Nothing is more persuasive than the illusion of reasoning

The answer must be in here somewhere

I have been working with ChatGPT to learn about how Large Language Models (LLMs, which include Claude and Gemini) work, and then on how to use them effectively. I was worried that ChatGPT wouldn’t be particularly good at this, particularly since human beings are so poor at explaining their own motivations to themselves.

We have trouble explaining ourselves to ourselves.

Rather, humans are good at explaining their motivations; it’s just that they are not that accurate. It’s not because we just lie to ourselves, though sometimes we do—often these assessments are pretty good. It’s just that we don’t have privileged access to our own underlying motivations.

It’s kind of the same way we can do a pretty good job of explaining the specific motivation of someone we know well when they make a decision, or respond in a certain way to a situation. We are always in our own company, and have watched ourselves make all sorts of decisions and acted in all sorts of ways, and know how we felt about those decisions and actions afterward. But we don’t perceive our own motivations directly.

Check your own revealed preferences to better understand yourself

It took me some years to realize that I didn’t know specifically what kind of woman I was attracted to—I just had to look at who I was and had been attracted to, and examine the regularities and the differences. This is true of a lot of things, but who you are attracted to has high visibility and has significant consequences on your happiness and life path, and so is an important choice. Sometimes you might recognize that what you are attracted to, in a mate, a choice of job, or who to hang out with, has bad long-term effects, and then have to work with a motivation system that you feel is not serving your long-term needs well.

But ChatGPT has no inner life. All it has are weights and tokens derived from a massive amount of linguistic training data. And that training data has a lot of examples of how to train, as well as a lot of content on LLMs themselves.

Why LLMs can seem to reason

So far it’s been doing a good job of teaching me, though I am early on in the roughly six-month initial training program it has suggested to me.

One key point is that LLMs don’t reason. They have incorporated vast amounts of written reasoning, and use those sequences to generate what looks very much like reasoning, but definitely isn’t. It explains where reasoning-like behavior comes from, even if there is no actual reasoning:

A token is plausible because it fits learned patterns, not because it is valid or true. Reasoning‑like behavior is imitation of structural constraints, not inference.

As it happens, I have been working my way through a book on reasoning and inference: The Enigma of Reason (2017), by Hugo Mercier and Dan Sperber. There are a lot of opinions in cognitive psychology and philosophy about reason, and they try to provide a context for their own positions by referencing both those researchers and thinkers they agree with and those they don’t, and why.

I find that comforting. But should I? Could an LLM produce a book that is just as persuasive, but misleading in some way?

Mercier and Sperber have already hit on the point in their chapter on “Metarepresentations”, and why we find explanations persuasive.

We have clear intuitions about the “goodness” of various explanations. As the psychologist Frank Keil and his colleagues have shown, these intuitions may not be very reliable when they concern our own ability to provide explanations…We are, however, better at evaluating the explanations given by others. Even children are typically quite adept at recognizing the expertise of others in specific domains of explanation, and at taking advantage of it.

Mercier and Sperber see a variety of lower-level modules in our thinking and responses, a model of the brain and mind I have long found persuasive. But because we don’t have any privileged access to the internals of those modules, we can be spoofed without realizing it. How does this particular module respond to the reasoning-like explanations of an LLM?

I feel that I have a sense of when someone has done the work and seems to understand what they are explaining to me. In fact, explaining something to someone else is a good way to see if we understand it ourselves, another example of how we sometimes need to move outside ourselves and then go back in to get a better idea of what we actually know, think, and feel.

Fake images, fake videos…fake explanations?

Most of us have at some point donned some off-the-rack explanatory reasoning to hide our own cognitive nakedness. We expected someone else to find that reasoning persuasive, when it was not that reasoning that gave us the position we hold.

It’s clear that it doesn’t take LLMs to flood that explanation-validating module into uselessness. Our general online cognitive environment has already done that. But LLMs may supercharge it.

The best way to defend ourselves might be to use our feeble reasoning powers to provide that sensitive module with some input hygiene, some rule like “don’t put anything in your brain that you wouldn't put on paper”. But reasoning is feeble and easily worn out, as well as frequently misapplied. We do need to learn how to defend ourselves.

I read Mercier and Sperber last night, and worked with ChatGPT this morning. Clearly, the topic is on my mind. Still, the fact that ChatGPT explained to me the possible implications of what Mercier and Sperber had been thinking indicates that I may be right to focus on these concepts.

What revealed preferences have you found in your own life?

What other ways have you found to understand your own motivations?

Fare evasion and moral preening

A couple of days ago, I was waiting for my wife at a T stop near our house. During the ten or so minutes I was there, I saw some dozen people enter through the gates. At least seven of them did not pay (I only started counting after a while, so the number is not precise). It was casual, expected behavior. From the appearance of most of these people, it wasn’t a matter of finding the fare a financial burden. It was a matter of being an entitled jerk.

Some chumps paying their fares

That entrance is a bit isolated from the main station, and I’ve never seen it staffed, despite there being a booth for someone. The MBTA supposedly started cracking down on fare evasion back in September, first warning people, then fining them. All I can say is, I think you could fund a few more fare enforcers on that station alone.

Now, fare evasion is as old as fares. The New York Transit Museum has an extensive display of turnstiles from throughout its history, and all got circumvented, jammed with slugs, or broken. This isn’t some kind of recent moral breakdown.

Still, it is always annoying if you are someone who has a sense of the duties we have to our fellow citizens.

An illustrative anecdote about fare evasion

For example, in Fare Share, a post on Addison del Mastro's Substack The Deleted Scenes, del Mastro describes an encounter he had at a housing conference. They got to discussing fare evasion, and one of his dinner companions

…said that when he rides the subway, he hops the turnstile whenever possible. He described it as a sort of gesture of solidarity with the poor. Nobody should have to pay, and since I have a choice, I’ll choose to throw my lot in with the poor turnstile-jumpers. (I don’t really think I’m exaggerating, but I wish I was.)

He then went on to say that for one of his previous jobs he had a monthly pre-loaded MTA card—in other words, his employer paid for his transit commute—and he still jumped the turnstile, “as a matter of principle.”

What principle is that, exactly? I asked him.

Del Mastro was just as irritated by this guy as I would have been. This isn't a principle, this is a pose. This isn't an attempt, even an inadequate attempt, to help anyone. It's a way of feeling like you're the kind of person who helps the poor, without actually needing to do anything.

How evading responsibility can provide a feeling of virtue

I think it's the "not needing to do anything" that is essential to the pose a lot of people strike. Is there a housing shortage? Compel developers to put unprofitable units in their multifamily buildings, so that we can provide "affordable" housing without paying anything. Are rents too high? Compel landlords to provide housing at a lower price. Public transit doesn't work well and is unpopular? Well, at least make it free, without any need to do serious work on improving service.

It also gives you a clear, definable, tweetable thing to point to. Reduced dwell times? A more comforting experience for a nervous woman or child traveling alone? Better on-time performance for someone whose pay gets docked if they're a couple of minutes late, so they don't have to take the bus that's half an hour early just to make sure? Too fuzzy, too dull, too down in the weeds. Those people are kind of boring, anyway. Not cool, like people who are dishonest or violent.

As you can tell, I can probably be defined as a “pro-public-order leftist” who feels that everyone benefits from public order, particularly those that progressives claim they are most devoted to. So, kind of a scold.

Besides paying your fare, return your shopping cart

But those people are always and everywhere with us, as Why Don’t People Return Their Shopping Carts? A (Somewhat) Scientific Investigation demonstrates. So those people evading the fare on the T were unlikely to have been making some kind of point. They were just self-centered jerks who didn’t think the rules applied to them.

But that doesn’t mean we have to make it easy for them to do it. One thing we should do is to push back against weird and self-regarding attitudes like that of the young dipshit del Mastro encountered. Show your fare-evading buddy how little you respect their behavior.

Do you evade fares just to prove some kind of point?

If so, what is it?

Progress and loss - a desk lamp story

Like an old man, inflexible, with back problems

I recently had to retire a V-Tech halogen desk lamp I’ve had for at least thirty years. The attachment point had pulled out of the base enough that I grew reluctant to adjust the arm, fearing the whole thing would come apart.

The joys of a finicky piece of technology

I like a bright light on my desktop, and a 100W halogen gave a nice, sharp, white light. I looked for a replacement, but new versions of the light, while still looking the same, are just not as bright. I presume that, aside from the energy inefficiency of the bulbs, safety regulations also played a role: these things get incredibly hot. That’s why it had an adjustment arm. The cover of the light grew too hot to touch, and I always risked burning myself on it.

I did enjoy the somewhat finicky process of replacing the bulb, taking off the glass diffuser by unscrewing three screws, pulling the old bulb out of the spring-loaded ends, and then carefully, making sure I didn’t get any skin oils on the quartz bulb (which raised the risk that the bulb would explode), putting in the new one and tightening the screws again. One of those had given out, and the other two were getting loose.

I don’t know what happened to mine

My taste for a bright light is long-standing enough that when I was in junior high school, my parents bought me a Tensor 1500, which used a bulb from a car headlight. I used that ugly little thing for years.

LEDs are beginning to actually fulfill their overhyped promises

But let’s face it, technology has moved on. Every year, LED lighting has gotten better. It was definitely overhyped when it was first widely introduced. Bulbs didn’t have the claimed lifespan, they tended to give a harsh light, they couldn’t be dimmed, and they often flickered. Over time, with improvements in the bulbs themselves, but even more in the driver electronics all LED bulbs require, lifespans are actually reaching numbers close to advertised values, colors have gotten much better, and decent dimmable bulbs that don’t buzz or flicker are increasingly available.

Energy efficiency gains are probably going to be incremental from now on, but improvements in driver electronics should make them significantly more reliable, and continue to improve color quality.

It doesn’t really make sense to buy cheap LED bulbs. There are pretty big quality, efficiency, and longevity differences, so you get annoyance without really saving much money, if any. It’s the driver electronics that tend to fail, not the LED light source itself.

And look carefully at the color temperature when you buy. It’s easy to buy one that’s whiter than the incandescent bulbs most of us grew up with. An LED with a color temperature of 2700K has the color most like an incandescent. My halogen was more like 3000K.

Imagine a bar of iron being heated up, from dull red through orange, yellow, and up into searing white, and you’ll understand what “color temperature” refers to. Hotter is whiter. Kelvin (“K”) is a temperature scale that starts at absolute zero, using Celsius degrees.

For brightness, to get what you want you have to look for lumens (a 100W incandescent bulb emits about 1600 lumens), rather than “Watts equivalent” since brands can vary quite a bit in how they measure that.

While there are certainly people who oppose LEDs on a kind of “things used to be the way they should be” principle, and will never be happy, while many others buy poor-quality bulbs, bulbs with the wrong color temperature, or non-dimmable bulbs for use with a dimmer, and then are dissatisfied. It does take a bit more attention than just buying a 60W incandescent bulb.

Good, cheap desk lamps

Two lights is probably overkill, but I really like the glow

So I replaced my beloved old halogen with a generic $19 swing arm lamp, very similar to the one I already had on the other side of the desk. Did I mention I like my light bright? It comes with a nice stable base, but I prefer to clamp it to my desk top.

Now that’s a beautiful lamp

This generic swing-arm lamp is a knock-off of Modernist designs of the 50s, like the Anglepoise. Those are still expensive, but not, I think, much better looking. The clamp desk lamp I truly desire is a French industrial-style lamp, a Jieldé LOFT, which comes in a wide range of colors and textures. It’s a bit more than $19, however, while using exactly the same bulb, so that’s just a fun thing to think about. And I’d still want two.

The one I just bought is a solid, well-working lamp and gives me the light I need. Getting one for this price is a benefit of global trade, and I’m always startled by how cheap they are.

I think part of the explanation is that they don’t need to be designed to handle a lot of wattage (a 1600-lumen, 100W-equivalent bulb uses something like 12W, less than half of what an incandescent refrigerator light would use). This is another reason you should spend more for a better LED—you’re already saving on the lamp itself, which no longer has to handle as much heat or current.

A really bright LED bulb is cool enough to touch. I can move stuff around and get to the USB sockets and DVD drive on my desktop computer without risking a burn. I can adjust it easily.

The shade is a bit too short for the bulb. I don’t like that the bulb sticks out. But I’m not planning to experiment with short-neck flood-style bulbs or anything. I just turn the shade a bit further away from me than I otherwise might.

Petty pleasures are real pleasures

I got this lamp a couple of weeks ago, and still notice how much I like it. It relates to the way I work, since I often write my notes by hand on a pad. It’s hard to judge what small change might make your life a bit easier or more pleasant, but it’s worth thinking about. I was oddly loyal to my old halogen for a bit too long.

What small change in your surroundings have you been considering but putting off?

On The Inappropriate Disposal of Coffee

I was recently struck by a story in The Guardian, about a woman in west London, Burcu Yesilyurt, who was initially fined £150 for pouring out what was left in her takeout coffee cup before getting on the bus, presumably to get to work.

The kind of thing we never used to know, or care about

Doing so violates section 33 of the Environmental Protection Act 1990, which is mostly about disposal on land, but does have the covering statement “a person shall not…treat, treat, keep or dispose of controlled waste (or extractive waste) in a manner likely to cause pollution of the environment or harm to human health.” The bolded phrase is what was cited as justification for the fine.

They told her she should have poured the coffee into a disposal bin. I generally dislike doing this because it leaves a giant, sloshy mess, and in places where actual humans have to pull out the bags and haul them somewhere, it makes their job even less pleasant than it already is.

How absurd is a fine for pouring out coffee, anyway?

So, I initially thought the whole thing was ridiculous. But a bit of looking around (with the help of ChatGPT 5.1) indicates that coffee, and particularly its caffeine, is an actual hazardous substance, particularly when poured into storm drains that pour unprocessed into rivers, lakes, or the sea. Even if it is poured down the sink, most waste processing doesn’t get rid of caffeine.

The volumetrics of unused coffee disposal is a shockingly understudied topic, but clearly, everyone drinks a lot of coffee, and buys even more than they drink, so there is always a lot of coffee going down the drain. Though I suspect much of that volume is from places that brew coffee and have to get rid of old coffee, because if they tried to sell it, they’d quickly have no business.

Fresh hot coffee at any time is one of the wonders of modern civilization. But, like most of the rest of our luxuries-turned-necessities, it comes with environmental externalities.

The environmental impact of staying awake in Sweden

Rosa Hellman, of the Swedish University of Agricultural Sciences, wrote a Master’s thesis (PDF) on liquid coffee waste in food-service settings. She estimated that Swedish restaurants and cafés discard something like 17,000 to 25,000 tonnes (that’s metric tons, 10 percent bigger than US tons) a year.

Hellman also cites a food diary study by the Swedish Environmental Protection Agency, which found that in 2020, households disposed of 190,000 tonnes of food and beverages through sewage, and 45 percent of that was coffee or tea. That works out to around 8 liters per person per year of coffee and tea, not separated in the study.

The restaurant and the food diary numbers are about two quite different things, so they can’t be combined to give a good estimate of how much coffee is getting into the water.

What’s the appropriate fine for inappropriate coffee disposal?

Ms. Yesilyurt felt that £150 was a bit steep, and I’d have to agree. The fines in the now-notorious section 33 seem aimed at businesses and other larger operations, not individuals getting rid of whatever is left in their takeout (well, this is London, so “takeaway”) cups. £150 is the minimum amount stated.

Interestingly, the Guardian article didn’t bother to tell us how often anyone in the UK gets fined for inappropriately disposing of coffee. I can’t find any source of information. Was Ms. Yesilyurt the first person in the UK ever hit with this fine? After she made an issue of it, they rescinded the fine, which seems to be the way these things always work.

The article does add a bit of back and forth over whether the cops were too aggressive, but that seems mostly because anyone confronted by three police officers and given a significant fine after doing what Ms. Yesilyurt did would be startled and nervous. Though this does stimulate the usual harrumphish but not inappropriate question: “Isn’t there some actual crime you people should be preventing or solving?”

What made them decide to do this? Ms. Yesilyurt seems a demure sort, not dangerous or likely to be aggressive, so maybe she was just an easy target at that time of day.

Who should bear the cost of preventing environmental damage?

Every action has an environmental consequence, but when is it worth tying that consequence to the small daily actions of individuals? A fine commensurate with the actual impact would be too small to be worth levying. And if cafés then have to bear the burden of undrunk coffee somehow, that makes it marginally less appealing to run such a business, and we like to have our cafés on every corner. Who is the lowest cost avoider here?

How well would you, bleary-eyed and about to get on a bus to get to work, have reacted in this situation?

A sphexish jabroni with sanpaku eyes: your vocabulary words for the day

I've been reading most of my life, and like to think that I have a decent vocabulary, but always get kind of a charge when I come across an unfamiliar word, particularly an odd or useful one. Recently I've come across three that were new to me, all of them, as it happens, in comments to blogs and Substacks that I read.

Sanpaku

This universally comes in the form "sanpaku eyes", which is how I saw it, in a comment to a photograph on Shorpy, a site that restores and enhances old photographs of the United States, a daily visit for me. In this case, the photograph was Dandy Duffers: 1900, and the comment referred to someone sitting on a fence rail as having sanpaku eyes.

This is the character in question:

Labeled with a term he could never have known

Sanpaku is a Japanese term and means "three whites", where the white of the eye, or sclera, is more visible than average, either above or below the iris. In medicalese that's called "scleral show". The pop culture figure easiest to see it in is Billie Eilish:

The celebrity most often used to illustrate sanpaku eyes

This is just a cosmetic feature, but there is no random feature that someone doesn't turn into a way to rank and judge people. Japanese face reading has all sorts of ways of telling you your fate from sanpaku eyes, though, like most such superstitions, it's inconsistent, so you can pick the fate you want.

I'm unlikely to use this term, even to describe a character's appearance, because I'd have to explain it, and that takes all the fun out of it. Unless they are then sclera-shamed by some other character, in which case it becomes yet another skirmish in the battle against ridiculous and sometimes harmful folk beliefs.

Sphexish

I found this is a comment to the Astral Codex Ten post Highlights From The Comments On Mentally Ill Homeless People, which is itself a follow-on to the post Details That You Should Include In Your Article On How We Should Do Something About Mentally Ill Homeless People.

Both, incidentally, are well worth reading, particularly if you are interested in urban issues. They are about demanding specific policy proposals for specific problems, in this case homelessness, specifically in San Francisco. What are you actually going to do? He is particularly amusing on what, exactly, getting "tough" actually means. And many of Alexander's commenters respond to his demands with actual specifics, not that common in a world where vibes are only product politically active people seem to either want to generate or consume.

Commenter CSW, a former public defender, about commitment hearings for mentally ill homeless people, discussed the amount of time it takes to deal with someone in that condition:

It was not uncommon to have hours worth of conversations with clients who had fixed delusionary beliefs, sphexishly returning to non sequiturs, and/or severe problems with logic and memory.

"Sphexish", what a delightful word. It refers to the repetitive actions of digger wasps (genus Sphex), and apparently came into usage via cognitive scientists Daniel Dennnett and Douglas Hofstadter. Though apparently the initial story underrates actual digger wasp behavior (there is no satisfying illustrative anecdote that is absolutely true as stated), the word has come to be used for repetitive behavior that shows a lack of self-reflection, and thus a certain lack of consciousness.

Jabroni

This is more slang than a word that delineates a previously vague concept. It seems to have been associated with Dwayne Johnson, The Rock, and just an insult to refer to a dumb person you feel contemptuous of. Probably more detail than you need. Though it comes from pro wrestling, and refers to some kind of apprentice wrestler whose job it was to lose to the talent, it doesn't seem to have much specificity to its meaning. It just has a nice jerky lilt to it.

I found it in a comment to this thoughtful analysis of the disaster of NYC congestion pricing: The Death of NYC Congestion Pricing, by Joey Politano, on his blog Apricitas Economics, which is generally interesting, though I read it mostly for housing and transportation issues.

The comment, by Edward Williams, was about problems facing the LA Metro, which Williams feels is actually pretty good. One of the problems he sees is:

public safety on trains and buses - every time a bus driver or passenger is assaulted, the LA times trots out the same nonprofit jabronis that say that the solution is free fares or jobs or whatever.

Politano liked the comment, and I do too. Safety on public transportation is key to wider use by a wider range of people, many poor and needing to get to work, childcare, and the grocery store, and those who oppose the effective enforcement of public safety laws are the common enemies of all transportation users, their motivations more about their own egos than about anything that actually benefits anyone. And free fares, in an attempt to solve an almost nonexistent problem without actually doing anything, cause a range of other real problems.

But "jabroni", pleasing though it is, will never been in my own insult dialect. I'll always worry I'm using it with a slight but distinct accent, as in "You are a sphexish jabroni with sanpaku eyes".

San Francisco and Boston: a tale of two trolleybuses

A couple of weeks ago, Miss E had a business trip to San Francisco, so I came along the weekend before. Her Airbnb was in the western part of town, an area called The Richmond (note: not Richmond, which is somewhere completely different). It’s a nice area, pretty quiet, mostly two and three-story residential, most with a garage taking up the first floor.

Pi.1415926535, CC BY-SA 3.0 <https://creativecommons.org/licenses/by-sa/3.0>, via Wikimedia Commons

The old part of town is way east, so we took a bus to tour around down there. Actually it was a trolleybus, an electric bus powered by dual overhead wires. I gather that trolleybuses are favored in San Francisco over diesel because of their better ability to climb hills.

The trolleybus was comfortable, but stopped pretty much every block. Short distances between stops seems to be a common problem with municipal bus systems. I presume it’s because locals all demand that the stops be close to their house, and claim hardship if they are farther way. But that’s just a guess. All I know is that the quiet, quickly accelerating bus could never get up to speed because it was stopping again almost immediately. This made the trip way longer than it need to be.

The ride did make me miss trolleybuses, or trackless trolleys, as they were once known in the Boston area. Even when I moved here in the 80s, most people just called them electric buses. A year ago, all the ones in my area (Cambridge, Watertown, Belmont ) were eliminated.

Adam E. Moreira, CC BY-SA 3.0 <http://creativecommons.org/licenses/by-sa/3.0/>, via Wikimedia Commons

The MBTA gave a lot of reasons why they had to go, and the reasons were familiar: the system was old and hard to maintain, and very few cities use them anymore. Some road work meant it would be expensive to replace the overhead wires, so why not just take them down since new battery-powered buses would be here soon? Using diesel buses would be a stopgap just until the new buses show up in 2024. Well, given the MBTA’s record of project rollout, we’ll have to see how long this stopgap is.

I got the impression the electric buses just seemed a bit dowdy. And I get the San Francisco keeps its system because of its hills. They are even converting diesel lines to trolleybuses.

I think battery-powered buses will have their problems, including the complex charging infrastructure, unanticipated maintenance problems (these are always unanticipated), and the disposal issues. People need to take a harder look at a working system before they decide it should be replaced by something sexier.

Like everyone else with an interest in public transportation, improving urban life, and mitigating climate change, I want the MBTA to make good decisions. I’m just not sure this was one of them.

Problems of Central Asian geography

My college library (Doheny Library, at USC) had old dark stacks that clearly dated from the era when you had to request a book and have it provided by a librarian. Actually, the same was true of my graduate school main library and here in Cambridge until they rebuilt what is somewhat oxymoronically called the Main Branch.

It was full of what can only be called tomes, and I remember seeing several evocative and challenging titles. One was Prosopography of the Later Roman Empire. Another was The Empire of the Steppes: A History of Central Asia, by René Grousset, with a severe black cover.

I did not essay either of these books at the time. I’ve been meaning to dive more deeply into geographies and eras of world history that I had little knowledge of. Earlier this year Razib Khan, a big reader of history as well as science, presented several reading lists, and I decided to start reading my way through the first one, on history.

The Eurasian steppe belt is in teal, with mostly forest to the north, deserts and mountains to the south

These are damned, thick, square books, and each takes a few weeks to read. So now I have read The Empire of the Steppes! It’s an old book (first published in French in 1939, translated into English in 1970), and extremely dense, covering steppe empires from the Scythians to the last remnants of the Mongols in the 18th century.

There are many tribes, many rulers, many betrayals, many overthrows. But it is also rewarding, because it really gives you a sense of the sweep of both geography and time. It is a vast overview of events between the southern Ukrainian steppe and Manchuria.

I can’t keep Central Asian geography in my head. Grousset provides useful schematic endpaper maps showing the different steppe regions as well as forested and mountainous areas, and a number of line maps for various empires and historic events. The dust jacket is taped down over the endpapers. Fortunately, front and back have the same map (perhaps anticipating this), so I could put the whole thing together by flipping back and forth.

The specific more-detailed line maps were extremely helpful for the events being covered in that section, but I still had a lot of trouble relating the schematic to the topography and landscape. I’ll keep working on it, though. It’s clear that the history we know well, essentially that of peripheral sedentary states, was strongly influenced by and often dependent on events in the center of the continent.

Oh, and I have still never read Prosopography of the Later Roman Empire. Roman prosopography studies social networks and groups to get at an understanding of social life where individual biographies are scarce or nonexistent. I see that the volume I would see has been joined by several more over the past few decades, so I am falling farther and farther behind.

Getting to transit

I live in a small city (Cambridge, MA) right next to a medium-sized one (Boston). Cities that are fun to live in and cities that are easy to drive in are pretty much mutually exclusive categories. Since Cambridge is fun, it is a nightmare to drive through. So I don’t. I don’t even own a car.

I would like it if fewer people drove here, and more people lived here. It would be a healthier, quieter, and safer place, with even more things to do, and more people to run into while doing them.

I live up near the circled T called Alewife, in the upper left of this map

One thing essential to moving in that direction is a fast, safe, and convenient public transportation system. That, unfortunately, is not something Boston’s public transportation system, the Massachusetts Bay Transportation Authority (MBTA), seems able, or, it seems, even willing, to provide. So most people really have to drive to get where they want to go, even if that is expensive, inconvenient, and frustrating.

While the MBTA seems to have been getting steadily worse over the past decade or so, this has stimulated a lot of attention. Trains are delayed, buses skipped, and conditions often unpleasant. The Boston Globe’s reporting on public transportation issues has been great. This has partly been stimulated by organizations like Transit Matters, which has been providing a lot of data on how the system has been operating, and thus compelling the MBTA to open up its opaque system.

This is a moment for me. I have realized that I, an educated citizen of mature years, have no idea of how to figure out what the problems are, how to decide what to do about them, and how to act in a way that helps provide solutions. Local politics are really the only way any of us has a chance to influence anything.

I’ve joined Transit Matters, as well as a local housing advocacy organization, A Better Cambridge. But it is still difficult for me to make a meaningful effort in making the city I live in an even better place. I’m going to try to discuss this, and the issues I have been dealing with, because I’m sure I’m not alone in my mix of ambition and flailing.

Writing With Intent: Muriel Spark on Sex and Prose Style

I'm still on my Muriel Spark kick. And, by the way, I did reread The Girls of Slender Means again. Once you start, you really can't stop.

I’m not sure if the schoolteacherish demeanor is deliberate, or just impatient

Now I'm in the middle of Loitering With Intent. Entertaining, of course, but not as easy a read as some of the others. It's about a writer, Fleur Talbot, who finds that aspects of reality begin the mimic events in a novel she has written. And Fleur gets accused of libeling people despite the fact that she wrote her novel before she met those people, or they did the things she wrote about.

It's inspired by the same period in Spark's life as A Far Cry From Kensington, though is more distant from the actual publishing milieu. It does have more sex, which plays a role in the quotation below. Fleur is talking to a frenemy, Dottie, whose husband Leslie Fleur has slept with. Despite their contentious relationship, which involves stolen manuscripts, among other misbehavior, Dottie and Fleur seem entangled for life. But I haven't finished the book yet.

Dottie and Fleur are talking on the phone. Dottie has just said that Fleur's book, Warrender Chase is "a thoroughly sick novel" (and, thus, we are reading a sick novel). In return, Fleur attacks Leslie's own literary work:

I could hear Dottie crying. I meant to tell her more about Leslie's prose, its frightful tautology. He never reached the point until it was undetectably lost in a web of multisyllabic words and images trowelled on like cement.

She said, "You didn't say this when you were sleeping with him."

"I didn't sleep with him for his prose style."

Muriel Spark herself has an economical way of telling a story, a manner I now aspire to, since all too often I am more Leslie than Fleur.

Muriel Spark on getting a cat to aid in concentration

Cigarette, nibbled pencil, stacked books: the signifiers of authorship used to be so much easier….

I finished reading A Far Cry from Kensington, by Muriel Spark. At a point in the middle of the book, Mrs. Hawkins is invited to a fancy dinner party, and finds herself seated next to "a red-face retired Brigadier General" (rather a stock character, which Sparks is totally aware of):

...I said something to the effect that he must have had an interesting life.

"I could write a book," he said.

"Why don't you?"

"Can't concentrate."

"For concentration," I said, "you need a cat. Do you happen to have a cat?"

"Cat? No. No cats. Two dogs. Quite enough."

(I have trouble explaining why I found that sequence of one and two-word sentences so funny, but I read it over and over out loud).

She then explains to him, at some length, how a cat aids in concentration, and at the end of it he says "Good. Right. I'll go out and get a cat."

Then she give us this coda:

(I must tell you here that three years later the Brigadier sent me a copy of his war memoirs, published by Mackintosh & Tooley [the publisher where Mrs. Hawkins works, though not for long, since her bête noire, the pisseur de copie, met in the previous post, will soon put in an appearance]. On the jacket cover was a picture of himself at his desk with a large alley-cat sitting inscrutably beside the lamp. He had inscribed it "To Mrs Hawkins, without whose friendly advice these memoirs would never have been written—and thanks for introducing me to Grumpy." The book itself was exceedingly dull. But I had advised him only that a cat helps concentration, not that the cat writes the book for you.)

An absolutely fantastic book, with a light surface and some darker undertones (typical of Spark), well structured, does not outwear its welcome.

Don't get on the bad side of a Muriel Spark character

I like to say that I like the writing of Muriel Spark, but then, if only to myself, have to admit that this is based solely on The Prime of Miss Jean Brodie, her most famous work, the movie version of which impressed me, at 13, with its sophistication...and, by extension my own, and on The Girls of Slender Means, which I love, and have read a couple times.

Muriel Spark in Rome, c1969

So, feeling somewhat at loose fictional ends, I got from the library one of those Book of the Month Club omnibus volumes where several separate books, complete with their own margins, fonts, and pagination, are squished together between hard covers, as through some mysterious geologic process. This one has, in addition to TGSM, which I plan to read for a third time, Memento Mori and the book I checked the volume out for, A Far Cry From Kensington.

It is told in a slippery time structure similar to the other two I have read, with the narrator (in this case, Nancy Hawkins, describing her life in 1954, when she worked at a dying publishing house) free to hint at what is to come, contrasting how she felt about something when it happened to how she feels about it as she is writing, and unexpectedly telling you a character's fate, or alternately, revealing something about her own past.

She has just met a character who will clearly be significant in her life (although I, the naive reader, have no idea why yet):

At this point the man whom I came to call the pisseur de copie enters my story. I forget which of the French symbolist writers of the late nineteenth century denounced a hack writer as a urinator of journalistic copy in the phrase 'pisseur de copie', but the description remained in my mind, and I attached it to a great many of the writers who hung around or wanted to meet Martin York [her boss, failing and now seemingly committing serious fraud]; and finally I attached it for life to one man alone, Hector Bartlett.

Then, somewhat further on:

Pisseur de copie! Hector Bartlett, it seemed to me, vomited literary matter, he urinated and sweated, he excreted it.

'Mrs Hawkins, I take incalculable pains with my prose style.'

He did indeed. The pains showed. His writings writhed and ached with twists and turns and tergiversations, inept words, fanciful repetitions, far-fetched verbosity and long, Latin-based words.

If you are a writer, you might sometimes wonder what the employees at your publisher really think of your work. Perhaps better not to inquire.

Balancing a cabinet on the edge of disaster: France 1940

I recently read Strange Victory: Hitler's Conquest of France, by Ernest R. May, about the Fall of France in 1940. May has written a number of books about executive decision making in foreign policy and intelligence assessment. Strange Victory is very much about that. May gives you an excellent view of the constraints facing the various participants, the constraints they thought they faced and maybe didn't, and how events can seem to go with glacial slowness, only to suddenly accelerate without warning.

On September 1, 1939 France and Britain, having promised to do so without anticipating that they would really have to follow through (the point was to dissuade Hitler from going ahead, they didn't really expect to have to do it), declared war on Germany in response to its invasion of Poland.

But then they didn't really do very much. There is some reason for this, there were a number of difficulties, but they both futzed around from then until May 10, 1940, blaming each other for things, when Hitler decided put an end to their prevarication.

So it's interesting to contemplate some of the problems facing Edouard Daladier, France's prime minister, after the declaration of war:

The autumn of 1939 had been frustrating for Daladier. He had tried to form a national government, hoping that in wartime he would not have to continue formulating every act or decree like a pharmacist preparing a complicated prescription. With the goal of at least having his own Radical Socialist Party [despite the name, a centrist party] united behind him, he asked Herriot, his old mentor and rival, to replace Bonnet as foreign minister. But Herriot said he would serve only if the cabinet also included Marshal Pétain, and Pétain refused to serve with Herriot on the ground that Herriot's appointment as foreign minister would alienate Mussolini and Franco. Socialist leaders also refused to serve either with one another or without one another. Paul Faure, himself ineligible because a pacifist and unregenerately munichois [as appeasers were known after the Munich agreement of 1938], swore to oppose a cabinet that contained Léon Blum or any socialist on Blum's side. He reportedly said that, if Blume entered the government, "then all Israel with him! That would be war without end!" [no surprise, Faure ended up serving Vichy, Blum in Buchenwald]. Faure and Blum alike threatened to vote against a cabinet that included anyone from the right; Flandin and others linked to employers' groups vetoed inclusion of even a moderate trade-union leader.

Soon after, Daladier yielded the premiership to Paul Reynaud—but Reynaud had to retain almost all of Daladier's cabinet, including Daladier as both minister of defense and minister of war. That's cabinet politics in the twilight of the Third Republic.

The closer you look at history, the less clear its lessons seem to be, and the more complex and tangled the lines of causation. May sees France's (and Britain's) failure as largely down to poor acquisition and management of intelligence. I'll try to take a look at that in a bit.

The poetical joys of Byzantine hierarchy

Steven Runciman's The Great Church in Captivity, is, as its subtitle explains, "A study of the patriarchate of Constantinople from the eve of the Turkish conquest to the Greek War of Independence". But it is actually more than that, because its first seven chapters, roughly 40 percent of the book, are a detailed explication of the Orthodox Church, its theology, structure, movements, and relations with the West.

Yeah, kind of a specialist read, and not the best introduction to Eastern Orthodoxy if you aren't familiar with it. Still, I have a weakness for Byzantine official titles, and while detailing Byzantine administration Sir Steven manages a magnificent hierarchical aria:

...five great offices remained throughout the Byzantine period. They were headed by the Grand Economus, who was in charge of all properties and sources of revenue and who administered the Patriarchate during an interregnum; by the Grand Sacellarius, who, in spite of his title, had nothing to do with the Purse but was in charge of all the Patriarchal monasteries, assisted by his own court and a deputy knows as the Archon of the Monasteries; by the Grand Skevophylax, in charge of all liturgical matters, as well as of the holy treasures and relics belonging to the Patriarchate; by the Grand Chartophylax, originally the keeper of the library but, after the disappearance of the Syncellus and the Archdeacon [vanished offices discussed earlier], the Patriarch's Secretary of State and director of personnel; and finally by the Prefect of the Sacellion, keeper of the Patriarchal prison and in charge of the punishment of ecclesiastical offenders. These five officials were members of the Holy Synod and ranked above all metropolitans [that is, bishops of large metropolitan sees].

If you survived that, you may well enjoy the book. But even I limit myself to small doses, lest I faint from the smell of frankincense.

In Search of the Boston Green Head, Part One: Basanite, Bekhen Stone, and the Wadi Hammamat

A moody and weird Julius Caesar, probably Egyptian, probably from the century or so after his death

This is the first of several posts related to a particular type of stone, and a small bust of an Egyptian priest made from that stone, in the Boston Museum of Fine Arts.

A few years ago I was browsing through Philip Matyszak's Chronicle of the Roman Republic (a series of short biographies of various figures during that era, with informative sidebars, stemmata, and timelines, it is made for browsing, at least if you are a Roman history geek), when I came across a picture of this striking bust of what is most likely Julius Caesar, in Berlin.

The caption in the book says it is carved from green "basanite", a stone from Egypt. Clearly the stone is sculpturally useful, dense, hard, of consistent texture, capable of being smoothed, and of striking color.

I've written about sculpture before, and am toying with a book about a sculptor in the late 19th century. So I was curious: what is basanite? Sculpture is the encounter between an idea and a specific physical reality. Well, OK, that describes art in general, but it seems clearest in the case of sculpture, whether stone, wood, metal, or clay, each of which is a demanding partner in the work.

But the identity of basanite is not as easy a question to answer as you might think.

The mysteries of geologic nomenclature

The stone in question is not always called basanite, and the history of the word "basanite" itself can cause confusion as well—after you read my detective work, I will reveal that the criminal is actually someone completely different. The material of the bust called the Green Caesar is sometimes described as greenschist, slate, or greywacke.

As we shall see later on, there is a completely different stone, an igneous extrusive called basanite, and which confused me for a long time. Here, "extrusive" just means the igneous rock formed on the surface, rather than underground. Confusingly to laypeople, geologists often give completely different names to the extrusive and intrusive versions of otherwise identical rocks—for example, what is granite when it forms deep underground is rhyolite if it forms on the surface. This igneous basanite is very much like basalt (an extrusive igneous rock whose intrusive version is called gabbro, if you must know).

There are also historical reasons this rock is called basanite, but the rock the head is made of is something else entirely. For now, I will call that stone "sculptural basanite" (a term of no validity outside this blog post—let it be our secret), though I will come up with a more precise name a bit further on, though be warned that that name will be just as much of a local variable as sculptural basanite.

On the track of sculptural basanite

In The Materials of Sculpture (an informative book covering stone of all kinds, as well as ivory, bronze, clay, and wax, among other things) Nicholas Penny says

The most highly prized Egyptian stone to be employed for sculpture on a large scale was basanite, commonly confused with basalt (and still usually described as such on museum labels [the book is from 1993]). This is a "greywacke", a by-product of the decomposition of basaltic rock.

"Decomposition" is geologically imprecise, but this places the stone at its place of origin, Egypt, and also demonstrates how the terminology can be vague or confused (it is the other kind of basanite that can be confused with basalt). Unfortunately for interested amateurs like me, geology has no equivalent of a species, where there is only one name for one thing, and the first discoverer gets absolute naming priority. Rock types are not even as clear a species, and it feels like there was always a lot of competitive naming going on in geology, with everyone convinced that the rock in their mountains was quite different than that allegedly similar rock in those mountains some sadly lame earlier geologists wasted their time investigating.

In "The building stones of ancient Egypt – a gift of its geology" (PDF), Klemm and Klemm say

green siltstones, dark green greywackes and conglomerates, which are best exposed in the Wadi Hammamat...[where] an impressive quarrying activity is documented by almost 600 rock inscriptions over a time interval from Predynastic until the late Roman period (about 4000 BC until 300 AD). These many inscriptions concentrated along a relative short distance in the wadi obviously indicate the uniqueness of this site and its extraordinary importance to ancient Egyptian culture. Consequently the rock type extracted here received a special name: ‘‘Bekhen-stone’’, as reported in many ancient Egyptian documents.

This Bekhen-stone is the basanite I got interested in. The Wadi Hammamat, far off in the Eastern Desert, required long overland transportation to the Nile, under difficult, waterless conditions. But I am not alone in being struck by the character of this stone:

Particularly the very dense, medium-grained dark green greywacke was used during the entire Pharaonic era and on until the Ptolemaic (Greek) period (from 332–30 BC), mainly for sarcophagi, archetraphes [sic— I'm assuming they mean "architraves", but am not sure] and for the finest carved sculptures of Egyptian antiquity. Scattered unfinished or broken sarcophagi indicate that at least the raw form of these vessels were worked out at the quarry site, which is understandable as they had to be transported over 90 km through the desert, until shipped on the river Nile to their final destination.

Most of the royal sarcophagi of the [Old Kingdom] and about 100 sarcophagi for private individuals of the Late Period (since 600 BC) and of Ptolemaic and Roman times were made of this rock variety.

Wadi Hammamat should totally be more famous

The Wadi Hammamat is totally cool, historically, geologically, and geographically, and has to have been the setting for some historical fantasy or other. Anyone know of one? My reading in that genre is pretty sparse.

OK, that's enough for now. Next time, we finally distinguish sculptural basanite from basalt, and then have a look-in on its dark twin, sculpturally useless igneous basanite. I bet you can't wait.

Musical imitations as their own form of art

Composers often play around by inhabiting an older style of composition. This seems to be particularly true of the modernists of the twentieth century. In part I suspect it's a way of demonstrating beyond question that they know exactly what they're doing, that what they normally write isn't just a bunch of random noise, no matter what you yahoos think.

Revivals as a way of getting into the essence of a style

You can revive the musical style, but the clothes just never work

Revivals often allow an artist an interesting perspective. Looking back, they can see more clearly the quiddity of a style, what makes it what it is, than practitioners at the time possibly could. In addition, there is no idiosyncratic patron, perhaps one who is an enthusiastic amateur performer on the flute or bassoon, breathing down their neck. Sometimes it is an homage in the form of habitation, of getting inside the mind of an admired predecessor. And sometimes it is just fun dance music, more an exercise in creative orchestration than anything else.

Hence works like Prokofiev's First Symphony, "the Classical Symphony", and Tchaikovsky's Fourth Orchestral Suite, "Mozartiana".

Cute and sometimes moving

I actually like a lot of this kind of work, and used to pick it up on used LPs. Much of the lighter side of it comes out in ballet suites based on Baroque composers, such as Thommasini's "The Good-Humored Ladies" (after Scarlatti) and Walton's "The Wise Virgins" (after Bach). and Beecham's "Love in Bath" (after Handel). Very light, but a great pleasure.

Also in the dance vein are a couple of Richard Strauss suites: "Divertimento" and "Dance Suite" (both after Couperin).

More melancholy and moving is a piece by the less-known Alfred Schnittke. of partly Volga German ancestry, who lived during the Soviet period. Suite in the Old Style is lovely. It is not in imitation of anyone in particular, just Baroque in general, and in fact is a reworking of various film work that he did, including for an animated children's film. There are many arrangements, some for chamber orchestra, some for cello and piano, but I favor the violin and piano version at the link. I particularly like the Vadim Guzman performance, but it does not seem to be available on YouTube.

Are there any stylistic revivals that you find particularly interesting?

Mine are mostly Moderns doing Baroque. Are there other combinations I should check out?

Audiobooks, nationalities, and accents

Like many people, I listen to a lot of audiobooks. I tend not to like listening to fiction, partly because I can get tired of a book, but want to check other things out about it. Sometimes a book has interleaved sections at two different times or two different points of view, and only one of them is interesting. So I want to be able to skip and skim. Fellow writer, I apologize, but you sometimes write just...a...bit...too...much.

An example is the much-praised Water for Elephants, by Sarah Gruen. The present of the story takes place in a retirement home, where the embittered older character remembers his adventures in the circus during the Depression. The circus sections were pretty good, but the retirement home sections were insanely boring, much like many retirement homes. Since I couldn't just skip them, I gave up on the book, potentially losing whatever there was about it that people really liked.

Narrative nonfiction is the way to go

So I generally go for narrative nonfiction. Michael Lewis and Sarah Vowell books are both great listens. Vowell reads her own, and also gets various performing friends to do the voices and even compose songs for the audio versions.

I've also recently listened to David Wootton's book about the Scientific Revolution, The Invention of Science. Andrew Jackson O'Shaughnessy's account of the British leadership during the American Revolutionary War, The Men Who Lost America, and Sharon Bertsch McGrayne's account of the history of Bayesian statistics. The Theory That Wouldn't Die, all worth a listen. If you don't think a book about Bayes' Theorem sounds interesting, then it probably won't be.

Do they need to do the police in different accents?

Do you think these guys are Germans?

The last two books I listened to were David Quammen's account of the science and the politics and personalities of horizontal gene transfer, The Tangled Tree, and Adam Zamoyski's history of elite paranoia between the Napoleonic Wars and the Revolutions of 1848, Phantom Terror.

Both are excellent books, but the readers of both indulge in something I find extremely irritating: when directly quoted, people who were born in various countries all speak in the stereotyped accents of their nationalities. French, German, and Russian are the most common.

This, despite the fact that much of what is being quoted was written down, or spoken in their own language, or, if not their own, then a language other than English. Most people don't write with an accent. If they are non-native speakers writing in English, their word choices and syntax might reveal that, but that will be in the original, with no need for the reader to add anything.

It doesn't help that one of the readers manages to mangle Russian names and terms even while affecting to talk like a Russian. For what it's worth, I think the French words and names are better.

And what do they think this is adding? A good reader certainly creates voices for various characters. but a stereotyped national accent is scarcely the best way to do that. It makes the whole thing sound like one of those national-stereotype-filled movies from the 1960s set just before the First World War, like Those Magnificent Men in Their Flying Machines or The Assassination Bureau. Sacre Bleu (which The Simpsons once helpfully subtitled as "Sacred Color Blue")!

I wouldn't have remarked on the accent thing, but it was two books in a row, and, unlike subtitles in a movie, I can't turn it off.

What do you think about the use of accents in audiobooks?

Or do you think I'm just too sensitive?

The protective force of cliche

Trees with departing leaves or birds are a common metaphorical image for dementia

My mother recently suffered a minor stroke. Combined with what seems to have been several prior, undetected strokes, she has gone from mild to fairly significant dementia. My sister and I were talking about her recently, and she listed a bunch of cliche phrases that my mother used, like "One day at a time", "time marches on", "so far so good", etc. You can lose a lot of vocabulary and mental capacity and still keep stock phrases.

In "Circles", one of the short sections that makes up the wonderful A Primer for Forgetting, Getting Past the Past, Lewis Hyde remembers his mother's failing intellect. She says a phrase to her husband over and over.

"You're going in circles," Father says. They say the CAT scan showed some atrophy of her frontal lobes, but the old material is still there. She is very much her old self. Her verbal tics and defenses remain. "Well, now Mrs. Pettibone," she says to herself, staring into the refrigerator before dinner. "We'll cope." "We'll get along." She is the shell of her old self, calcified language and no organism alive enough to lay down new layers.

Would it be possible to live in such a way as to never acquire habits of mind? When my short-term memory goes, I don't want to be penned up in the wickerwork of my rote responses. If I start being my old self, no heroic measures, please.

My brother was just over at my mom's. When he brought some boxes in from the storage area, she looked at them and said that she is empty box with nothing inside.

Sometimes, behind the wickerwork, you can see eyes peering out.

How the 19th century Austrian secret service proved that people will give up private information if you save them time and money

The period between the Congress of Vienna in 1815 and the wave of "revolutions" in 1848 was one of repression, police action, and poorly organized, chaotic attempts at revolution that were quickly suppressed, with the exception of a few changes of government in France. That last year, 1848, was marked by much larger but still poorly organized and chaotic attempts at revolution, that resulted in retrenchment, and the replacement of the moderate monarchy of Louis-Phillipe with the less moderate if possibly more colorful Second Empire of Louis-Napoleon (there is no burden the people find more onerous than that of having to make choices).

Not that the repressive post-Napoleonic regimes were much more impressive than the attempts at revolt against them. According the Adam Zamoyski in Phantom Terror: Political Paranoia and the Creation of the Modern State, 1789-1848 (I listened to the audio), every leader from Metternich and Tsar Nicholas I on down spent their time obsessing over large-scale conspiracies of Masons, French revolutionaries, and Illuminati that did not exist, without recognizing that the various civil disturbances that they had to keep dealing with were really a response to the hopes of liberation and national autonomy that had been raised by the events of the quarter century after the French Revolution, and then dashed.

And it is kind of weird to think that the period from the Fall of the Bastille to the exile of Napoleon Bonaparte to St. Helena is only 25 years long. In his superb podcast The History of the Twentieth Century, Mark Painter calls the Congress of Vienna "an attempt to hit Ctrl + Z on the French Revolution" (this is based on memory, and may not be word-for-word accurate). Unlike a typo, it kind of stuck around.

But, still, the state apparatus of control put in place over that period worked pretty well at keeping the regimes in place until after an entire century they were all dumb enough simultaneously to start a gigantic war that none of them could win, and all of the multinational empires came apart like tissue paper you left in your back pocket when you threw your jeans in the washing machine.

The inescapable attraction of cheap and fast

In addition to his diplomatic skill, Metternich had quite the reputation as a lover. He is around 60 in this portrait.

Intelligence, opening people's mail, breaking codes, and other familiar practices grew through this period, though they look fairly amateurish from the point of view of intel operations over the twentieth century and into our time. Much of the letter reading seemed mostly to provide princes and leaders with salacious material about colleagues and rivals for personal entertainment.

Prince Klemens von Metternich was the dominant figure in Austria, and he wanted to make sure he could read everyone's mail. How he managed this is entertaining, and instructive.

According to Zamoyski, in the early part of the period, during the 1820s:

Metternich...identified control of the postal service as a key element in the invigilation of Europe. In the course of the eighteenth century Vienna had, by providing the most efficient postal service throughout the Holy Roman Empire, gained access to the correspondence passing through central Europe. Although the Empire had been abolished, much of the post carried around its former territory still passed through Austrian sorting offices. Metternich managed to extend this to Switzerland, a natural crossroads as well as a meeting place for subversives of every sort. All Swiss post passed through Berne, whose postal service was in the hands of the conservative patrician de Fischer family, with the result that all mail between France, Germany, and Italy was accessible to the Austrian authorities. Most of the mail going in and out of Italy passed through Lombardy, where it came under Austrian police scrutiny.

Metternich attempted to extend this to the rest of Italy, but failed, due to Papal opposition. Incidentally,"invigilation" seems to now be used only for proctoring exams, but it should probably be revived in the Zamoyski's wider sense.

Later on, in the 1840s, Metternich had to work a bit harder:

To ensure that as much European mail as possible continued to pass through Austrian domains, Metternich saw to it that the Habsburg postal service was cheaper and faster than the alternatives.

This apparently put a huge strain on the intelligence operatives, who had to open, copy, reseal, and return mail of interest to the post office without incurring additional delay. Zamoyski has some fun with how overworked this small set of bureaucrats was.

People knew there was a good chance their letters were being read, and tried various subterfuges to make it harder, but they still used the Austrian post, because it really was cheaper and faster than the alternatives.

While entertaining, the book is fairly narrow on the topic of government responses to subversion, and is not anything like a general history of the period. It may go into more detail on surveillance that some readers (or listeners) might like, but I found it extremely entertaining. Zamoyski knows how to feed in a lot of information without getting tedious, a rare skill among historians.

What privacy do you give up for cheap and fast?

I'm not even sure that's worth asking, because it's pretty much everything, for all of us. Metternich seems to have pioneered this form of big data gathering.