Hacker News Reader

Comment by oytis | Hacker News Reader

A study (if it's a real study) can't tell you what should be done, it should give you new knowledge about the state of affairs.

(while I agree that drug offense should be decriminalized).

View on HN

Was about to comment this myself - totally agree.

The single quickest way to make me suspicious of an empirical study is to give it a prescriptive title. I agree with this title wholeheartedly, but I want (1) studies that provide objective evidence and (2) position papers that recommend policies based on that evidence. Both are legitimate copes for researchers to publish within if they call it what it is. This is a position paper, not a study.

Exactly. Any form of science cannot tell us what should or should not be done. Science informs our choices. It doesn't make them.

While I agree with the fundamental, the is-ought divide, I think this paper is basically saying "based on a load of measurements that are obviously goals of our society, criminalisation of drug offences is acting against our own interests".

This is only a utilitarian ethical argument. If people approach from a deontological perspective, the measurements may not matter.

Suggesting a policy prescription in the title of a scientific study puts the legitimacy of the study in complete question. The researchers must approach the question as one which can be falsifiable, else it's not science.

> Suggesting a policy prescription in the title of a scientific study puts the legitimacy of the study in complete question.

Does it?

> The researchers must approach the question as one which can be falsifiable, else it's not science.

Isn't that what this study was—they had a hypothesis which they wished to test: "We should do X." Then they did whatever they reported in their paper, and found that their hypothesis was supported by the evidence.

A policy prescription requires weighing of values, which a scientist should avoid at all costs. It's quite literally the point of science, to understand how the world works without injecting your personal biases about how the world should work.

The point of science is, ideas are tested by experiment. If you don't start your experiment with a hypothesis about the way the world works, then you're just p-hacking.

If the authors were striving for objectivity, they'd skip discussion of values and interests entirely and talk solely about what effect drug policies had on certain quantifiable metrics --- all in a factual and dispassionate way.

A scientific paper is a report on the state of reality and should not contain a case for any sort of action.

That's fine, but it should say that (and probably does; the journalism is probably the one changing the story).

How the heck is this comment downvoted? The is-ought gap is unpopular here?

HNs username must be Sam Harrises biggest fans...

It's not that the is-ought gap is unpopular: it's that the gap doesn't even make it into the debate. We're dealing with an epistemic regime in which it's acceptable to engineer desired "ought" outcomes by controlling who gets to make statements about what "is". That is, is no "is" independent from power dynamics, oppressor and oppressed. "Is", in this model, is just another form of narrative that serves the powerful.

In this world, you should expect downvotes for arguing that science informs us about objective reality, because people of a certain mindset see scientific results that contradict their worldview as just another form of oppressive discourse intended to subjugate the powerless. To them, science is just one of many equal and arbitrary forms of rhetoric and power gaming.

Ehhh, most studies in this realm have a 'results' section and 'discussion' section. I disagree that experts in their field can't have an opinion laid out in the discussion section, given that it follows from the analysis and results from the previous sections.

That being said, the page is currently timing out, so I can' even read the article...

'Results' present the new knowledge gained in a short form. 'Discussion' is normally an invitation to scientific discussion, not a political one. But even if it has the latter, it's still not the result of a study, just a piece of opinion that authors saw appropriate to include.

The paper [1] really seems more like a call to action from ethicists published in a scientific journal as a study for whatever reason.

[1] https://www.tandfonline.com/doi/full/10.1080/15265161.2020.1...

Yes, the authors published their article in the Journal of Bioethics, which is exactly where this type of article/argument is supposed to be published.

What is your complaint exactly?

I disagree that the purpose of ethics as an academic discipline is to tell people what to do.

You started off with the idea that research papers "can't tell you what should be done", which it seems like you disagree with now?

> 'Discussion' is normally an invitation to scientific discussion, not a political one

To think that those are so distinct is perhaps naive.

Thanks for the link, it does seem to be a call to action as you said, but backed by many studies across many disciplines. Which would make it both a scientific and political.

It was more meant as an imperative, as in, "a president can't despise his own citizens", even though we empirically know it's not true, but...

To me is a huge abuse of the authority of science. The results of science normally lie above any democratic discussion. No mater how you vote and what your values are COVID-19 exists, and global warming exists either. If you publish an opinion piece - even if it's a well-informed opinion - as a scientific paper, you are basically saying "here's what the policy should be, and you're not entitled to discuss it, because we know better".

I'm not against scientists having an informed opinion an political matter, and I'm not even against scientific journals publishing opinions if they consider them important enough, but there should be a distinction in this case between the part where science is published and the opinion column.

They do that already... Distinguishing between 'results' and 'discussion'...

Eh, plenty of studies I've seen, from economics to environmental science, has some policy recommendations at the end. It's a problem but if you're going to discard this study you have to discard many others.

Reply on news.ycombinator.com