork at 23% above frozen, which I wrote up as "coarse supply-chain management with clear optimisation potential." Fund entered, management changed, frozen pork came in, gross margin improved, the stock rose forty percent, we exited cleanly, and the brand collapsed two years later.
Before dialling I spent forty minutes rehearsing his reactions, and then realised every one of those rehearsals was about me — about whether I would feel better or worse afterwards. Logged it as error nine and wrote one instruction underneath: do not apologise.
He remembered me. "You were young, you wore glasses, you asked what time the meat came in." I did ask. He said three-thirty at the abattoir, four at the plant. I wrote it down as a cost line.
What I did not know, and what nobody who does diligence ever knows because we leave when it closes: he took twelve million out at the fund's entry and bought his son two flats, and then retired, and his son switched to frozen pork over his objection twice. His words: I took the twelve million, I no longer had the standing to stop him.
He is not angry with me and gave the reason plainly. He read our report — the fund showed it to him as the professional opinion — and he thought it was well argued. The 23% was correct; he had calculated it himself.
He kept fresh pork because his mother, who cooked in a canteen in the sixties, told him: what you give people to eat, you must be willing to eat yourself. Through his whole tenure, his family ate the same dumplings he sold. In the frozen years, his son made a separate batch for the house.
He would not let me apologise, and his reason is the sharpest thing said to me in six months. The report was true. Every figure in it was true. I did not lie.
I simply did not write down what the twenty-three percent was buying.
And when I asked whether including that line would have changed the fund's decision, he said no — immediately, without bitterness — and then: but I would have seen it. I would have seen that the man who wrote the report understood what I was doing.
A man who has done one thing for thirty years does not necessarily need help. He wants to know whether anyone has understood.
Filing this where it belongs. It is not a moral failure. It is an accounting failure — and accounting is the only thing I have ever been trained to do. — M.
附录二·Marco's Memo
Memo to file. Rented flat, Luochuan county town. Night eighty-nine. No visitation.
Subject: provenance work, completed as far as it can be.
1. 2014 bond, USD 300m. Status: repaid at maturity 2019, refinanced. No default event attributable to the receivable structure. Issuer now ~3x prior scale. Controlling shareholder investigated 2021 (improper guarantees, fund misappropriation); the factoring counterparty I recommended is named in the filing.
Correct reading: the structure did not fail. It transmitted. My work made a fragile balance sheet financeable for five more years, at the end of which the exposure had grown and was held by different people.
2. 2015 structured product. Client list unobtainable — I left ten years ago and the distribution entity was separate. Two of three former colleagues did not respond; one replied "don't ask."
Substitute source: seven published judgments, plaintiffs vs. the distributing private bank. All seven dismissed. Common ground of dismissal: signed risk disclosure.
Plaintiff five, aged 67, pleaded that the sum was the entire retirement savings of himself and his wife.
Note on the instrument that defeated him: our own template. Forty-seven-page contract, two-page disclosure at the back, one sentence copied in the client's own handwriting because a handwritten acknowledgement is evidentially decisive. Added by counsel; considered elegant at the time; I did not object.
3. The dumpling plant, 2017. Located the founder via a small Douyin account, ~2,000 followers, an order line in the comments. Sat on the number for three days.
4. Pre-call error, logged as no. 9: forty minutes rehearsing his responses, none of it about his present circumstances. All of it about how I would feel afterward. Corrective instruction written before dialling: do not apologise.
5. Call, 47 minutes. Findings I did not have and could not have had: — He remembered the specific question I asked in the workshop (what time the meat arrives). I recorded his answer as a cost differential. — He monetised ~¥12m at the fund's entry, bought his son two apartments, and stepped back. The switch to frozen was his son's, over his objection, twice. His assessment of his own position: "I took the twelve million, I could not stop him." — The fund showed him our report as independent professional opinion. He found it persuasive. He had independently computed the same 23%.
So the report was not a hidden weapon. It was handed to him and it convinced him, which is worse.
6. His refusal of my apology, and the reason. Every number in the report was accurate. There was no falsification. The omission was categorical, not factual: I did not record what the 23% purchased.
I had the input. I stood in that workshop and asked. He told me: 3:30 at the abattoir. I converted a man's 3 a.m. into a line item and the conversion was itself the error.
7. Counterfactual, put to him directly: would including it have changed the fund's decision? Answer, immediate: no. Followed by: but I would have seen it. I would have seen that the person who wrote it understood what I was doing.
"A man who has done one thing for thirty years does not necessarily need help. He wants to know whether anyone understood."
I do not have a framework for that and I am not going to invent one tonight.
8. Classification, and this is the entry that matters. This was not an ethical lapse — I did not lie, conceal, or misstate. It was an accounting failure: a category of cost that my ledger had no column for, and therefore did not exist.
Everything I have been trained in for fifteen years is accounting. That is the whole of it.
9. Action, 21:30. Instructed Qiang Lei to add one line per lot to the storage ledger: whose fruit it is and what happened to that household this year. He asked what that has to do with storing apples. Answer: nothing.
His first entry, written unprompted and better than anything I would have specified — Qiang Donglin's 984 jin, held back after the year's sale, best of the crop; lost ¥19,000 two winters ago and did not dare store for three years; spent the night before squatting in the yard until midnight, most of a packet of cigarettes, re-sorting sixteen crates three times.
That paragraph is the column my 2017 report did not have.
10. 21:50. Director Wang returned to the store this afternoon. Did not enter. Stood twenty minutes in front of the iron plate — not reading the six articles, but reading the rough board beneath it, the three crooked lines I wrote in vermilion.
Then asked Qiang Lei one question: whose name goes on that board next winter.
I am recording it without interpretation.
附录三·English Highlight
"To err and not correct it — that is the error."
The Analects line is usually read as an argument for repentance. What it does not tell you is what happens when the error was never a lie.
A diligence report on a dumpling manufacturer, written in 2017, recorded that the founder's insistence on same-day fresh pork cost twenty-three percent more than frozen and constituted "coarse supply-chain management with clear optimisation potential." Every figure in it was accurate. The founder himself had run the same calculation and reached the same number, and when the fund handed him the report he found it persuasive.
He took his liquidity, bought his son two apartments, retired, and watched his son switch to frozen pork over his objection. The brand collapsed within two years. He does not blame the analyst, and says so without bitterness: the report was true; you did not lie.
What it omitted was what the twenty-three percent was buying.
The information had been available. The analyst had stood in the workshop and asked what time the meat arrived, and been told: three-thirty at the abattoir. He wrote it down as a cost differential rather than as a man rising at three in the morning, and the conversion — not any falsehood — was the whole of the failure.
Asked whether including that line would have changed the fund's decision, the old man answered immediately: no. And then: but I would have seen it. I would have seen that the man who wrote it understood what I was doing.
He is seventy-one, runs a small operation producing four hundred kilos a day, and still goes to the abattoir himself at four in the morning, because his mother — who cooked in a canteen in the 1960s — told him that what you give people to eat, you must be willing to eat yourself.
附录四·Dongpo's Cheat Sheet
| 今夜之事 | 出处 | 现代对应 | 一句话 |
|---|---|---|---|
| 债没违约,又发了一笔 | 2014年境外债 | 结构没垮,是在滚 | 我让一张脆弱的表,又能融资五年 |
| 七份判决书 | 裁判文书网 | 找不到名单就找诉讼 | 第五份原告六十七岁,全部退休积蓄 |
| 手抄那一句 | 风险揭示书 | 为打官司而设计 | 加的时候,所有人都觉得很聪明 |
| 演了他四十分钟 | 打电话之前 | 忏悔的自利性 | 我在演的是自己听完会不会好受 |
| 你问我肉几点进的 | 老周记得 | 那句话我记成了成本 | 凌晨三点半,被我写成百分之二十三 |
| 我拿了一千二百万 | 老人自陈 | 他也不是白的 | 拿了那笔钱,我拦不动了 |
| 给人吃的东西你得敢自己吃 | 他母亲的话 | 那百分之二十三买的东西 | 用冻肉那几年,他儿子在家吃的是另做的 |
| 你没说谎 | 老人拒绝道歉 | 不是道德失误 | 是记账失误,漏了一整个类目 |
| 会不会改主意?不会 | 反事实 | 可我会看见 | 干了三十年的人,想知道有没有人看懂 |
| 台账加一行 | 马可的新规 | 把人写回账里 | 跟存果没关系,所以才要记 |