楽園こちら側」の「事実に誠意を」をほぼdeepLで翻訳してみた その2


12 Dr. Hiroshi Nishiura is one of the few professionals of mathematical models of infectious diseases in Japan, and it is well known that his ability is outstanding. However, many people don't understand mathematical models themselves (I must confess that I can't say that I understand all of the findings because I'm not a professional of mathematical models either), so his findings and comments are easily deified. Because the contents of the mathematical model are a complete black box to many people, it makes it seem like the oracle is coming out like a shrine's oracle. Much of Japan's infection control policy relies on the Nishiura theory. So there is nothing wrong with that, but one of the problems in Japan is that there is no plan B in case plan A goes bust. Dr. Nishiura is an excellent scholar. It is not God. Hence the need to have that Plan B with the possibility of making a mistake. I am greatly concerned that bureaucrats and politicians who are prone to infallibilism will mistake science for an oracle. It is only when falsifiability is assured that science can continue to be scientific.


13 数理モデル演繹法活用産物である演繹法帰納法アブダクションで補完するのが、学問の基本であり、臨床医学常識である演繹法的にどんなに正しく見えても実はそれは違っていた、ということはこの業界ではよくあることなのだ。ヘーゲルマルクスのような巨大な知性でも演繹法オンリーでは間違うのである

Mathematical models are the product of deductive methods. The deductive method is complemented by the inductive or abduction method, which is the basis of scholarship and the common sense of clinical medicine. It's a common occurrence in this industry that no matter how deducibly correct it may seem, it's actually not true. Even a huge intellect like Hegel or Marx can make a mistake by deduction alone.


14 モデルを使うな、といっているのでは決してない。ぼく自身モデルを用いて論文を書く。しかし、モデル無謬ではなく、そこには前提である仮定があり、仮定はしばしば間違っている。グラム染色活用するとは、グラム染色にできないこと、分からないことを知悉していることであり、グラム染色万能論者にグラム染色は使えない。同じことだ。英国でも数理モデル活用されているが、だからこそ英国人はその結語には非常に懐疑的で、常に反論異論が起きている。健全科学的な態度である

 I'm not saying don't use the model at all. I myself write a paper using a model. However, the model is not infallible, there are assumptions that are assumptions, and the assumptions are often wrong. Making use of Gram's stain means having full knowledge of what Gram's stain cannot do and does not understand, and Gram's stain cannot be used by Gram's stain universalists. It's the same thing. Mathematical models are also utilized in the UK, which is why Brits are very sceptical of their conclusions, and there are always counter-arguments and objections. It is a sound and scientific attitude.

感想:「前提たる仮定」がうまく訳せていなかったので「前提である仮定」にしたが、assumptions that are assumptionsになってしまった。


15 Japan's "now" is a well-controlled state of infection, which is much better than Wuhan at its worst, or Italy, Spain, France, England, or New York at the present time. The problem is that it doesn't guarantee that it will "always work".


16 懸念されるのは東京だ。感染報告が増えたことだけが問題なのではない。クラスター形成できない、トレースできない感染者が増えているのが問題である。そして、その陽性患者数に比べて検査数がずっと少ない。47人の感染者を捕捉するために100人未満(陽性者の検査日が不明だが、おそらくこのへんだろう)しか検査していないのは少なすぎる。

It is Tokyo that is of concern. The increase in reports of infection is not the only problem. The problem is that more and more infected people are unable to form clusters and cannot be traced. And the number of tests is much lower than that number of positive cases; it's too little that they only tested less than 100 people (the date of testing for the positives is unknown, but it's probably around here) to capture 47 infected people.

Again, it's not necessary to figure out all the infected people. However, it is troubling that the flow of infection, movement and clusters are out of sight. Therefore, the threshold for testing must be lowered in Tokyo. The threshold for testing varies with the circumstances. That's what I explained with the Korean example. Sticking to the Ministry of Health, Labour and Welfare's "standards" will lead to a misunderstanding of the phenomenon itself. Already in the Kansai region, infected people have been found with taste and smell abnormalities, and clusters have been detected from there. I would like to make more use of the athletic sensibilities of these clinicians. I'm not sure "where" in Tokyo is the barrier to lowering the number of inspections, but that barrier needs to be removed immediately.



17 This conceptual diagram that everyone is looking at - lowering the peak of the infection and shifting it to the side. This is all a product of deduction, and I don't know if it's really true. As mentioned above, the UK estimates already suggest that this is not enough. It is possible that the damage that was shifted to the side could simply be "extra-long damage".


18  そして、ここが肝心なのだが、ピークを下げるという理念が、「ピークを下げなければいけない」という観念になり、「ピークは下がっているはずだ」という確信になり、「ピークは起きていないんだ」という自己暗示に転じてはいけないということだ。プランAに固執する日本あるあるの失敗のパターンで、ダイヤモンドプリンセスでは「二次感染が起きてはいけない」が「起きているはずがない」に転じてノーガード下船を許してしまった。「ピークが起きてはいけない」が「ピークなんて見たくない」にならないように現実を見据える必要がある。たとえ、それが我々の見たくない不都合な真実であったとしても。

And this is the key point: the idea of lowering the peak should not become the notion that the peak must be lowered, or the belief that the peak must be lowered, or the self-implication that the peak is not happening. In a pattern of Japanese failure to stick to Plan A, Diamond Princess allowed no-guard disembarkation by changing "secondary infection should not occur" to "it can't have happened". We need to keep our eyes on reality so that "peak shouldn't happen" doesn't become "I don't want to see a peak. Even if it is an inconvenient truth that we don't want to see.



19 Repeatedly. It's common knowledge in this industry that deductive methods are complemented by inductive methods. Nevertheless, PCR is often false-negative and has little power to determine the status of infection. That's why "testing everything" is so wrong. However, a serum test measuring immunoglobulin IgM and IgG would provide a more accurate picture of the "status of infection in the population. This, however, is not infallible. It is difficult to use for individual cases because it misses early infection, which is why it misses early HIV infection.Whether antibody testing is useful in individual cases remains to be tested, but it is well suited for epidemiological studies on a population basis. Roughly speaking, we can confirm whether the "infection is rampant" in Tokyo right now, or whether it's just an unfounded fear.


As a precedent, serology tests in London showed that the 2009 pandemic flu was 10 times more likely than previously predicted. Antibody testing is often performed after an outbreak, but now is a good time to examine COVID-19, which is becoming a chronic pandemic.


20 英国さらアグレッシブだ。家庭で抗体検査を行い、「感染である」とわかればそれを自宅での自己隔離根拠に使おうというのだ。ロックダウンが起きている中で、検査陰性は「自己隔離不要」を意味しないため、その戦略に欠陥はある。が、考え方としては「感染全体を抑え込みたい」というもので、検討価値はあると思う。

The UK is even more aggressive. The idea is to test for antibodies at home, and if they are found to be infected, they will use it as a basis for self-isolation at home. That strategy is flawed because with the lockdown in place, a negative test does not mean "no self-sequestration". However, the idea is that we want to control the infection as a whole, and I think it is worth considering.


21 東京でどのくらいの感染が起きているか帰納法確認必要であり、有用だ。その結果がどうなるかは預言者ではないぼくには分からない。が、どんな結果が出てきても、それを受け入れ、場合によっては自説を変えて、プランBに移行することにも躊躇しない態度が科学者には必要だ。科学者は、首尾一貫していないことにかけて、首尾一貫していなければならないのだ。形式においては首尾一貫していなくても、プリンシプルプロフェッショナリズムにおいて一貫しなければならないのだ。事実に誠意を。 

Inductive legal confirmation of how many infections are occurring in Tokyo is necessary and useful. I'm not a prophet, so I don't know what the outcome will be.However, no matter what the outcome, scientists need to accept it and not hesitate to change their thesis and move on to Plan B in some cases. Scientists have to be coherent in their inconsistencies.They may not be coherent in form, but they must be coherent in principles and professionalism. Good faith in the facts. 




記事への反応 -


ログイン ユーザー登録
ようこそ ゲスト さん