A man with a conviction is a hard man to change.

From When Prophecy Fails -- A Social and Psychological Study of a Modern Group That Predicted the Destruction of the World by Leon Festinger, Henry Riecken, and Stanley Schachter. Festinger's was a ground level investigation of a cult interpreted through a new theory about the psychology of belief and knowledge. These are the opening paragraphs of the book.

A man with a conviction is a hard man to change. Tell him you disagree and he turns away. Show him facts or figures and he questions your sources. Appeal to logic and he fails to see your point.

We have all experienced the futility of trying to change a strong conviction, especially if the convinced person has some investment in his belief. We are familiar with the variety of ingenious defenses with which people protect their convictions, managing to keep them unscathed through the most devastating attacks.

But man's resourcefulness goes beyond simply protecting a belief. Suppose an individual believes something with his whole heart; suppose further that he has a commitment to this belief, that he has taken irrevocable actions because of it; finally, suppose that he is presented with evidence, unequivocal and undeniable evidence, that his belief is wrong: what will happen? The individual will frequently emerge, not only unshaken, but even more convinced of the truth of his beliefs than ever before. Indeed, he may even show a new fervor about convincing and converting other people to his view.

How and why does such a response to contradictory evidence come about? This is the question on which this book focuses. We hope that by the end of the volume, we will have provided an adequate answer to the question, an answer documented by data.

Let us begin by stating the conditions under which we would expect to observe increased fervor following the disconfirmation of a belief. There are five such conditions.

1. A belief must be held with deep conviction and it must have some relevance to action, that is, to what the believer does or how he behaves.
2. The person holding the belief must have committed himself to it; that is, for the sake of his belief, he must have taken some important action that is difficult to undo. In general, the more important such actions are, and the more difficult they are to undo, the greater is the individual's commitment to the belief.
3. The belief must be sufficiently specific and sufficiently concerned with the real world so that events may unequivocally refute the belief.
4. Such undeniable disconfirmatory evidence must occur and must be recognized by the individual holding the belief.
The first two of these conditions specify the circumstances that will make the belief resistant to change. The third and fourth conditions together, on the other hand, point to factors that would exert powerful pressure on a believer to discard his belief. It is, of course, possible that an individual, even though deeply convinced of a belief, may discard it in the face of unequivocal disconfirmation. We must therefore, state a fifth condition specifying the circumstances under which the belief will be discarded and those under which it will be maintained with new fervor.
5. The individual believer must have social support. It is unlikely that one isolated believer could withstand the kind of disconfirming evidence we have specified. If, however, the believer is a member of a group of convinced persons who can support one another, we would expect the belief to be maintained and the believers to attempt to proselyte or to persuade nonmembers that the belief is correct.

These five conditions specify the circumstances under which increased proselyting would be expected to follow disconfirmation.

Deceit, misdirection, and obfuscation

This is fascinating.

As a consequence of the unexpected loss of their candidate to Donald Trump, the mainstream media has been in overdrive coming up with increasingly preposterous reasons for why that occurred. Regret and remorse are one thing, but the response has become spectacularly unhinged.

All of the responses are laughable and many of them are irresponsible. Some almost dangerous. The one that has most irritated me has been the brouhaha over "Fake News." First, there is little evidence that any news materially changes peoples opinions other than solidifying them in their preconceptions. There is certainly no evidence that semi-mythical sources of "fake news" proliferated at the end of the campaign and had a measurable effect. It is a claim with no empirical evidence.

More importantly, this smacks so clearly of totalitarianism and a repudiation of our cultural and constitutional commitment to freedom of speech and the marketplace of ideas. It is repulsive to see so many mainstream media outlets trying to claim that their product is real news while the news product from others with whom they disagree is "fake news." It is evidence of either willful ignorance, ideological blindness, or venal commercial interest.

Beyond all that, there is the interesting epistemological exercise of identifying "fake news." It is in this context that I read Now you can fact-check Trump’s tweets — in the tweets themselves by Philip Bump. His own blinders seem to preclude him seeing that he comes across as a democratic operative with a byline. But it also highlights the nitpicking and linguistic gymnastics that these advocates have to indulge in.

Take this tweet as an example.

So I made a Chrome extension that slips context and corrections into Trump's tweets themselves. https://t.co/7H6SokzHTB

— Philip Bump (@pbump) December 16, 2016

Click to enlarge

Donald Trump's claim is "Are we talking about the same cyberattack where it was revealed that head of the DNC illegally gave Hillary the questions to the debate?"

The interesting thing here is how much information can be conveyed within the constraints of 147 characters.

How might a person understand Trump's claim. Donna Brazile was, until she resigned in disgrace, the temporary head of the DNC. The DNC was hacked for their e-mails (or the emails were leaked, it is still not clear). The e-mails did reveal extensive wrong-doing on the part of various members of the DNC and the establishment. Brazile had, earlier in the campaign before she became interim head of the DNC, shared CNN questions with the candidate she supported, Hillary Clinton. Brazile almost certainly had a nondisclosure agreement precluding her from sharing the confidential questions.

So where are the factual weaknesses in this 136 character tweet?

1) Probably should have made clear that Brazile was head of the DNC after the occasion when she shared the confidential questions from CNN with Clinton.

2) Should have provided evidence that Brazile had a non-disclosure agreement with CNN that precluded her from sharing the questions with any campaign members.

Of these two, I see only the second as material. Does Trump know that Brazile had a nondisclosure agreement? Certainly, that is a standard and routine clause in handling sensitive data. His statement that she "illegally" shared is true if she was subject to a non-disclosure agreement which I suspect most people assume she was. If, for whatever reason, she had not agreed to non-disclosure, then shame on CNN. However, in terms of the substance of Trumps tweet, it becomes a marginal semantic issue. Without a non-disclosure agreement (or its equivalent), then it would be more correct to say that she "immorally" shared, or "inappropriately" shared, or even "she breached everyone's trust by sharing."

From this man-in-the-street perspective, there is nothing "fake" about Trump's tweet, and certainly not within the constraints of 147 characters. A little more clarity and specificity would have been nice but the tweet gets to the substance of what most people would consider the issue - a senior DNC insider and political ally of Clinton's breached convention (and maybe her legal contract) by inappropriately sharing confidential information in order to privilege and advantage Clinton in a debate. That is a serious allegation and it is supported by the publicly available information.

Now Bump wants to claim that there is something wrong with these facts and has built a chrome extension to recast Trumps tweet. What is the substance? Bump's claim is that:

Click to enlarge

He seems to consider three things important in judging the appropriateness of Brazile's actions.

1) Brazile leaked the questions before she became temporary DNC leader.

2) Brazile was not obligated to maintain the confidentiality of the questions.

3) The possibility that the hack revealing the illegal and inappropriate insider actions might have (but not yet determined) originated with someone in Russia.

To someone not in the inner circles of incestuous power and intrigue, these seem picayune at best.

If you commit a wrongful act, it is wrong regardless of your position. Granted it might take on greater salience, but the wrongfulness remains as the foundational issue. A Principal of a school lying about his educational credentials and a student in that school lying on their college application have both lied. The lying is the issue even though we might (would that it were true) accord greater consequence and accountability to the Principal than to the student. Thus the distinction about when Brazile committed her transgression is of less importance than simply having committed it.

Whether Brazile was contractually bound not to reveal the questions is the most salient issue. Certainly, everyone assumes she would have been so bound. Bump merely asserts that she had no such contractual obligation and therefore there was nothing illegal. That seems improbable but it awaits confirmation. You cannot simply assert your way to Truth in an argument.

Finally, the issue of Russia is a classic logical fallacy, a non sequitur. In judging Clinton and Brazile's actions, it doesn't matter who revealed the wrong-doing. It doesn't matter whether the evidence emerges because of free-lance Romanian hacking, state sponsored Russian hacking, or whether it was a disgruntled DNC insider as Wikileaks insists. Bump's flourish is also a standard rhetorical strategy of misdirection.

So what are we left with? It appears that Trump's tweet is materially accurate. He could have clarified timing of Brazile's position but that wouldn't change the substance of the charge. He could have provided evidence of a confidentiality contract, but it is reasonable to assume that there was one. His claim seems, on balance completely accurate.

Bump's Chrome extension does nothing but introduce obfuscation. It works if you already want to believe that nothing wrong occurred but it does not actually refute anything in Trump's claim. It implicitly acknowledges that Brazile leaked the questions. It assumes that, against standard conventions, there was no confidentiality contract. It makes a misdirecting and unsubstantiated claim about who might have been responsible for shining light on the DNC misdeeds.

No wonder no one trusts the press. Even their fake news strategy is based on deceit, misdirection, and obfuscation.

As if to cement this negative impression, Bump offers a second example.

Click to enlarge.

Trump makes three claims.

1) His electoral victory of 306 to 232 represents a landslide.

2) That there were millions of people who voted illegally.

3) That if you subtract those who voted illegally, Trump would have a popular vote majority as well.

I certainly think Trump is on weaker ground here but when we are dealing with facts, Bump is in an even weaker position.

Bump's fact-check response to the three claims is:

1) 306 to 232 doesn't constitute a landslide.

2) "There is absolutely no evidence that there were a significant number of votes cast illegally."

The claim that 306 to 232 constitutes a landslide is not fact checkable. It is an opinion. For Bump to assert that Trump "did not win a landslide in any sense" is simply untrue. I agree with Bump that I would not characterize 306 to 232 as a landslide. A "commanding victory" perhaps. A "solid victory" perhaps. But that's me. The fact is that Trump received 57% of the electoral votes, or, alternatively, he received 32% more electoral votes than Clinton.

I am sure that at least some people might judge that as a landslide. Regardless, the claim rests on an opinion as to what constitutes a landslide and therefore is not capable of being determined as True or False. If Bump had claimed that Trump's characterization of 306 to 232 was disputable, Bump would be on firm ground. However, Bump destroys his position by claiming that there was no way in any sense to claim that 306 to 232 is a landslide. Bump makes his own claim false.

Similarly, Bump overreaches with the claim that "There is absolutely no evidence that there were a significant number of votes cast illegally." 136 million people voted. At least some illegal votes were cast. Were there hundreds, thousands, tens of thousands of illegal votes? I have no idea but I am pretty confident that there was some material number of illegal votes given that many states operate substantially on the honor system in terms of voting and given that we have video recording of political operatives bragging about how they increase illegal voting.

But for Bump to claim that there is absolutely no evidence is absurd. Given that 37% of Detroit's precincts recorded more votes than there were voters, and that 95% of the votes went to one candidate, there's more than enough real evidence to suggest that there could have been large-scale illegal voting. How widespread it might have been is a different question.

Between these two examples we have zero evidence of the Chrome extension serving as a fact-checking exercise. What the extension does is give an opportunity for Bump to introduce his opinion to others, not address anything that is factual.

My interpretation is that Trump is circumventing the mainstream media completely by his tweets, his dominating press conferences and by his rallies. Fact checking and fake news allegations are coming across as the traditional mainstream media insisting on their relevancy and grasping at trying to control the narrative which is slipping away from them. Paradoxically, their actions, such as with this Chrome fact-checking extension seem more likely to discredit them further.

An ideology of joylessness and depression

The mind of the socialist is inimical to human happiness.

I bet the Christmas party at The Guardian is a riot pic.twitter.com/45a4uOd33i

— James (@JamesFl) December 5, 2016

Don't turn your back on experience

My father's career was in the international oil industry and I grew up around the world, wherever there was oil or the hope of discovering oil. I was born in the US but was two months old when I made my first international flight, to Venezuela where we lived for another four years.

Living in countries with track records of civil unrest, autocracy, dictatorships, mob-rule, xenophobia, revolutions and coups, there was always the real prospect of trouble and the possible need for emergency evacuations. My father was a red-blooded American but also an engineer and a pragmatist. One of the lessons drummed in to us from an early age was that while the American Embassy in any particular country might look big and imposing, as an American citizen you could not count on assistance from the Embassy. You had to have your own Plan A (and B and C) because the Embassy was unlikely to be willing or able to help you. If you had to seek help, go to the British, Canadian or Swiss Embassies.

As a child, that seemed demeaning of America though I sort of understood that that there were constraints on a big country that might not apply to a smaller one. I didn't like the implications of the advice but I knew it came from experience.

My father is no longer with us but I thought of him today, and imagined his smile, when I saw this alert from the State Department.

U.S. State Department warns Americans in Venezuela: You're on your own... https://t.co/bQGhSpeWLr pic.twitter.com/PaaE4SMau0

— Nathan Crooks (@nmcrooks) December 16, 2016

Some things just don't change.

In peace you reward the military for avoiding risk and in war you want the same commanders to aggressively exercise risk

From LCS Shock Trials Were Less Severe Than Navy Standard by MarEx.

In written testimony for the Senate Armed Services committee, Dr. Michael Gilmore, director of operational testing under the Secretary of Defense, said that the shock trials for the Independence and Freedom-class Littoral Combat Ships were conducted at "reduced severity" due to concerns about the possibility of damage.

"The Navy argued that the reduced severity approach was necessary because they lacked specific test data and a general understanding of how the non-Grade A systems . . . would respond to shock," he wrote.

In addition, for the test on the Freedom-class vessel, the Navy stopped the shock trials at the second of three shots. The third trial would have shocked the ship with a blast one-third less powerful than the vessel is designed to survive, but the Navy still deemed it to be too risky.

"The Navy viewed the third [Freedom-class] trial as not worthwhile because the Navy was concerned shocking the ship at the increased level of that trial would significantly damage substantial amounts of non-hardened equipment, as well as damage, potentially significantly, the limited amount of hardened equipment, thereby necessitating costly and lengthy repairs," he wrote. The service opted for a simulated third test instead.

Gilmore suggested that the shock trial results were consistent with the LCS' less-hardened design and construction. When combined with the cancelations and delays affecting the platforms' stand-off mission packages, he wrote, the vessels' limited survivability would make them vulnerable in a front-line combat environment.

I have recently seen a number of articles on the lower than expected hardiness of the LCS.

This account was particularly striking to me as I have just finished Bruce Henderson's Down to the Sea. Henderson's story is about the December 1944 incident in which Admiral Halsey sailed the Third Fleet directly into a typhoon, resulting in the loss of three destroyers, 790 sailors, nine other vessels were damaged and a hundred planes lost or damaged.

In relating the history, Henderson provides the details that reinforced a particular lesson. It is reasonably well known that the US fleet had particular shortcomings starting the war. As a particular example, US torpedoes were enormously faulty owing to inadequate testing in the inter-war years, in turn as a result of cost savings. The consequence was that for the first couple of years of the war, we sent submarines into action who, at great peril, attacked Japanese shipping only to see 30-70% of their torpedoes fail.

Henderson's account brings forward a more immediate lesson. My summary is that in peacetime, the Navy was highly focused on saving money and avoiding accidents. Over time this rewarded very cautious commanders. Beyond the material deficiencies discovered in the first year of the war, there was also a cultural deficiency. Vessel commanders who had spent their entire careers avoiding risks were being sent into battle where success is contingent on aggressive action. In tracking individual commanders, it is very clear from the accounts in Down to the Sea that one of the first year makeovers in the Navy was a changing of the guard among the cadre of commanders. Those who adhered to the entirety of their training (and avoided risks) were demoted and transferred state-side to make room for more aggressive commanders (who often had been at the margins of their career tracks owing to their aggressive nature).

My take-away from Down to the Sea was that any military going into war after a long duration of peace has to be mindful of both the adequacy of their equipment but also, and probably more importantly, has to be prepared to rapidly shift their culture from risk-containment to risk-engagement.

The LCS story above looks to me like a re-enactment of exactly the same dynamics of the inter-war years leading up to 1941.

Failures in forecasting

Paul Krugman has long been an ideological commentator. I did not know that the conversion from Nobel prize winning economist to failed pundit occurred as long ago as this but, if true, this is a pretty spectacular fail. So spectacular, I am inclined to believe he is being framed.

@mijaresluis419 @GayPatriot Yes... he knows so, so much. Defer to his brilliance. pic.twitter.com/SBozKrnzHp

— Ginger Taylor (@GingerTaylor) December 9, 2016

Well, apparently it is real, though with a minor context that mitigates it a small amount. But not much.

Balancing deep awareness of existing knowledge vs. making new connections and developing new interpretations

As the floodgates of information have opened and data and detritus flowed through since the internet, I have tried to be conscientious in my construction of a disparate but focused, traditional but innovative, exploratory but confirming portfolio of knowledge sources. I want to know more about things I am interested in, I want to discover things I don't even know I don't know, I want help connecting things that are related but appear discrete. I also want to screen out noise and low value information. There is no perfect mix and indeed the goals are inconsistent. Finding new ideas entails reviewing a lot of speculation that is simply wrong. Understanding more about things I am reasonably informed about involves getting through a lot of duplicative material. There is no easy solution.

In fact there is no algorithmic solution at all. It is simply a matter of constant personal behavior management, adjusting behaviors and practices to needs within a more strategic construct of ultimate goals. All being done within the real constraint of available time.

I found Exploration and exploitation of Victorian science in Darwin's reading notebooks by J. Murdock, C. Allen, and S. DeDeo interesting.

From the abstract:

Search in an environment with an uncertain distribution of resources involves a trade-off between exploitation of past discoveries and further exploration. This extends to information foraging, where a knowledge-seeker shifts between reading in depth and studying new domains. To study this decision-making process, we examine the reading choices made by one of the most celebrated scientists of the modern era: Charles Darwin. From the full-text of books listed in his chronologically-organized reading journals, we generate topic models to quantify his local (text-to-text) and global (text-to-past) reading decisions using Kullback-Liebler Divergence, a cognitively-validated, information-theoretic measure of relative surprise. Rather than a pattern of surprise-minimization, corresponding to a pure exploitation strategy, Darwin's behavior shifts from early exploitation to later exploration, seeking unusually high levels of cognitive surprise relative to previous eras. These shifts, detected by an unsupervised Bayesian model, correlate with major intellectual epochs of his career as identified both by qualitative scholarship and Darwin's own self-commentary. Our methods allow us to compare his consumption of texts with their publication order. We find Darwin's consumption more exploratory than the culture's production, suggesting that underneath gradual societal changes are the explorations of individual synthesis and discovery. Our quantitative methods advance the study of cognitive search through a framework for testing interactions between individual and collective behavior and between short- and long-term consumption choices. This novel application of topic modeling to characterize individual reading complements widespread studies of collective scientific behavior.

My interpretation of this dense description: Darwin invested his time early in his career in laying down a base of knowledge from work by others. Having mastered the fields, he then shifted his epistemological focus to exploration, seeking new ideas and new connections and interpretations.

Darwin's consumption of knowledge resources in has later life were more discovery oriented than was reflected in the nature of knowledge materials being produced. In other words, most books published focused on disseminating received knowledge whereas most books Darwin read were more focused on establishing different interpretations, new connections, and simple discovery.

First seek the data and then spot the patterns

I came across Climate Scientists Take to the Streets to Protest Trump by David Kirby.

On Sunday, Donald Trump told Fox News that “nobody really knows” whether climate change is real. But scientists who staged a rally in San Francisco on Tuesday had a stinging rebuke for the president-elect: It’s real, it’s here, and we are bracing for an epic battle against your policies and appointments.

“The election result was not the one that many of us would like to see, but the battle goes on and must go on,” Michael Mann, director of the Earth System Science Center at Penn State University, said at the rally.

Mann spoke outside the Fall Meeting of the American Geophysical Union, where 26,000 scientists are gathering in the first major convening of climate scientists since the presidential election. He was joined by other climate researchers and activists representing Native Americans and people living in “frontline communities” near polluting refineries, power plants, pipelines, and other fossil fuel infrastructure.

Michael Mann is a fairly discredited spokesperson for climate change having been caught out in the East Anglia email leaks and propounding a climate hockey-stick research paper which was later withdrawn.

But I was interested not so much in the merits of the debate as in the reporting. Wow, 26,000 scientists turned out to protest the president-elect's threat to the Federal government financing of the research gravy train. That makes perfect sense from a human perspective. These people are trying to protect their livelihoods.

But then, just as my eye was about to sweep away to the next article, I got a nagging sense that something wasn't quite right. That 26,000 number. Is it 26,000 protestors or is it 26,000 who attended the convention, some portion of whom might have joined the protest. As it turns out, it is the latter. 26,000 at the convention and a smaller number at the protest. How small?

That took some searching. Nobody seems to be reporting the actual crowd size other than some vague imputations that it was large. I went looking for some pictures. The best source seemed to be here. From the pictures, it appears that there are about 30 actual protestors (the people up front with signs.) How about people in the crowd? That is a lot harder to estimate. For one, you can't tell who are rubberneckers and who are actual protest participants. But let's take the whole crowd. I would estimate, judging from all the pictures, that there might be perhaps 250 people in the crowd. Let's double it for a really safe margin of error - 500 people. 530/26,000 = 2%.

2% of climate change scientists thought it was worthwhile protesting the perceived antagonism of the incoming administration. That almost seems the inversion of the common, though inaccurate, claim that 97% of scientists support anthropogenic climate change. Instead, the protest would suggest "98% of climate change scientists don't consider their research worth defending."

That's quite a different story than the first spin. Where does spin end and fake news begin?

Piping the penguin

Brilliant.

Who would have believed that the perfect Wikipedia photo caption could have been improved upon? pic.twitter.com/pLedKWbs1o

— Alan Ferrier (@alanferrier) December 13, 2016

Open and transparent communication leads to superior outcomes

This is interesting but I can't tell how interesting because it is behind a firewall. From the abstract of Durable Coalitions and Communication: Public versus Private Negotiations by David P. Baron, Renee Bowen, and Salvatore Nunnari. Key elements in bold.

We present a laboratory experiment to study the effect of communication on durable coalitions – coalitions that support the same allocation from one period to the next. We study a bargaining setting where the status quo policy is determined by the policy implemented in the previous period. Our main experimental treatment is the opportunity for subjects to negotiate with one another through unrestricted cheap-talk communication before a proposal is made and comes to a vote. We compare committees with no communication, committees where communication is public and messages are observed by all committee members, and committees where communication is private and any committee member can send private messages to any other committee member. We find that the opportunity to communicate has a significant impact on outcomes and coalitions. When communication is public, there are more universal coalitions and fewer majoritarian coalitions. With private communication, there are more majoritarian coalitions and fewer universal coalitions. With either type of communication coalitions occur more frequently and last longer than with no communication. The content of communication is correlated with coalition type and with the formation and dissolution of durable coalitions. These findings suggest a coordination role for communication that varies with the mode of communication.

One might infer that realized communication among parties, whether via open public communication or through private communication channels is superior to no communication. It appears that the health of the coalition is assisted by open and transparent communication.

Sounds like a thorough endorsement of the First Amendment and the marketplace of ideas.