[infoseek12]

It will be interesting if a lot of research gets published in Chinese journals and not the West. If this happens I wonder if Western researchers will get their information from Chinese journals or lag a bit behind Chinese researchers on the answers to these questions.

[fnordpiglet]

If you’re willing to cut ethical corners in one dimension there’s no reason to believe you won’t in other dimensions. Doing rigorous research with honesty, transparency, and ethically is interrelated. That’s why being a reputable journal is so important - presumably the standards are high and the research is more reliable, and citing it is more credible. While not consistently true or universally true, it’s at least true the less reputable journals are known to be … less reputable. So, yes, you could republish in less reputable journals, but other researchers will assume a low quality result intermixed with fraud, unethical practice, and lack of transparency. Fair or not Chinese journals don’t have a reputation for quality like mainstream journals, even among Chinese researchers. Finally a Chinese language journal has almost no readership outside of China, further diminishing the impact of your research. Even citing the research in an international journal would be problematic.

[roenxi]

> If you’re willing to cut ethical corners in one dimension there’s no reason to believe you won’t in other dimensions.

There is reason to believe. People have different ethical standards and some standards are more conducive to getting honest useful results. In this case I think free and informed consent is important. But the ethical framework our medical system works to is counterproductive and destructive; so there is reason to believe that the ethical option is to ignore the official rules on ethics under some circumstances. I personally am of the opinion that the single biggest threat to my health and wellbeing is the West's culture of forcing everyone to conform to the highest possible safety standard and paying no attention to the cost-benefit of that. As I measure it, our absurd standards appear to be crippling (with respect to some hypothetical entity that had more reasonable standards) the west's:

* Manufacturing capabilties.

* Energy security.

* Medical research.

So while someone with different ethical standards might also cut corners on the quality of the reporting I find it extremely easy to believe that someone with different standards might just produce better research. We don't all have the same ethical frame.

[fnordpiglet]

The issue is less related to morality, which I think is what you are referring to, and more following the ethical guidelines and standards for a journal. The former is relative, and the latter is definite and concrete. It’s related to “following the rules,” less about “conceiving of right and wrong as I do.”

That’s where the interrelation with following standards regarding honesty, transparency, etc. You can fall afoul of these out of inability to follow the rules rather than duplicity - there are standards for record keeping, statistical rigor, etc, which are prescriptive and fairly inconvenient. If you can’t follow the ethics rules because you find them inconvenient per your cost benefit analysis, what about rigorous record keeping? Not ignoring inconvenient outliers? What calculus do you use that lets you choose which inconvenient rule you won’t follow? As someone who doesn’t know your standards, how do I trust you’re making well informed decisions that don’t skew results in your favor in your convenience calculus?

A simpler way is hold a standard and rules and when you vary you assume variance exists everywhere.

This is science, not business. Science isn’t just about discovering results, it’s about conveying your results in a rigorous and believable way following high standards that build confidence others can build ontop of your work. Cutting corners damages your reputation on all dimensions.

I would also note these standards come about for a real reason it’s not just a bunch of anti progress bureaucrats devising ways to make everything expensive from a “western” culture perspective. These are best practices derived from observations of real and genuine horrors. They are absolutely restrictive. But when the rules don’t exist, some jackass somewhere does something inhuman and gets rewarded for it. Maybe it creates progress, but it leaves real victims whose lives are destroyed. This being unacceptable isn’t a western culture thing. It’s a human thing. No one wants to be the victim, and most people agree it’s not ok to exploit and harm people for your benefit. The difference is some societies (not cultures) have weak systems that protect the individual over profit, power, and greed. One could make the argument that the current society in China doesn’t prioritize human rights very highly, but I think it’s wrong to say Chinese culture doesn’t. There’s a difference between a people and their government.

[roenxi]

Interesting post, but no I mean the ethical guidelines and standards of the journal. I suspect that the scientists treat them like people treat EULAs - they make assumptions about what is in there and often don't actually read them unless someone rejects their paper because of the guidelines.

[fnordpiglet]

My grandfather was a career medical researcher and medical school founder, with a career spanning heart surgery, kidney transplant pioneering, to blood diseases, and pulmonary systems. He had hundreds of papers published, hundreds of patents, and published until he died in his 80s. In my experience he was very well versed in the ethical standards for medical research, sat on the human research standards boards, and oversaw likely thousands of graduate students. My belief from existing in his orbit and sitting in his labs most my summers growing up was people were very concerned about the ethics of human research. I think generally the culture of biomedical research tends to reject people who aren’t from the field, and journals tend to hold fairly uniform standards that are pretty broadly agreed upon. I don’t think the researchers in question here were ambiguous as to whether forced or coerced consent to genetic forensic research is ethical or not. It was just convenient and didn’t require sourcing subjects and engaging with them sufficient to gain positive informed consent. People that traffic in slave labor goods are almost certainly aware it’s considered wrong, people who traffic in forced medical research also know it’s considered wrong. It’s not the same as skipping through a ten thousand page boilerplate of garbage.

[hnfong]

It's very weird that you seem to have such faith in the so called reputable journals, and speak in such a seemingly elitist tone, at a time where the "replication crisis" and in general academic fraud is so well known.

[npunt]

Agree, unethical research also makes replication harder thus any results are more suspect.

[quotemstr]

"Ethics" isn't one thing. One can be committed personally to honesty and scientific rigor yet opposed to curiosity-killing IRB culture that purports to think that a questionnaire somehow might rise to the level of atrocity but not for IRB review.

[fnordpiglet]

An ethical code and standard however is. Cutting corners in following the rules in one area for convenience is indicative of not following rules that inconvenience you. It’s not relevant if you agree with the morality of the ethics, which are rules and are singular and expressed as a standard. If you can’t comply, you can’t be assumed you follow the other rules as well.

[quotemstr]

https://www.simplypsychology.org/kohlberg.html#Level-3-%E2%8...

[fnordpiglet]

Except this is science, not daily life. There is a process and procedure expected to ensure integrity of results and consistency in the process. Yes the ethical guidelines are there to attempt to achieve some morality, but from a scientific process perspective it doesn’t matter if you agree or disagree or have a different morality. It’s a process integrity question above all else. You aren’t prevented from doing science however you feel is moral, but if you don’t conform to the process requirements front to back, don’t expect journals to publish your work. It’s like I tell engineers - you can prefer tabs or spaces or place braces wherever you want on your own time, but when working in a shared code base, show your creativity in your ideas not your coding style. Likewise show your moral beliefs in your life, but when doing biomedical research, follow the ethics rules because they’re part of the shared process.

[ametrau]

Great. So you say. Now how do we tell them apart?

[BlueTemplar]

How readily does reputation change with scandals ? Because we seem to be in the middle of a pretty big one, where the functioning of the whole system is in question.

Starting with a milder example :

https://www.ageofinvention.xyz/p/age-of-invention-how-to-be-...

Followed by a more... radioactive one :

https://alexwasburne.substack.com/p/the-sausage-factory-of-p...

BTW for this last one, HN seems to be outright blocking submissions, with the dreaded (and in this case disingenuous) "you are posting too fast" error. Now I'm willing to leave some benefit of the doubt that this is done in order to minimize flamewars, but it does still leave a bad taste...

But to conclude, some wise (??) words :

https://slatestarcodex.com/2017/10/23/kolmogorov-complicity-...

(The links to Paul Graham's and Scott Aaronson's takes are also worth reading.)

[dekhn]

In the late 1800s and early 1900s there was a lot of world-class science going on in England and Germany and most of it didn't get translated, which led to both countries being somewhat behind the advances of the other. The Dutch and Danish played a big role in mediating between those two countries, which were in intense competition at that point.

I've long assumed that if China continues to invest in its research culture, eventually (probably even now) there is research being published in Chinese (I assume that means Mandarin, but I am not knowledgeable on specifics) that would be extremely valuable. Areas like material sciences, optics, EE, MEMS, etc.

But I think since the US has so much cultural exchange with China, and so many Chinese-speaking residents, that probably any really important stuff will get translated.

[zelse]

Re: Publishing in Mandarin, just in case you're curious:

Mandarin, Cantonese, Hakka, Hokkien, Wu, etc are spoken languages (or dialects, or varieties), but the writing system is common to all of them, albeit in two major forms -- Simplified and Traditional. If I write 恭喜发财, a Mandarin speaker will read that aloud as gongxi facai, a Cantonese speaker as gung hei faat coi, and a Hokkien speaker as kiong-hi hoat-chai, though I left off tone markings even though they're important. In that sense, unless something's romanized, written materials are pretty much just written in 'Chinese', and not any particular variety.

Learning Chinese is funny that way: the spoken language has very simple, regular grammar, but then you also need to learn how to read, write, and pronounce ~3,000 characters to read a newspaper, with a well-educated reader in the language typically recognizing between 8 and 15 thousand characters of the 80,000-100,000 or so that have ever existed (including obscure characters, obsolete characters, regional characters, and variant characters).

It's similar to how if you show a Frenchman, an Englishman, a German and a Pole '456', they'll all know the meaning even if the Englishman thinks "four-hundred fifty-six", and the Frenchman thinks quatre cent cinquante-six.

[hnfong]

This is mostly true, but not exactly.

Most Chinese speakers can understand (written) Mandarin, but all the spoken languages can be written in theory. Mandarin was just the first one to have been popularly written and it was the official dialect, so it became the de-facto standard (before it became de jure).

People in Hong Kong for example have been writing Cantonese for decades, albeit more often recently than before. But if you looked you'd find published works written in Cantonese 100+ years ago. That said, if you only knew Mandarin it will probably only take you a couple days to get the hang of reading Cantonese since it's not that different when written.

[zelse]

Oh neat. Do these writings use Hanzi or something else?

[hnfong]

廣東話基本上係用漢字但係台灣嘅情況有啲複雜。可以睇下 https://en.wikipedia.org/wiki/Written_Hokkien

(Written Cantonese basically uses Hanzi, but the situation of Taiwanese is more complicated [or maybe I don't understand it enough]. You can check out the link above.)

[soundnote]

The writing is still fundamentally Mandarin, is it not? Hanzi just disconnect the writing somewhat from the sound of the individual language, and the languages are closely related enough that you can read written Mandarin reasonably easily as a speaker of other Chinese languages.

At least eg. Hannas - far from an impartial voice on the subject, mind - clearly distinguishes the written standard as a separate language from dialect speakers' native ones:

"By the same token, the 'unity' that Chinese characters allegedly impart to the language by allowing speakers of different 'dialects' to read a common written language turns out to be an illusion. These so-called Chinese dialects have less in common than the Romance languages of Europe, meaning that speakers of nonstandard Chinese (some 30 percent of the Han population) are not reading their own language or even a common language, but what is to them a Mandarin-based second language written in Chinese characters. Granted the characters allow non-Mandarin speak- ers to read segments of written Mandarin in their own regional pronuncia- tions. But, far from unifying Chinese, this practice only perpetuates differences that would have been leveled out long ago under the influence of a phonetic script."

— Hannas, Asia's Orthographic Dilemma, Ch.8, Appropriateness to East Asian Languages

[worik]

Mēh!

Let the Chinese authorities burden themselves with questions of DNA markers and ethnic groups

Eugenics has not done anyone any good, and when used as a means of controlling populations it goes to very dark places

If the whole stack is not kept ethical, it is not ethical at all. If not ethical then it tends to evil

[aydyn]

> If the whole stack is not kept ethical, it is not ethical at all.

This is such a naive, borderline illiterate take. How much of modern human biological knowledge is built upon a foundation of unethical practices?

[kibwen]

From a practical (rather than moral) perspective, the reason that scientific ethics are a thing in the first place is so that the public does not sour on scientific endeavors and view them as illegitimate. If you eschew ethics, then, like Dr. Frankenstein, eventually you will have the public beating down your door with torches and pitchforks, regardless of the results you achieve.

Which is to say, if you care about advancing science in the long term rather than merely in the short term, you must care about ethics. Accusing others of naivete here is a bad look.

[aydyn]

From a practical perspective, nobody cares unless it affects them.

Ethics? AI research today is plainly not ethical, yet its full steam ahead.

Whats a bad look is some idealized notion that everything can be perfect and ethical and there are no gray areas. That is plainly magical thinking.

[hnfong]

From a practical (rather than moral) perspective, as long as the research is approved by the government and the government is authoritarian, there won't be any public beating down your door with torches and pitchforks.

[xdennis]

> Eugenics has not done anyone any good

It's reduced genetic defects in the population. For example in many countries you can abort babies with Down syndrome up to the time of birth.

[BiteCode_dev]

Are writing this comment from a phone made with no unethical issues?

Cause I don't know any.

[cocacola1]

No evil is preferable, but a little evil is acceptable. A lot of evil, however, is not (unless compared to pure evil, I suppose).

[BiteCode_dev]

So the whole chain must be ethical, but children labor to make your phone is a little evil that is not counting.

Absolute statement, then relativism.

I'm convinced.

[cocacola1]

Rather, a recognition of the current state of affairs. And not just my phone; my clothes, some chocolate I had recently, and probably most things I own. The overseer is more evil, but we’re all a little evil (or a lot?), though I doubt we’re all about to go off grid and be self sufficient, weaving our own clothes, building our own shelter, foraging our food, etc.

[BiteCode_dev]

We can apply this argument to the original scientific frauds.

At this point I have no idea what your opinion is.

[yogorenapan]

How are you measuring “evil”?

[cocacola1]

By the tried and true methods of “I know it when I see it” and the smell test.

[COGlory]

No. No one trusts Chinese research to be:

A) accurate B) reproducible

Even in Western journals it's frequently suspect.