rba
Playing Possum: The Variability Hypothesis
I’d lean toward the latter, but I just don’t think we know.
Or you’d be persuaded if they switched “a” to “i” back to “a” in biologically implausible time?
This is handwaving. To make such a claim, you need reference to the mechanism or at least the anatomy. As a source, sperm whales don’t have a larynx, but they have phonic lips. As a filter, they don’t have tongues or lips or throats in the way we do, but they have a distal air sac. Is that what they’re “changing the length” of ? Is that what they’re “changing the tension” of?
“I think for your speed argument to be relevant, you need to show that the whales are switching between sounds at a biologically implausible speed”. Yes, this is exactly right. Look at Fig 6 above.
The spectral pattern switches from “a” clicks to “i” clicks in at most 100 ms (there might be 30 ms between clicks and each of those click samples is 5 ms). That’s really fast. It rivals the human motor system governing our articulation. It’s ambitious to claim whales have such capabilities without detailed anatomical references that the authors don’t make.
As I noted elsewhere in these comments, the intracoda timescale is borderline as well, and would rival human abilities. A huge claim.
I hadn’t considered the overtone issue. I need to think about that…
The intraclick and intracoda “adjustment by muscles” is precisely what I’m disputing.
There are several possibilities that I hope sharpen this up.
The whales have motor articulatory control at the intraclick level, which is why you see the different spectral “vowels” in a single click. Since you see this in 5 ms intraclick samples, they must have control on those timescales or smaller. As the bird people I cite above state, this seems completely implausible, possibly by several orders of magnitude in time. For what’s it’s worth, my correspondence with the authors on pubpeer makes it clear they’re not claiming this.
The whales have motor articulatory control at the intracoda level, which is why you see the mixed coda types (Fig 6.). However, the spectral change you see in these example occurs within at most the 100 ms range (like you said, say maybe 10 ms per click plus 30 ms in between) which is about human-level abilities. Unlike 1, this isn’t completely implausible, but it seems like a very ambitious claim.
The whales don’t have motor articulatory control at the intracoda level, and the mixed codas actually represent a beaming artifact/interference pattern of the kind outlined in the post. However, once you concede that, it becomes parsimonious to say that actually “i” vowels are themselves a beaming artifact. This also explain the intraclick pattern as well.
It’s a little unclear, but I think the authors are claiming 2 is correct. I think they’d need to concede intracoda articulatory control to get this to work (or at least to explain the large minority of mixed-type codas). I’m claiming 3 is correct.
The mixed codas seem like strong evidence against this view, or at least strong evidence against the claim that the “a”/”i” are an example of what you’re describing.
This is a great point with regards to how plausible the various scientific animals communications claims actually are, but to my eye, the CETI people look motivated in the same way as the Koko people or your local crazy cat lady.
How Articulate Are the Whales?
This!
It does look like the PR department drafted this response...
I don’t think this is right, but maybe I will calculate new indices to account for this. For one thing, I think the idea is that Pentagon employees are less likely to frequent gay bars than other DC residents. For another, there exist tweets in the dataset that look like this:
HIGH activity is being reported at the closest Papa Johns to the Pentagon. Freddies Beach Bar is reporting abnormally low activity levels for a Saturday at 7:11pm ET. Classic indicator for potential overtime at the Pentagon.
Does Pentagon Pizza Theory Work?
Yes. I mention the differential son career path explanation briefly. You can see evidence of that in there.
Thanks for the thoughtful comment. I’ll try to address these remarks in order. You state
Furthermore you only examine all cause mortality, whereas the study examines deaths from specific diseases.
They also use overall mortality (Web Table 10), which is what I was trying to reproduce and screenshotted. The significance figures aren’t really different than those for the regressions broken down by mortality cause (Web Table 15), but the rate ratios for the all cause mortality ratios are clearly smaller in the disaggregated regressions because people die from other stuff. I mostly ignored the rate ratio sizes and focused on the significances here, but agree the most powerful effects probably result from the disaggregated regressions.
Removing all 14 controls will kill your statistical power further, because the presence of things like “adequate sanitation” and “health expenditure” will definitely affect the mortality rate. … they seem plausible on priors (I’m not familiar enough to know if they’re standard), and seem to improve the predictive power of the model.
This is a fundamental misunderstanding of how controls work and what they’re supposed to do. Just yesterday Cremieux wrote a pretty good piece on this very topic. The authors include these controls with little thought to the underlying causal mechanism. Their only remarks on them at all in the supplemental material are
These controls, combined with country fixed-effects models and other isolation factors, help refine the relationship between USAID per capita and changes in mortality over time.
This isn’t necessarily true at all. Consider one of the controls: Health expenditure as a percent of GDP. Is that a confounder, influencing USAID spend levels and mortality rates simultaneously? Is it a collider, caused by increased or decreased USAID spend and mortality changes? In the former case, yes, go ahead and control for it, but if it’s the latter, it screws up the analysis. Both are plausible. The authors consider neither.
I did apparently miscount the controls. It’s unclear why their own spec on page 13 miscounts them as well.
You mention the Monte Carlo simulations aren’t comparable. This is a fair, and I really like the explanation. I didn’t really touch on that aspect of this analysis in the post, but you’ve persuaded me I’m making a cheap point.
“Also it makes no sense to say “some of the choices they make to strengthen the result end up being counterproductive”.
This was unclear, and I regret it. I meant counterproductive in the sense of complicating the analysis. I’m still not clear how they got such strong, consistent results. My suspicions is careful control selection as alluded to above.
I tried reproducing that Lancet study about USAID cuts so you don’t have to
Correct. The data is suggestive, but does not make for very strong evidence of poisoning. Personally, I think the case for increased osteoporosis is strong for a variety of reasons, but we need information to answer if this was because of a mass-poisoning. .
This is a very interesting post.