Monday, July 24, 2023

Is AI Better Regulated After the White House Meeting?

 

The easy answer is, it's too soon to tell.  But for a number of reasons, the July 21 meeting between President Biden and leaders of seven Big Tech firms, including Google, Microsoft, and OpenAI, may prove to be more show than substance.

 

Admittedly, there is widespread agreement that some sort of regulation of artificial intelligence (AI) should be considered.  Even industry leaders such as Elon Musk have been warning that things are moving too fast, and there are small but real risks of huge catastrophes lurking out there that could be averted by agreed-upon restrictions or regulations of the burgeoning AI industry.

 

Last Friday's White House meeting of representatives from seven leading AI firms—Amazon, Anthropic, Google, Inflection, Meta (formerly Facebook), Microsoft, and OpenAI—produced a "fact sheet" that listed eight bullet-point commitments made by the participants.  The actual meeting was not open to the public, but one presumes the White House would not publish such things without at least the passive approval of the participants. 

 

Browsing through the items, I don't see many things that a prudent giant AI corporation wouldn't be doing already.  For example, take "The companies commit to internal and external security testing of their AI systems before their release."  Not to do any security testing would be foolish.  External testing, meaning testing by third-party security firms, is probably pretty common in the industry already, although not universal. 

 

The same thing goes for the commitment to "facilitating third-party discovery and reporting of vulnerabilities in their AI systems."  No tech firm worth its salt is going to ignore an outsider's legitimate report of finding a weak spot in their products, and so this is again something that the firms are probably doing already. 

 

The most technical commitment, but again one that the companies are probably doing already, is to "protect proprietary and unreleased model weights."  Unversed as I am at AI technicalities, I'm not sure exactly what this means, but the model weights appear to be something like the keys to how a given AI system runs once it's been trained, and so it only stands to reason that the companies would protect assets that cost them a lot of computing time to obtain, even before the White House told them to do that.

 

Four bullet points address "Earning the Public's Trust," which, incidentally, implies that the firms have a considerable way to go to earn it.  But we'll let that pass. 

 

The firms commit to developing some way of watermarking or otherwise indicating when "content is AI generated."  That's all very well, but the answer to such a question is rarely just yes or no.  What if some private citizen takes a watermarked AI product and incorporates it manually into something else that is no longer watermarked?  The intention is good, but the path to execution is foggy, to say the least.

 

Perhaps the commitment with the most bite is this one:  "The companies commit to publicly reporting their AI systems’ capabilities, limitations, and areas of appropriate and inappropriate use."  The wording is broad enough to drive a truck through, although again, the intention is good.  How often, how detailed, and how extensive such reports may be is left up to the company.

 

The last two public-trust items commit the firms to "prioritizing" research into the societal risks of AI, and to use AI to address "society's greatest challenges."  If I decide not to wash my car today, I have prioritized washing my car—negatively, it is true, but hey, I said I'd prioritize it!

 

So what is different about the way these firms will carry out their AI activities after the White House meeting?  A lot of good intentions were aired, and if the firms happened to enjoy a lot of public trust in the first place, these good intentions might have found an audience who believes that they will be carried out.

 

But the atmosphere of cynicism which has gradually encroached on almost all areas of public life make such an eventuality unlikely, to say the least.  And this cynicism has arisen due in no small part to the previous doings of the aforementioned Big Tech firms—specifically, their activities in social media.

 

When you compare the health of what you might call the body politic of the United States today with what it was say, fifty years ago, the comparison is breathtaking.  In 1973, 42% of those U. S. residents surveyed said they had either a "great deal" or "quite a lot" of confidence in Congress.  Only 16% said they had "very little" confidence.  In 2023, the number of people with either a great deal or quite a lot of confidence is only 8%, and fully 48% say they have "very little" confidence in Congress.  While this trend has been happening for years, much of it has occurred only since 2018, after the social-media phenomenon overtook legacy media as the main conduit of political information exchange—if one can call it that.

 

Never mind what AI may do in the future.  We are standing in the wreckage of something it has done already:  it has caused great and perhaps permanent damage to the primary means by which a nation has of governing itself.  Not AI alone, to be sure, but AI has played an essential role in the way companies have profited from encouraging the worst in people. 

 

It would be nice if last Friday's White House meeting triggered a revolution in the way Big Tech uses AI and its other boxes of tricks to encourage genuine human flourishing without the horrific side effects in both personal lives and in public institutions that we have seen already.  But getting some CEOs in a private room with the President and issuing a nice-sounding press release afterwards isn't likely to do that.  It's a step in the right direction, but a step so tiny that it's almost not worth talking about. 

 

Historically, needed technical regulations have come about only when vivid, graphic, and (usually) sudden harm has been caused.  The kinds of damage AI can do are rarely that striking, so we may have to wait quite a while before meaningful AI regulations are even considered.  But in my view, it was already high time years ago.

 

Sources:  The White House press release on the July 21 meeting of AI firms with President Biden can be found at https://www.whitehouse.gov/briefing-room/statements-releases/2023/07/21/fact-sheet-biden-harris-administration-secures-voluntary-commitments-from-leading-artificial-intelligence-companies-to-manage-the-risks-posed-by-ai/.  I also referred to a PBS report on the meeting at

https://www.pbs.org/newshour/politics/watch-live-biden-announces-ai-safeguards-after-meeting-with-tech-leaders.  The Gallup poll historical data on confidence in Congress can be found at

https://news.gallup.com/poll/1597/confidence-institutions.aspx.

Monday, July 17, 2023

Air-Condition Texas Prisons!

 

An Open Letter To:

 

The Hon. Greg Abbott, Governor of the State of Texas

The Hon. Dan Patrick, Lieutenant Governor and President of the Senate

The Hon. Members of the Texas Senate

 

A July 9 editorial in the Austin American-Statesman alerted its readers to the fact that of the 128,000 or so prisoners incarcerated in Texas prisons, only about 42,000 sleep in air-conditioned cells.  Only 31% of the state's prisons provide air conditioning for prisoners as of May of this year, which has been one of the hottest ones in recent memory.

 

Just a few weeks ago, facing a $32 billion state surplus, the Texas House voted to spend less than 2% of that sum on the project of air-conditioning most Texas prisons.  At long last, the state which witnessed the first air-conditioned church building (First Presbyterian in Orange, Texas in 1914) and the first custom air-conditioned automobile, was going to extend the benefits of that characteristically Texas technology to its prisoners.

 

But to your shame, Texas Senators and Lt. Gov. Patrick, you declined to act on that measure, and to the extent Gov. Abbott failed to make it a priority during the special sessions, he shares some of the blame. 

 

Here are some objections I can think of to air-conditioning Texas prisons, with the rebuttals to each:

 

* It costs too much to retrofit old structures to be air-conditioned.  It didn't cost too much to retrofit the nineteenth-century pile called the Texas Capitol so that legislators can work in air-conditioned comfort back in 1955.  And 2% of the surplus doesn't strike me as "too much."

 

*  Prisons haven't had air conditioning before now, so why change? By that argument, we should never have installed indoor plumbing or electricity in them either.  If primitive conditions are what we want prisoners to have, why not make them go to the bathroom in open latrines and rely on kerosene lanterns?  Then we can add plagues of cholera and fires to heat prostration and the other hazards prisoners experience already.

 

* Many prisoners work outside in hot conditions anyway, so they're used to it.  Some people got used to living in Nazi or Soviet prison camps, too, but that doesn't mean it was a good thing.  Most farmers and other outdoor workers manage to work in air-conditioned equipment and environments whenever possible.  Withholding air conditioning all of the time because it can't be experienced some of the time is illogical.

 

*  Prison staff have to endure the heat too.  I am willing to wager that the chief official of each prison has an air-conditioned office.  The Statesman cites a 40% turnover rate for lower-level prison employees, which the lack of air conditioning surely contributes to.  If we simply passed a law that the average temperature in the warden's office cannot be more than two degrees less than that in the hottest cell, we would see a revolution in prison budgets overnight, even those operated by contractors, directing funds toward air-conditioning prison cells.

 

Many of you legislators make much of your Christianity.  Do you think God wasn't looking when you neglected to take up the prison air-conditioning bill?  Didn't Jesus say, "Depart from me, you cursed, into the eternal fire prepared for the devil and his angels; for I was hungry and you gave me no food, I was thirsty and you gave me no drink, I was . . . in prison and you did not visit me"?  Air conditioning isn't mentioned by name in that list, but I don't think any of you are literalist enough to miss the implications.

 

I lived in Massachusetts for many years before returning to my native Texas in large part because of the Christian-influenced culture that in many ways reflects my deepest beliefs about life and the universe.  And I still think Texas is the better place to live, at least if you're not in prison.  But if anyone from out of state asks me why Texas, of all places, the leader in the development of air conditioning, has failed to apply this blessing of humanity to some of its most neglected residents—its prisoners—I can only hang my head in shame. 

 

Besides the moral aspect, there are practical ones too.  Sooner or later, confinement to an un-air-conditioned cell in South Texas will lead a lawyer to convince a judge and jury that such is "cruel and unusual punishment," and the Feds will take over the prisons.  Nobody wants that, not even the prisoners, probably.  So before it's too late—and as your average age is 60, too late may be sooner than you think—remedy this sin of omission, and air-condition Texas prisons.

 

Karl D. Stephan, Professor

Ingram School of Engineering

Texas State University

Monday, July 10, 2023

Is the Bloom Off the Self-Driving Rose?

 

Pardon the mixed metaphor—roses don't drive—but I couldn't think of another way to summarize the current prospects for truly autonomous vehicles.  For several years now, we have been promised that self-driving cars are just around the corner.  In particular, Tesla has marketed an expensive option for their electric vehicles called "Full Self-Driving Capability."  But as the NHTSA is asking Tesla for ever more detailed information about how it has deployed and modified its autonomous-driving software since 2014, it's beginning to look like the promised future of leaving all the driving to robot chauffeurs while we nap or play cards in the back seat is nothing more than hype.

 

What is reality right now regarding self-driving cars?  The Society of Automotive Engineers has established six levels of autonomy, ranging from 0 (what a Model T had, requiring you to do everything yourself) up to Level 5.  A Level 5 car could drive you anywhere that is physically accessible by a car without your having to lift a finger. 

 

The most advanced self-driving systems currently on the market are Level 3.  A Level 3 system such as that in the Honda Legend, which was introduced only in Japan in limited quantities, really does drive the car without human intervention, but reserves the right to ask the human to take over if things get too hairy.  "Hairy" in this case can mean something as mild as a rainy day, which sets up weird reflections on streets and can confuse even deep-learning AI systems. 

 

Tesla's much-publicized "Full Self-Driving Mode" is only a Level 2 system, because the occupant is supposed to be prepared to start steering at any time.  People routinely violate this rule, however, which is how several accidents involving supposedly self-driving Teslas have happened.

 

Computer scientist and engineer Anthony Levandowsky ought to know the self-driving score if anybody does, as he got into the business way back in 2003 when he teamed with some fellow U. C. Berkeley engineers to enter a self-driving motorcycle in DARPA's 2004 Grand Challenge.  He went on to work for Google, left Google for Uber amid lawsuits charging theft of intellectual property, was convicted of same, and pardoned by President Trump on Trump's last day in office.

 

If anybody has an insider's perspective on self-driving vehicles, Levandowsky does.  What is he up to now?  He runs a company that converts giant open-pit quarry trucks to be autonomous vehicles.  And in an interview with Bloomberg News, he says that the kind of highly restricted environment in which such trucks operate may be the best that truly autonomous driving can do for the foreseeable future.

 

While there have been numerous technical advances in sensors, computing power, and AI in the two decades or so that Levandowsky has been in the business, self-driving cars are up against a truly astounding opponent:  the average driver.  As the Bloomberg article points out, suppose you see a couple of pigeons on the road ahead.  As an experienced human, you know that pigeons almost always fly away before your car lands on top of them.  And even if you are contending with a particularly dopey or hung-over pigeon whose situational awareness isn't up to snuff, running over a pigeon is not going to ruin your car.  So it's no big deal to see some pigeons in your path if you're driving.

 

But to an autonomous vehicle system that may never have encountered these particular pigeons on this particular stretch of road under these particular lighting and weather circumstances, it's a totally novel experience.  And most prudent programmers will insert a default "brake when in doubt" operation when the system encounters something that might be dangerous.  So what may well happen is that the car suddenly brakes, and the driver following you may not notice in time, leading to a rear-end collision or even a pileup on a busy freeway—all because of a pigeon.

 

Multiply this scenario by the thousands of other ones that come up all the time, and you begin to understand why self-driving taxis are found only in highly restricted areas of certain cities, and why so many of the demonstrations of self-driving cars take place in California, Arizona, and other places where clear skies can be counted on.  Levandowsky takes the position that every so-called self-driving car out there is really just a pilot project, and Tesla even makes this condition explicit, calling their system a "beta version," meaning it's still under development. 

 

So when, if ever, are we going to get to have our little cocktail among friends in the back seat, heedless of the weather or the traffic?  I can picture only two situations in which even some people can get to experience this in what remains of my lifetime.

 

One is if cities establish self-driving-only zones in which only self-driving cars with similar operating systems are allowed.  This type of operation will never be a mass-market phenomenon, but it essentially transfers Levandowsky's self-driving quarry trucks to dense downtown areas, where the environment can be controlled, perhaps with strictly-enforced rules on pedestrians.

 

The other would extend this principle to an entire region or country.  Maybe we should start with Lichtenstein.  Only self-driving cars would be allowed on the roads.  This will probably never fly in the U. S., but it might take place in some dictatorial environment such as China or North Korea—assuming North Koreans ever get rich enough to buy their own cars.  And even then, the rest of the environment is still going to cause lots of problems—pigeons can't read highway laws, and they're going to land on the road anyway. 

 

It's kind of a shame, really.  Part of me was looking forward to handing the whole responsibility of driving over to some system that had proven itself at least as trustworthy as your average taxi driver.  (Not the taxi driver I wound up with in Amherst, Massachusetts once, who had been out "fishing" with a six-pack and hit a curb, blew out a tire, stopped at a gas station, got the tire fixed, and still got us to our flight to China somehow.)  But now it looks like that's going to go the way of other Jetson-inspired dreams, like personal flying saucers.  Of course, if they do get batteries good enough to operate self-flying one-person drones, we could rewrite the FAA rules a lot more easily than redoing all the traffic laws.  But you'd still have to deal with those pesky pigeons.

 

Sources:  The Bloomberg News article "Even After $100 Billion, Self-Driving Cars Are Going Nowhere" appeared on Oct. 5, 2022 at https://www.bloomberg.com/news/features/2022-10-06/even-after-100-billion-self-driving-cars-are-going-nowhere.  I also referred to the J. D. Power website for the SAE's six levels of autonomous-vehicle operation at https://www.jdpower.com/cars/shopping-guides/levels-of-autonomous-driving-explained

and https://www.cbsnews.com/sanfrancisco/news/tesla-autopilot-driver-assist-system-nhtsa-seeks-details-recent-changes/, as well as the Wikipedia article on Anthony Levandowsky.

Monday, July 03, 2023

Is Science Basic Enough to Support Engineering?

 

The word "basic" is often misused, but its most straightforward meaning is to denote something that lies at the bottom of other things, causally speaking.  Without basic math, you can't do calculus.  Without basic English, you can't make head or tail of Milton's Paradise Lost (and maybe not even with it).  Modern science-based (there's that word again) engineering depends on science in manifold ways:  not just to provide the physical understanding of the materials and processes that engineers work with, but sometimes to give inspiration to novel ideas as well.

 

When the word "basic" is applied to science, we get one of a pair of phrases that are often used to emphasize a contrast, namely basic versus applied science.  J. Britt Holbook is a philosopher with an appointment in the Department of Humanities at the New Jersey Institute of Technology.  In a new collection of papers on science, technology, and society, he muses on the contrast between basic and applied science, and asks what role the state should play in supporting either one, both, or perhaps neither.

 

In the beginning, there was no state support of science, or as it was called then, natural philosophy.  Philosophers had to make a living in some practical arena (Socrates was reportedly a stone-carver), and do philosophy in their spare time.  A few managed to become teachers of philosophy, but still, one could not expect to earn a government salary by just thinking about the stars.  Your lectures had to attract enough students to pay the rent.

 

It was in nineteenth-century Germany that the state began to realize that supporting science scholarship might actually benefit not only the scientists, but society at large.  Holbrook cites the great Alexander von Humboldt (1769-1859) as one of the main early proponents of such support.  What Humboldt wanted the state to support was not what we would call science, narrowly defined.  Rather, it was the entire range of disciplines worthy of graduate study by both professors and their students, who would not only acquire what the professors had to pass on to them, but would discover new knowledge as well.  This included what we would call social and even moral sciences as well as the natural sciences.  And by 1870 or so, several German universities managed to put Humboldt's vision into action, and put Germany in the lead of what the Germans called Wissenschaft.

 

When we in the U. S. tried to import this pattern, we put our own twist on it, a notably practical one.  The birth of "big science," funded at huge levels by focused government programs, took place during World War II, largely overseen by Vannevar Bush, who directed the Office of Scientific Research and Development (OSRD) during the war.  While the OSRD's most historic product was the nuclear bomb, it also established precedents of funding levels that Bush attempted to maintain after the war.  Although the ensuing National Science Foundation, founded in 1950, was not exactly what Bush had in mind, it embodied elements of both Bush's ideas and Humboldt's notion that supporting scientists to investigate knowledge for knowledge's sake alone would benefit society both directly, in terms of new applications of science via engineering, and indirectly, by raising the tone of the educated populace in general.

 

In the following decades, Holbrook notes that the NSF has steered ever closer to the applied, technological, or engineering side of things.  In 2022, the agency formed its first new directorate in thirty years.  The Technology, Innovation, and Partnerships Directorate will further emphasize research focused on practical applications and technology transfer to industry.  This is only the most recent move in a long-established trend which some observers see as betraying the NSF's original purpose of funding research without regard to its possible future applications.

 

Holbrook sees a real danger in allowing public needs, as perceived by the government, to dictate the direction of science research.  Toward the end of his article, he says " . . . we run the risk that whatever intellectual activity we argue the state should support will be reduced to a mere means to satisfy the desires of the state.  There is no surer way to guarantee an absolutist state than to give in to the idea that all 'science' should support the needs of the state as defined by the state."

 

From my own worm's-eye view of university-based research, I have seen the NSF move from a position in which the technical science content of a proposal was almost all that mattered, to a position in which one's political credentials and even sex and race have a significant effect on the chances of getting funded.  While Holbrook did not directly address these issues, they are a part of the shift he is criticizing.

 

Humboldt's original vision was broader than just funding a technological arms race between countries (i. e. the U. S. versus China) or even providing the direct benefits that new scientific knowledge conveys by means of development in technology and engineering.  He had what may now sound like an old-fashioned and even naive trust that more knowledge will lead to an enlightened public and an atmosphere more favorable to human flourishing.

 

Although Holbrook didn't point this out, Humboldt's idea approaches Plato's notion that evil is simply the result of ignorance.  That is why Plato's ideal government would be put in charge of a philosopher-king, because the philosophers know more than anyone else and therefore have the best chance at being good.

 

I think the public's recent experience with scientific expertise during COVID-19 showed that scientists, far from being the most moral people of all, have flaws just like the rest of us do.  I agree with Holbrook that putting blinders on scientists and making them study only things of direct interest to the state is liable to cripple both science and the engineering on which it depends.  I believe we have strayed much too far from the seldom-realized ideal of paying scientists to study whatever interests them, whether we can imagine a payday resulting from their efforts or not.  But on the other hand, we shouldn't expect their discoveries to make us better people.  That's not their job.  That's up to us, to do with whatever natural and supernatural help we can find.

 

Sources:  J. Britt Holbrook's essay "An Effective History of the Basic-Applied Distinction in 'Science' Policy" appears on pp. 483-497 of G. Miller, H. M. Jerónimo, and Qin Zhu, Thinking Through Science and Technology (Lanham, MD:  Rowman & Littlefield, 2023).

Monday, June 26, 2023

An AI Church Service: Whom Will You Serve?

 

Every couple of years, the Evangelical Church in Germany, one of the world's largest national Protestant church federations, holds a convention in a different location.  This year, it was in Fuerth, outside of Nuremberg in Bavaria, and the event that attracted more media attention than anything else was a church service produced almost entirely by the AI chatbot ChatGPT.  The service was the brainchild of 29-year-old Jonas Simmerlein, who is identified as a "theologian and philosopher," not a computer expert.  This may explain some of the complaints heard from the congregation about the rather wooden delivery style of the four avatars—two women and two men—who delivered the prayers and the sermon and selected the music. 

 

Some of the younger attendees among the 300 or so present said on the other hand that the service was surprisingly good, even though "a bit bumpy at times."  The service was not interactive in any real sense, so when the congregation found some of the avatar's words inadvertently humorous, as when it said with a deadpan expression, "to keep our faith, we must pray and go to church regularly," it kept right on going despite the laughter. 

 

To be fair, some of the problems people noted could be fixed by more resources.  Hollywood studios have access to animation techniques that reliably pull emotions from millions, so the expressionless delivery style is not a fundamental issue, but simply reflects Simmerlein's inexperience with the process.  But suppose the avatars had been much more emotionally engaging, and the whole service felt like real people were up there talking to you?  Would that have been better?  It all depends on what you think a church service is for.

 

If you view attending church as just one of a number of ways to spend your time searching for happiness, and if you judge a service by how happy it makes you, then the emotional buzz it produces is all-important.  By this criterion, going to your favorite band's rock concert is probably going to do you more good than any number of church services. 

 

But if you view church attendance as intentionally placing yourself in the presence of a supernatural God, the picture changes.  If a human being is not running the service, some branches of Christianity regard it as worthless, because only a human priest can do certain things that have to be done for a mass to be a mass, for example.  In the Catholic mass, a duly ordained priest must consecrate the bread and wine to change it into Christ's body and blood.  No AI system can do that.  And when the members of the congregation take the elements, as they are called, the priest is acting as Christ for them, which is why he has to be male. 

 

This odd little AI experiment highlights an issue that we are going to face more and more as time goes on, and as so-called generative AI develops more abilities that mimic human ones.  There are all kinds of work for human beings in the world, and none of them, perhaps, is entirely without some negative aspect.  Even Michelangelo got a tired back while working on the Sistine Chapel.  But if he had been able to call upon some wizard to do the work for him, would he have done so?  I seriously doubt it.  An artist uses tools, but the quality of the art depends not primarily on the tools but on how well the artist uses them and transmits the vision in his or her mind to canvas, paper, or sound.  Art, or a church service, produced almost entirely by a machine is going to be missing the essential ingredient that makes it a human achievement. 

 

But not everybody will miss that ingredient.  In the dystopian novel Nineteen Eighty-Four, if my memory serves, one of the characters made a living by operating a "pornography machine" which cranked out dirty books for consumption by the masses.  The work was entirely physical, like running a printing press, and involved no mental effort at all.  Despite the completely mechanical source of the books, they proved to be very popular, as pornography usually does.

 

The vast traffic in pornography on the Internet, except for the fact that it is digital instead of mechanical, fulfills George Orwell's prophecy of 1949, and shows what can happen when modern technology is twisted to deliver what some people consider happiness.  The connection between Internet pornography and church services is not obvious, but if you place them both on a spectrum of things designed solely to make the user happy, the question is simply how effective each one is in achieving the goal of happiness.  Some people may like church services run by AI-generated avatars, and others may like AI-produced Internet porn—it just becomes a matter of taste.

 

We have Jonas Simmerlein to thank for creating an extreme example of the wrong kind of church service.  The right kind is truly a service, but not primarily to the individuals involved.  If God truly is the Creator of the universe and everything good in it, we owe Him everything, in principle, and an hour or so of devotion and worship every week seems like only a pittance.  If there is some way (right now I can't imagine how, but maybe there is some way) to incorporate AI into a worship service that is truly and obviously devoted to God, maybe that can happen.  But turning the whole thing over to a machine is no different than showing a video of a recorded service—in fact it's a good deal worse, because a video of a live service at least shows real people doing and saying things that have a chance of pleasing God.

 

The Fuerth ChatGPT church service is a first, all right, but it may turn out to be one of those things that we have to try in order to find out that this is not really the direction we want to go.  Of course, there will always be an audience for such things.  In fact, one of the fears of AI developers is that AI will come up with false religions to deceive the masses into doing things like Jim Jones did with his Peoples Temple cult, from which we get the grim phrase "drink the Kool-Aid."  Let's hope that most of us, anyway, have more sense than that, and that AI developers put safeguards in place to make such things impossible.

 

Sources:  The Associated Press carried the article "Can a chatbot preach a good sermon?" on June 10, 2023 at https://apnews.com/article/germany-church-protestants-chatgpt-ai-sermon-651f21c24cfb47e3122e987a7263d348. 

Monday, June 19, 2023

The Davenport Apartment Collapse

 

Around 5 PM on Sunday evening, May 28, if you had been standing near the corner of West 4th and Main Streets in downtown Davenport, Iowa, you would have seen The Davenport apartment building yield up about a third of its facade to gravity.  With a loud rumble and a cloud of dust, more than a dozen apartments in the building collapsed, taking three lives and trapping Quanishia Berry in her fourth-story apartment for over twenty-four hours before she was rescued.  Like many engineering disasters, this one was the outcome of a series of actions, or inactions, that could have prevented it.  But now that it's too late for the victims, we can at least learn something from the sequence of events that preceded the tragedy.

 

The Davenport was built in 1907 and at six stories, was probably one of the most prominent structures in the city at that time.  The residents living in it just before the collapse complained of odd creaking noises, cracks in the floor underneath carpets, and window frames pulling away from walls. 

 

The Davenport's owner, Andrew Wold, paid a structural engineer in February to conduct an inspection of the building.  The engineer's report stated that although bricks in the facade were cracked, the flaws were not an "imminent threat to the building or its residents."  In later reports, the last one dated four days before the collapse, the same engineer said that now, several patches of bricks "appear ready to fall imminently, which may create a safety hazard to cars or passersby."  Evidently, a portion of a load-bearing wall behind the brick facade had collapsed, transferring loads onto the facade itself, which was not intended to carry such loads.  This would explain the bulging and other issues noted by the residents.

 

Some residents became so alarmed by the deterioration that they broke their lease and moved out before the collapse.  But others took the word of the landlord that there was no immediate danger. 

 

It is always a judgment call to determine when a building's condition is so bad that it must be evacuated.  Wold had obtained estimates for shoring and supporting parts of the building, but at $50,000 he regarded them as too costly.  The owner of the masonry repair service who provided the quote said that he wouldn't allow his workers on the site without such repairs.  So here was a technically competent person saying he wouldn't even let his employees near a place where residents were living twenty-four hours a day. 

 

How could things have gone differently?  There is evidence that Trishna Pradhan, the city's chief building official, thinks she might have been able to prevent the tragedy by issuing an evacuation order, something she was presumably authorized to do.  The evidence is that a few days after the collapse, she resigned her position.  Another official gave the reason for her resignation as having "wrongly categorized an inspection of the building."  While orders to evacuate a structure immediately are serious things and involve potential legal liability, failing to issue such an order can have worse consequences, as it did in this case. 

 

A building is an extremely complicated structure, and predicting down to the minute when something as large as The Davenport will collapse is essentially impossible.  But when large cracks appear in significant places and the facade begins to bulge visibly, even the average non-technical person would agree that something bad may be about to happen. 

 

One article on the collapse interviewed engineers who cite modern technologies such as vibration sensors that can detect abnormal shifts and movements that occur before serious structural failures.  These can be installed in suspect structures and monitored, but then the question becomes, "How much movement is too much?"  These methods are still experimental, and the expense of installing and monitoring them is one more reason to simply leave things as they are.

 

And that was the main sin of omission committed here.  No one wanted to be the person who stuck his or her neck out and said, "Okay, this place is about to fall down, and we need to get the residents out before it does."  Only a few occupants made this decision on their own, which is not to blame the ones who stayed.  After all, they were told that there was no imminent danger, and up to 5 PM on May 28, that statement was correct, depending on how you define "imminent." 

 

That word did come up in the engineering report four days before the collapse, and should have motivated someone—the landlord, the city—to act.  But yet another decision point was passed, and events moved on to their tragic conclusion.

 

Whoever takes over from Trishna Pradhan as chief building official is going to be a lot more careful about classifying the results of building inspections in the future, one would hope.  And as the technology for sensing vibration in buildings gets cheaper and more available, it might be a good thing if buildings with structural problems were required to have such sensors installed.  The most valuable thing about that would be that we could get lots of data from normal wear and tear, and could get a better idea about what happens with regard to vibrations as a building gets closer to falling down.  Electronic infrastructure sensors like these are available for new buildings, but it would be a challenge to install them in old structures, although not impossible.

 

I can't close this blog without remarking on the likely class of residents who occupied The Davenport in its final days, and whether their complaints might have received more attention if they had been, frankly, richer.  But rich people don't live in 116-year-old run-down buildings in the middle of deteriorated city centers.  Just this morning I was reading in the Old Testament book of Proverbs, and came across one that seems to sum up what happened here:  "The poor man utters supplications, but the rich man answers roughly."  Next time, the rich man needs to listen more carefully, and take more thought for the safety of his fellow human beings.

 

Sources:  I referred to the AP article "Many warnings before collapse" at https://apnews.com/article/iowa-davenport-building-collapse-621844c0d6c206ced4639ddcf443eb80 and the Wikipedia article "Davenport apartment collapse."  The Proverbs verse is Prov. 18:23 from the New American Standard Bible.

Monday, June 12, 2023

Prudence, Justice, Democracy, and Technocracy

 

In the June 12 issue of National Review, writer Noah Rothman takes aim at the web of intrusive regulations that, as the Lilliputians tried to bind Gulliver with hundreds of strings, try to micromanage our personal lives with regard to the kinds of technologies we use around the house.  In recent years, the following well-established items have become targets for government regulation above and beyond the usual safety concerns:  incandescent light bulbs, gas stoves, gas water heaters, air conditioners, lawn mowers, and even paper bags and plastic straws. 

 

Any attempt to elucidate a coherent motivation behind these regulations, which emerge from all levels of government (city, state, and federal), tends to end in confusion.  Various reasons are given ranging from suspicious increases in childhood asthma for gas stoves, to air and noise pollution from gasoline-powered lawn mowers, to alleged environmental harms from plastic straws.  None of the regulators claim that any of these technologies are bad because they are directly responsible for the deaths or injuries of those who use them (although people have died in fires caused by gas stoves and cut off their fingers with lawnmowers).  The harm is always more remote:  vague and poorly-implemented studies showing correlation between asthma and the use of gas ranges, increased levels of environmental pollution that are so small as to be negligible, or the ever-present excuse of climate change. 

 

As Rothman points out, neglected in all these calculations by the regulators are the inconveniences, or worse, placed on the consumer by the byproducts of regulation.  LED lighting costs more than old-fashioned incandescent lighting.  This is a clear burden on poor people.  Some kinds of cooking are hard or impossible to do with electric or induction ranges, which also cost more to operate in many locales than gas ranges do.  In the name of various vague political causes, then, the regulators ask the much larger number of ordinary citizens to endure a set of individually slight inconveniences, which however add up to a considerable burden if totaled over the entire U. S. 

 

If we look at this situation from an ethical point of view, two cardinal virtues come to mind:  justice and prudence.  Justice sees that each person gets what is coming to him or her, in either a nice way or a not-so-nice way.  Prudence is a sort of all-around virtue of balancing various factors so that justice is served.  In a democracy, both of these virtues should be possessed by the body politic so that whatever is democratically enacted embodies them. 

 

Before the era of micro-managing regulations, all of the technologies mentioned above were adopted not because they were politically correct, but because enterprising inventors and manufacturers developed them in response to needs and wants of the public.  And the public went for them without any help from government.  In the name of prudence, certain basic safety regulations and voluntary schemes such as the Underwriters Laboratory label were put in place to ensure that direct harm from these technologies was kept to an acceptably low level, but that was it as far as regulation is concerned.

 

However, there is another philosophy of government that is more or less opposed to democracy, and that is technocracy:  rule by experts.  In this view, the great unwashed masses are too ignorant to choose domestic technologies wisely, and need to be guided by their elitely-educated masters into the path of enlightenment.  Exactly which path that is depends on the expert, so in practice the regulation that is enacted depends on just who grabs the levers of power in the increasingly complex web of bureaucracies that we live under today.  This probably explains the geographically diverse nature of these regulations:  what's fine in Nevada may be banned in California and vice versa, depending on which technocrat is running the show. 

 

Both prudence and justice are violated when an ideologue imposes his idiosyncratic whims on a large group of citizens who have been minding their own business.  Whatever allegedly meritorious political goal is served by the regulation is usually far outweighed by the sum of small (or not so small) inconveniences suffered by those who are regulated.  Voluntary suffering and privation, as many citizens did to sacrifice consumer products during World War II, is one thing.  But to have it imposed on you in the absence of a national emergency is both unwise (imprudent) and unfair (unjust).

 

Rothman ends his article in the hope that voters will turn to the ballot box to alleviate some of the ills imposed by bureaucratic micro-regulation of technologies which have proved both popular and safe, by and large.  Unfortunately, the same diffuseness that marks the regulations (small harms caused to a lot of people) means that such matters are unlikely to become dominant issues in political campaigns.  I doubt that even the most progressive candidates have made banning gas stoves a major plank in their platforms. 

 

The most we can hope for in this regard is that one party or the other will take up the cause of pushing back against rule by technocrats in general, and a vote for that party will work against the creeping regulations that threaten to deprive us of things that work just fine, thank you.  While this might work on the national level, the entrenched nature of local politics, especially in large cities and certain states, means that there are only limited hopes in this direction. 

 

Sometimes other types of public-relations actions can have effects all out of proportion to their actual size.  I think of Mahatma Gandhi's famous Salt March of 1930, which publicized the Indian populace's opposition to the British monopoly on salt production.  Salt was something everybody used and had to pay for, and the British unfairly prevented Indian citizens from making it themselves, no doubt using reasons of sanitation and quality to justify their injustice.  After the arrest of some 60,000 protesters, Gandhi was able to negotiate with the British leaders for the right to make salt by ordinary citizens, and the march was a success.

 

I don't know how effective a lawnmower march on a state capital would be, but I'm sure it would garner some publicity.  When ordinary democratic means either fail or are unavailable, more unorthodox ways should be tried.  But the best thing is to re-establish both a working democratic system, and a populace who will not stand for any nonsense.

 

Sources:  The article by Noah Rothman "The War on Things That Work" appeared in the June 12, 2023 issue of National Review, pp. 21-24.  I also referred to an article on the Gandhi-Irwin Pact at https://www.britannica.com/event/Gandhi-Irwin-Pact.

 

Monday, June 05, 2023

Train Crash in India: An Avoidable Tragedy

 

On Friday evening, June 2, the Coromandel Express carrying about 1250 passengers was traveling through a stretch of track just southwest of Balasore, about 130 miles southwest of Kolkata, India.  A freight train full of heavy iron ore was parked on a siding that paralleled the main line.  Somehow, the express train was signaled to travel past the siding, but the actual switch over which the train passed was set to shunt it into the siding, forcing it to head straight for the stationary freight train.

 

The express train derailed, throwing about a dozen passenger cars over both the siding and the mainline tracks like a spilled box of toothpicks.  To make things worse, another express train, the Yesvantpur-Howrah Express, arrived before emergency signals could reach it, and crashed into some of the derailed Coromandel cars, leading to further derailings and deaths.

 

As of today (Sunday June 4), the total fatality count stands at 275, with about 800 injuries. A preliminary investigation has pinned the cause to faulty signaling and switching.  Obviously, if the signals had matched the position of the switches, the Coromandel Express would have been ordered to stop.  But somehow, perhaps in a last-minute error, someone or something changed the switch from the main line to the siding, leading to the disaster.

 

Though the deaths of 275 people are horrific, this is not the worst train disaster in India's history.  In August of 1995, two trains collided near New Delhi and 358 people died.  This leads to the question of whether an Indian passenger is taking his life in his hands anytime he travels by rail.  Surprisingly, the answer is "not necessarily."

 

There are several ways to measure the relative hazards of travel in different modes and locations.  From an individual's point of view, the number that is most meaningful is probably the passenger-miles traveled for each fatality.  The farther you can go without dying, the better.  Surprisingly, by this measure it is about twice as dangerous to travel by rail in the U. S. as it is to travel by rail in India.  The fatality rate in the U. S. is one for every 3.4 billion passenger-miles, and in India it is one for every 6.6 billion.  Yet when even one passenger gets killed in a train wreck here, it's big news.

 

The difference can be partially explained by the fact that intercity passenger travel by rail in the U. S. is something only a few people do, whereas most of the middle and poorer classes in India travel by rail.  The passenger-miles traveled in India every year is about 35 times that in the U. S.  So even though the absolute number of train-related passenger fatalities in the U. S. is much smaller, we take the train even less than they do in India, and so our ratio of deaths to passenger-miles is worse.

 

Another difference is the density of people per train.  I haven't taken a ride on Amtrak lately, or a regional line such as the Long Island Railroad, but it would surprise me if you could find a regularly scheduled train anywhere in the U. S. that carries 1200 people at a time.  The more people on a single train, the more people can be hurt or killed if that train gets into trouble.

 

For a long time, railroads have been trying to devise foolproof systems of track switching.  The so-called interlocking switches were originally devised in the 1850s.  The idea was to guarantee that no combination of switches could be set that would lead to a collision.  At first the systems were entirely mechanical, but by the 1890s electric-motor-activated switches were introduced, leading to electromechanical and later entirely electronic systems.

 

But even the best-designed systems can fail.  Unless the trains themselves are controlled or monitored remotely (an element of what in the U. S. is called positive train control), a determined or confused individual may change a switch at the wrong moment, for example, just as an express train is heading toward the switch that leads to a siding.  While this will automatically change the signals ahead of the switch, if the train has passed all the signals already, it's way too late to do anything about it. 

 

Something like this may have happened in the Coromandel Express tragedy.  Once the relevant parties are interviewed and records examined, a clearer picture of the cause should emerge.  But at this point, that is small solace for the thousands of people who have lost relatives or loved ones in an accident that did not have to happen. 

 

The tolerance for violent technology-related death varies from culture to culture.  Here in the U. S., we have more or less passively accepted a rise in auto-related deaths over the last few years, despite numerous technological improvements such as safety belts and air bags that make accidents survivable which formerly would have been surefire killers.  In 2020 and 2021, the fatality rate increased about 17%, partly no doubt due to COVID-19.  But in 2022 the rate stayed about the same, which seems to indicate we changed something for the worse during the pandemic and haven't changed back.  Whether the reason is increased drug and alcohol use, more reckless driving by depressed individuals, or some combination of factors, an extra 4,000 or so deaths per year has not really seemed to concern anybody, except when it's you or someone you love.

 

At any rate, I hope that this latest train accident will lead to further improvements in the already comparatively good safety record of Indian railways.  Considering the huge numbers of people they transport every day, they are doing a better job by some measures than we are doing in the U. S.  But there is always room for improvement, and as technology and organizational integrity improve, maybe this can be the last major train wreck in India for a long time.

 

Sources:  I referred to an NBC News article on the wreck that appeared at https://www.nbcnews.com/news/world/india-train-crash-cause-signal-error-blamed-derailment-rcna87609.  I also referred to statistics on comparative train-passenger fatality rates worldwide at https://pedestrianobservations.com/2011/06/02/comparative-rail-safety/ and the Wikipedia article "Interlocking" and https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in_U.S._by_year.