Comment by MisterMower

Ok, I'm pretty sure I understand your claims. I'll divide up my response in sections with `---` addressing each of them. I'll use ILN for "international laws and norms". (My edits are finally done as of 10:33 am eastern.)

---

> Why do you think international law constrains nation states, despite much evidence to the contrary, including today’s events?

For clarity, I didn't say nor mean «ILN determine/constrain actors in all cases». I conveyed, to the best of my ability: «ILN are a useful factor to include in a model»".

In the text below, I'll elaborate quite a bit on why ILN matter as factors to include.

---

As I read your comment, I'm interpreting it as claiming to argue against «ILN is a useful factor to include in a model». You put forth an argument saying it doesn't matter. I'll argue against your argument.

But I should pause. I should not assume... I don't know your age, your nationality, your educational and/or professional background, and so on... Do you know what I mean by a quantitative model? Even people with Ph.D. can easily talk past each other on this. I'll give just one example of what I'm talking about: [1]. It is likely easier to grok than a dense statistical analysis 'locked up' in an academic paper.

Here is some context about why I care about this. Professionally I've worked as a software developer, entrepreneur, machine learning & statistical researcher and analyst, and lots more. At my core I both (a) build technology to solve human problems and (b) enjoy building things because it is fun and enabling. About ~10 years ago I build a search engine to surface quantitative models. One of my key underlying drivers is to help people to move past merely talking about stuff. Talk is cheap and imprecise. There are many kinds of models, none perfect, but the use of them is vastly better than pretending like any one framing _is_ reality. Recognizing a model as a model is the first step. Then you can step back and figure out "what is this model useful for?". With models we can put things 'on paper' and point and them and say things like "what happens when we factor in X"? We don't have to fixate on one model. We can be fluid and think about what we're trying to understand and predict.

---

> When convenient they will use international law and norms as justification for actions they would take regardless. When inconvenient, they will just ignore them.

This is both a false binary and (for lack of a concise phrase) 'starting very late in the causal chain'. I'll start with the first and then explain the second afterwards.

1. The above is a false dichotomy. There are at least three other cases. ILN are (to some degree):...

A. ... imbued in a leader such that they don't even _want_ to venture too far outside the parameters of 'acceptable'.

B. ... perceived to have consequences that need to be accounted for. Over time the leader will more or less compare their perceptions of reality to what happens in reality.

C. ... these consequences (perceived and actual) affect the decision space of a leader. A change in the decision space, in general, may change the decision. (not always of course).

There is variation in how much A, B, C apply and play out. This variation provides informational value -- a foundational concept in modeling. More informational value, ceteris parabus, makes model predictions better. We're on the same page?

2. You may have noticed that above I've already implicitly explained my second criticism. If you want to predict how leaders act, it is unwise to start the analysis at the point of 'where they made their mind up'. Instead, you want to predict how and why they form their views and goals. My claim is that factoring in ILN (international laws and norms) is useful -- it is better to factor it in than to exclude it. To skip past it is 'starting too late in the causal chain'. It would be analogous to saying "Once the trigger is pulled, the laws against murder cannot stop the bullet."

2'. If one wants to build an even _better_ predictive model, one would want to predict what kinds of leaders come to power. Imagine some counterfactuals. What if there was no UN Declaration of Human Rights? In such a world, what kind of leaders would come to power? In general, they would be different than the current slate. I'd predict to see more warlord-like behavior; there would be less trust and more military spending. Without trust, more force and threats of force are necessary. I hope you can see I've sketched out an argument for why ILN provide some shared basis for countries to cooperate based on shared values. [2]

In short, if you are arguing -- and want to continue doing so -- that ILN have no predictive value, I challenge you to build a predictive model and prove it. But I don't think you really will hold such a view once you step away from the keyboard and reflect.

I don't confuse my approximations for reality. I once rigidly held a view not too different from the one you probably do. I thought my model was 'real'. But no longer. I've found better models for predicting.

---

> My priors before this were that international law mattered a little, but this event has convinced me it’s all a farce. Exhibit A: the UN’s increasing irrelevance as we move toward a multipolar world.

Two responses:

3. It is too soon to assess the scope of international responses to Trump's invasion of Venezuela. My response here is the same as the section above. The problem is reasonable people will struggle to find consensus on how to operationalize "farce" into a prediction. It is too squishy. We can do better than this; we can build models. I've already covered this ground above.

---

> Appeals to scholars or analysts is meaningless in this context. You should post why those people think they matter, or what their reasoning is, not, “hey, guys that I think are smart say this matters.”

I don't think I'm really fully tracking you here, so I'll respond as best I can... We're standing on the shoulders of giants. We cannot discount the work of scholars throughout history.

My claims are not arguments from authority. They are arguments along the lines of 'smart people have taught us that thinking about the world in terms of models is superior to not doing so.' To use another phrase that conveys the same idea: don't confuse the map with the territory.

No charitable person would claim that my argument is anything like "hey, guys that I think are smart say this matters". If you go back and reread, now, you don't really think I'm saying that, do you? To make such a claim would be to ignore large parts of what I wrote -- it would involve tossing those out -- and myopically and forcing an uncharitable interpretation onto my words. Hacker News works better when people are charitable. [3]

---

I'll gone to some lengths to try to understand your position and explain mine. At this point, I hope you will reciprocate.

[1] https://conflictforecast.org ... this is just an example to give flavor. I'm not holding it up as a great model, but at least it is relatively clear in how it works – compared to what you'll typically see when some pundit says something about some risk of conflict: """Our forecast uses millions of newspaper articles to make the conflict forecast. In our analysis of the content of the newspaper articles we rely on a so-called topic model which summarizes the millions of articles and words into topics using unsupervised machine learning. The topic model allows us to calculate 15 topic shares for each country/month which we display in the bubbles to the right."""

[2] Some people can't or don't see this. Some people fixate on isolated individual behavior. They ignore evolutionary foundations that people exist in a social context. They don't understand or appreciate game theory or theory of mind; see https://en.wikipedia.org/wiki/Theory_of_mind .

[3] """Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith.""" https://news.ycombinator.com/newsguidelines.html