Unknown Unknowns

This article discusses an important methodological issue of wide interdisciplinary importance: how to deal with "unknown unknowns" so as not to be overconfident about scientific results, without throwing out the baby with the bathwater and retreating to a nihilist position that we know nothing. 

It demonstrates an approach to estimating the uncertainty of results even though we don't know the precise sources of the uncertainties, including possible researcher fraud.

Uncertainty quantification is a key part of astronomy and physics; scientific researchers attempt to model both statistical and systematic uncertainties in their data as best as possible, often using a Bayesian framework. Decisions might then be made on the resulting uncertainty quantification -- perhaps whether or not to believe in a certain theory, or whether to take certain actions. 

However it is well known that most statistical claims should be taken contextually; even if certain models are excluded at a very high degree of confidence, researchers are typically aware there may be systematics that were not accounted for, and thus typically will require confirmation from multiple independent sources before any novel results are truly accepted. 

In this paper we compare two methods in the astronomical literature that seek to attempt to quantify these `unknown unknowns' -- in particular attempting to produce realistic thick tails in the posterior of parameter estimation problems, that account for the possible existence of very large unknown effects. 

We test these methods on a series of case studies, and discuss how robust these methods would be in the presence of malicious interference with the scientific data.

Peter Hatfield, "Quantification of Unknown Unknowns in Astronomy and Physics" arXiv:2207.13993 (July 28, 2022).

The Ethics of Lunar And Mar Exploration

I predict that planners of new missions of the Moon and Mars will completely ignore the anti-colonial ethical analysis of this paper.

[Submitted on 15 Oct 2020]

Ethical Exploration and the Role of Planetary Protection in Disrupting Colonial Practices

Frank Tavares, Denise Buckner, Dana Burton, Jordan McKaig, Parvathy Prem, Eleni Ravanis, Natalie Trevino, Aparna Venkatesan, Steven D. Vance, Monica Vidaurri, Lucianne Walkowicz, Mary Beth Wilhelm

We recommend that the planetary science and space exploration community engage in a robust reevaluation concerning the ethics of how future crewed and uncrewed missions to the Moon and Mars will interact with those planetary environments. This should occur through a process of community input, with emphasis on how such missions can resist colonial structures. Such discussions must be rooted in the historical context of the violent colonialism in the Americas and across the globe that has accompanied exploration of Earth. The structures created by settler colonialism are very much alive today, impact the scientific community, and are currently replicated in the space exploration communities' plans for human exploration and in-situ resource utilization. 

These discussions must lead to enforceable planetary protection policies that create a framework for ethical exploration of other worlds. 

Current policy does not adequately address questions related to in-situ resource utilization and environmental preservation and is without enforcement mechanisms. Further, interactions with potential extraterrestrial life have scientific and moral stakes. Decisions on these topics will be made in the coming decade as the Artemis program enables frequent missions to the Moon and crewed missions to Mars. Those first choices will have irreversible consequences for the future of human space exploration and must be extremely well considered, with input from those beyond the scientific community, including expertise from the humanities and members of the general public. Without planetary protection policy that actively resists colonial practices, they will be replicated in our interactions and exploration of other planetary bodies. The time is now to engage in these difficult conversations and disrupt colonial practices within our field so that they are not carried to other worlds.

Comments:	A submission to the Planetary Science and Astrobiology Decadal Survey 2023-2032
Subjects:	Instrumentation and Methods for Astrophysics (astro-ph.IM); Earth and Planetary Astrophysics (astro-ph.EP)
Cite as:	arXiv:2010.08344 [astro-ph.IM]

The Ethical Obligations Of Scientists

Yes, it is funny. But, should scientists spend more time thinking about whether they should, instead of merely whether they can? This certainly isn't part of the culture we instill as we train scientists.

The quote is from Jurassic Park (1993), which is closer to the realm of the possible than you might think.

Remembering A Real Scientific Hero

Four years ago [as of August 18, 2019], Syrian archaeologist Khaled Al-Asaad was murdered by Daesh after refusing to give away the location of hidden artefacts in Palmyra. He gave his life for the heritage he had dedicated his life to, in defiance of brutality, extremism, and authoritarianism.

From here.

Sargin and Faizal Refuse To Amend Physics Paper While Knowing It Is Wrong

Good physicist Sabine Hossenfelder rightly calls out the authors on a new paper about the black hole physics of loop quantum gravity, who make a claim in the title of the paper that they later admitted to her was incorrect when she contacted them and pointed out their error.

But, they wouldn't change the paper and minimized their error, even though the claim that they have wrong is in the very title of their paper. This pretty much totally offends the entire process of posting pre-prints at arxiv.org, and the entire peer review process. So, she wrote a blog post that will track back to the paper.

You might get away with the impression that we have here two unfortunate researchers who were confused about some terminology, and I’m being an ass for highlighting their mistakes. And you would be right, of course, they were confused, and I’m an ass. But let me add that after having read the paper I did contact the authors and explained that their statement that the LQG violates the Holographic Principle is wrong and does not follow from their calculation. After some back and forth, they agreed with me, but refused to change anything about their paper, claiming that it’s a matter of phrasing and in their opinion it’s all okay even though it might confuse some people. And so I am posting this explanation here because then it will show up as an arxiv trackback. Just to avoid that it confuses some people.

The offending authors are Ozan Sargın and Mir Faizal, and the paper is: "Violation of the Holographic Principle in the Loop Quantum Gravity" http://arxiv.org/abs/1509.00843.

My Genome (An Executive Summary)

My 23 and Me personal genome testing preliminary results are in (much more promptly than advertised, for what it is worth):

Highlights

Y-DNA haplogroup: E1b1b1a2* (E1b1b1 is also known as E-M35, E1b1b1a is E-M78 and my own haplogroup is E-V36/V13).

mtDNA haplogroup: H1b

Broad ancestry composition (out of 31 regional groupings including speculative estimates):
European 99.9%
*Northern European 94.8%
**Finnish 38.1%
**Scandinavian 16.4% (Sweden-Norway-Denmark)
**British and Irish 6.3% (UK and Ireland)
**French and German 1.1% (Germany, Netherlands, France, Switzerland, Austria, Belgium)
**Broadly Northern European 32.9%
*Broadly Southern European 2.6%
*Broadly European 2.6%
East Asian - Yakut 0.1%
Unassigned < 0.1%

Neanderthal Ancestry 2.9% (79th percentile; 82nd percentile for Europeans).

Feel-Good Gene: AC (mixed ancestral "C" and feel-good "A" type)

Analysis

My Y-DNA haplogroup, which is most common in Greece and Albania, is surprising as it is rare in the places that are home to my ancestors (Protestant Germany near the former West German-East German border), but not shocking.  It could have been a part of either a first wave early European farmer migration, or a Bronze Age migration, to the north and west from the Balkans.  The source of this Y-DNA was probably not recent as my genome lacks any East European and only a very modest amount of Southern European ancestry, and no Near Eastern ancestry.

My combined Y-DNA and mtDNA haplogroups, while not terribly remarkable for someone of my ancestry, would also be quite normal for a man of Berber origin, until one gets to the very last couple of mutations in the Y-DNA haplogroup that are European specific.

I was surprised to be more Finnish than Scandinavian.  My Finnish ancestors are Swedish language speaking Finns and I'd expected the Scandinavian component to be half or more of the total.  But, apparently, Swede-Finns absorbed more native Finnish ancestry than I'd expected, although the possibility that Swedish ancestry which due to founder's effects is most common in Finland counts as Finnish could also be a factor.

The trace Yakut ancestry isn't too surprising given my substantial Finnish heritage, with a presumed cryptic circumpolar migration producing an ancestor hundreds of years ago.  This trace, which appears only on one of my chromosomes in a single bunch, also highlights the fact that you can have ancestry which is not reflected in either of your uniparental haplogroups.

The higher than average Neanderthal ancestry was also expected as Finns have above average hunter-gatherer ancestry in Europe and European hunter-gathers had higher levels of Neanderthal ancestry than post-Last Glacial Maximum migrants from the Near East.

There is less specifically British and Irish ancestry than I'd expected and less German ancestry than I'd expected, and I hadn't expected any Southern European ancestry.  But, as most of my father's ancestry is ascribed to "Broadly Northern European", it is fair to say that he's pretty much a Northern European mutt.

The "Feel-Good Gene" is one where I'd thought it was likely that I'd have a mixed or completely derived type, as this gene is particularly common in Finns and is associated with phenotypes that fit my personality.

Some spot checking has revealed that there are a not insignificant number of SNPs for which no result was obtained, although they are clearly a small minority and I don't expect 100% reads on everything in a mailed in spit test analyzed for an average price of less than $99 per genome.

Ethics and Motivation

I'm aware that by posting this information, I am also posting without their consent, by inference, information about the uniparental genetics of my father, one my two children, my brother, my uncle, my male paternal first cousin, two maternal uncles, a maternal aunt, four maternal first cousins, and two children of one of my maternal first cousins, for a total of at least fourteen people, in addition to large numbers of deceased people (e.g. my grandparents and mother) and many more distant relatives.

On the other hand, the information is intrinsically a part of me and hence morally belongs to me to use as I see fit, anyone using it has to tease out the family relationships to do so, and there is not strong stereotypes attached to Y-DNA and mtDNA haplogroups at this point, although there is arguably mild anti-Neanderthal stigma that this post my help to dissolve (or enhance, depending upon what you think about me).  And, I think it is worthwhile for someone who blogs a great deal about genetics to disclose his own, after benefiting so much from the disclosures of others.

Moreover, the kind of information that I have disclosed already really provides very little information that couldn't be inferred from publicly available information on my family tree, ancestry, appearance, and demeanor.  I would think twice before posting my whole genome, or portions of it with greater medical significance.

I also think it is important for the public to put a face on genetic information, and make my own contribution here.

Also, knowledge of your own genes inevitably colors your interest in our research, something which I should disclose to my readers.

For example, Maju just blogged on mtDNA H1 in Cantabria from 19,000 years ago, something that it had previously not been clear existed.  This area in Northern Spain was a refuge in which modern humans survived the deepest part of the last Ice Age, with glaciers that reached their greatest extent around 20,000 years ago.  So, this find disproves the hypothesis that everyone in Europe who survived the last Ice Age was mtDNA U with all other haplogroups arriving more recently from the Near East, which had been the prevailing paradigm based upon ancient DNA findings to date, before this discovery.

Knowing that I bear this mtDNA haplogroup makes that discovery one of more personal significance than it otherwise would have been, although it was surely noteworthy in any case.

Spanish Baby Snatching

A Spanish Civil War era scheme that continued until the 1990s where almost 300,000 "Red" women's babies were stolen from them shortly after delivery and placed in conservative families in an effort to make their children more conservative and reshape the nation is starting to be unraveled.

This was one of the grandest nature v. nuture experiments in modern history, although the results are unknown and the methods violated every principle of experiments on human subjects.