After reading, and writing my longest ever post about, Helen Lewis’ The Genius Myth, I entered a slightly fallow period in terms of reading engagement and motivation. I’d started reading The Genius Myth alongside Rutger Bregman’s Moral Ambition, but my interest in that book faded,1 and I struggled to find something as engaging as the book I’d just finished.
Early in the third week of September, however, this fallow period finally ended, when I heard Tim Berners-Lee (TBL) being interviewed by Rory Stewart and Alastair Campbell in The Rest is Politics Leading podcast series, to promote his new autobiography, This is for everyone, which in a sense is also an autobiography of the world wide web, his invention. I’d ordered the book within a few minutes of the podcast ending, and finished reading it in about three days (which is fast for me).
Here are nine thoughts I had prompted by (primarily) the book and the podcast:
Idea 1: “Why are you only a millionaire, not a billionaire?”
This seemed to be the subtext of (what felt like) at least half a dozen questions from on the TRIP Leading podcast, especially from Rory Stewart. To Stewart and Campbell, and I’m sure many who swirl in their social circles, the web is an engine of supreme wealth generation, a place that has created a new billionaire class, whose products affect almost everyone on Earth every day of our lives in some ways. So Stewart and Campbell seemed perplexed that the creator of the web, someone who ‘ought’ to have had the supreme first mover advantage, was ‘only’ a millionaire and not a billionaire. Even more perplexing to Stewart and Campbell: why does TBL appear not to want to be a billionaire, and not appear in any way envious of those who used his technology to become thousands of times wealthier than he is? Linked to this - something Stewart warned and apologised to the listener for at the start of the podcast - why is TBL not the highly articulate and polished media performer ‘we’ (?) are now conditioned to expect from Silicon Valley?
These questions weren’t, to my mind, hostile questions so much as questions orthogonal to TBL’s personality, pursuits and interests. To Stewart and Campbell, not seeking to use one’s ability to maximise wealth, power and influence appeared simply not to make sense. The best explanation they could come up with was simply that TBL is an eccentric, though really this is just a (borderline derogatory) label rather than an explanation.
Having read the book, however, I think the real answer is that TBL is, was, and always will be deeply passionate about building and promoting systems that connect information, ideas, and people, and that so long as his material circumstances are never so adverse as to impinge on this lifelong calling, they will always be sufficient. Sometimes money helps, as with ensuring TBL’s web-standards body, W3O/W3C, remains financially afloat; but often it risks making the web ‘worse’, which to TBL seems to have consistently meant something like ‘able to allow free exchange of information’. One of TBL’s proudest achievements, for example, seems to have been to convince CERN to give the world http, in perpetuity, for free [p. 103]; another was to fight, through W3C, for the web to never become de-facto proprietary through intentional disalignment on standards between browsers run by either Microsoft or Netscape, both of which wanted to introduce additions and features to ‘their version’ of the web which were incompatible with other providers. TBL doubtless had many opportunities to make decisions which would have have allowed him to own and profit from a more proprietary variant of the web (such as if he worked with Netscape to build ‘the new Netscape web’), but would have slowed down growth and access to the web: whenever given the choice between owning a larger share of a smaller pie, or owning but influencing a more vibrant and faster growing pie, he seems to have chosen the option that makes the web better, rather than himself richer. His vision, always, seems always to have been much grander than personal fortune or political power. If that’s ‘eccentric’ for Stewart and Campbell, then in a sense that speaks mainly to their lack of genuinely transcendent vision, rather than TBL’s inherent strangeness.
Idea 2: CERN and http: Is organisational efficiency fundamentally a myth?
I vaguely remember seeing and hearing an online lecture by TBL many years ago, where in his excitable, self-distracting, and frenetic style he said something like “I took my proposal [for http] to my line manager. He said ‘I don’t really understand what you’re talking about. Why not spend a few weeks working on it?’” Within the book I can’t find the exact same quote (which of course I might have confabulated), but there are still plenty of references to those around TBL have difficulty understanding his vision, and the proportionality and relevance of his proposals to his specific remit at CERN, but nevertheless recognising his passions and ability, and supporting him in developing his ideas into technologies.
To this end, TBL writes of effectively being ‘protected’ by his bosses and his bosses’ bosses, despite them often having little understanding of what he was doing.
What does this say for the very idea of organisational efficiency? Though there were likely some genuine benefits to developing http within CERN, the specific examples - such as being able to find people’s phone numbers, linked research publications, and brief biographies of fellow staff at CERN a bit more easily - may well have never, in a strict sense, have been proportionate to the cost of investment. If CERN had been a more ‘efficient’ organisation, an organisation where all costs and expenses have to be justified clearly within work programes and cost centres, and audited carefully to justify the organisation’s costs to the European tax payer, it may well have killed the web before it was ever born.
Even considering the discovery of the Higgs Boson, however, the web is almost certainly the most important contribution CERN has made, or will ever made, to humanity. And it would never have done this if CERN were a truly ‘efficient’ organisation. Is the very idea of organisational efficiency therefore something of a dangerous myth?
I suspect the answer’s both yes and no. Efficiency seems a useful concept when it’s clear what we want to maximise, but an horrific and crushing concept when we don’t. One idea I’ve been fascinated by for many years are fitness surface optimisation algorithms, and the ways that the most ‘efficient’ algorithms tend to lead us to local optima, rather than global optima, when the fitness surface is complex rather than simple. The most efficient approaches are akin to ‘blind hill climbers’: place one of these devices on a hill within the landscape we want to explore, and it will detect locally the gradient around it, and start trundling to ever higher elevation, until the top of the hill is reached. Then it will stop. But if the hill on which the hill-climber is placed is an undulation in the shadow of a mountain, it won’t see or detect the mountain, and forever find itself stuck on the hill instead. At every step along the way, the hill-climber can report incremental improvement, but it will never understand the broader landscape, and the opportunities that might present themselves if it were less ‘efficient’ and allowed occasionally to simply ‘wonder around’.
Any approach that allows the possibility of finding the global optima consistently will necessarily be less ‘efficient’. One approach is to employ an ensemble of hill-climbers, each dropped off in different locations of an invisible and unexplored landscape. (i.e. to give the algorithms different starting values.) If there were 10 hill climbers, and they all end up on the same place as a single hill climber, then ten times as much effort will have been spent on finding the same optima. Clearly very ‘inefficient’. But if one of the ten hill climbers happens to have started at the foot of the mountain, whereas the rest were all somewhere along the hill, and the mountain’s peak elevation is twenty times as that of the hill, then the reward will have been worth the additional effort.
Another, related, related approach is to intentionally build into the algorithms an element of ‘forgetfulness’ and ‘play’, which clearly seems a very inefficient thing to do. Think of these algorithms as more like ladybirds, with both a ‘flying’ phase and a ‘crawling’ phase. 95% of the time (say) the algorithms are in the crawling phase, and behave just like the standard hill climbers. But then 5% of the time these algorithms suddenly start flying around, and land somewhere completely different. Often, this means that the apparent progress towards the top of the last hill’s peak has been lost - something that seems very wasteful - but maybe during the flying phase the ladybird has found its way to an even higher section of the same hill; or found its way to the base of the mountain - the true global optima, rather than the local optima the simple hill climbers would have dutifully crawled.
By analogy, the most genuinely innovative organisations support staff to pursue both ‘flying’ and ‘crawling’, and to this extent trade-off some ‘efficiency’ for more possibility of creativity. Within large organisations, the freedom to fly has often found its way through permitting or actively promoting ‘skunkworks’. And within small tech outfits, through the idea of ‘pivoting’: recognising that an adjunct of one technology and use-case might actually be where the genuine value of a technology and business exists. Getting the balance right between flying and crawling is a perennial challenge, but to only value the measurable ‘efficiency’ of crawling over the immeasurable potential benefits of flying makes for a bad organisation and bad society.
Idea 3: Top down and bottom up modes of enterprise need to know how to dance
The web (or rather http) is not the internet, which was developed in the US through military funding, before spreading through US universities. The military: top-down organisation based on hierarchy and standards operation processes; universities: bottom-up institutions based on exploration and emergent connections. Within the US, top-down started, then bottom-up followed.
Another theme of questions from Stewart and Campbell, once they’d tired of asking TBL why he wasn’t a billionaire, was “Why did the US, not Europe, capitalise so much on the European technology of the web?” The short answer seemed to be that the web was the next step of the internet dance that was taking place in US universities, and so the bottom-up, freewheeling sensibilities of these US research institutions were highly compatible with TBL’s vision of the web. Http was intended as a means of allowing people to tell machines something that people can identify easily but machines cannot: the way that data sources are connected and linked in the human mind. And it was intended to do so in a way that was akin to conversation: originally TBL wanted web pages to be editable by default, rather than just viewable, allowing new readers to themselves become editors, and find and refine the associations between nodes even further. Http was a lightweight framework offering a lot of possibilities for how people build links with each other. This open, democratic (but meritocratic) mentality was much more compatible with the often-anarchic researchers already using and familiar with the internet in US research institutions.
By contrast, it seems Europe was relatively reluctant to adopt the web for the same reasons the hippies and nerds over the Atlantic were keen on it: it was too bottom-up. The European instinct was instead to support and enforce a more top-down model of information sharing, based much more on conventions and standards. And conventions and standards mean meetings and bureaucracy, and meetings and bureaucracy takes time for deliberation and agreement. In short, it seems like Europe lost the advantage of its own technology by trying to tame, organise and categorise it, to make it work better in theory, rather than to adopt and use it in practice. After teaching the US a new series of dance moves, European bureaucracies then refused to dance any further.
Idea 4: The Web was for Information, not Engagement
For TBL, Wikipedia is the clearest working example of the web that he had envisaged. Wikipedia is not, by and large, a particularly addictive website. Instead, it’s something people will use to find information on something, and then once they’ve found that information, they will stop using it. Disengagement is therefore a feature of Wikipedia, not a flaw in Wikipedia.
But much of the modern web is the other way around: intended to engage, not to inform. The reasons for this are obvious once given the types of monetisation that have prevailed over the last twenty or so years: the profitability of a website is broadly proportional to the amount of time people spend on it, because users’ attention is the resource such companies sell to advertisers. Whereas the original aim of the web was to help people find what they are looking for, the engagement imperative of the commercial web instead militated towards almost the opposite: ensuring that users want more, but find less.
Idea 5: Would the semantic web have led to smarter (but less polite) AIs?
Though the web itself saw explosive growth, some of TBL’s later ideas, promoted either personally or through the W3C, have found much less enthusiastic adoption. This includes what TBL referred to as ‘the Semantic Web’, or ‘Web 3.0’. In essence this was an argument, and a recommended set of standards, for embedding more machine readable data and information about associations between entities into websites. From the perspective of most viewers of a website, almost nothing would look any different. But within each site, so went the vision, was an extensive categorisation of information about entities and their relationships which a computer could understand. Eventually, if everyone building and maintaining websites went to the trouble of formally expressing information about what things are, and how they are are associated, then big thinking machines, crawling across the internet, could also become loaded up with all this information, and with enough entities, definitions, and associations, start to learn enough about the world to reason about it.
Perhaps the idea that individuals and businesses would go to a great deal of extra trouble to help computers know such facts as “Dresen is a city” and “Socrates is a man” was always fanciful, and so the concept was doomed from the get-go. TBL blames specific big players, like Microsoft, for not wanting to adopt semantic web standards, including RDF.
Perhaps modern LLMs, and how they work, has shown that the Semantic Web was not just unrealistic in terms of the amount of work it was expecting people to do behind the scenes to be nice to machines, but also turned out not to be especially valuable in helping machines to ‘think’ in any case.
There are broadly two paradigms guiding artificial intelligence research: semantic, and Markovian. The semantic paradigm was based around the idea that, if we explicitly instruct machines with enough facts and propositions, then eventually the machine will be able to take this nexus of information and reason on its own. The semantic web seems clearly most wedded this this paradigm for artificial intelligence.
The Markovian paradigm, by contrast, just requires streams of text, or other forms of data. The text doesn’t have to say anything rational or sensible - it could exclusively comprise verbatim transcripts of the ramblings of Schizophrenics,, for example - but there does need to be a lot of it. From this paradigm, the intelligence of AI is - depending on perspective - illusory or emergent, and just comes about from the data getting processed ever more effectively and comprehensively to predict the next word, or the next sentence, or the next paragraph.
Of the two AI paradigms, the second definitely sounds a lot dumber. But over the past five years (at most) pretty much all progress in AI has come from researchers, mostly commercial, pursuing this second paradigm. The results have been spectacular, and perhaps indicate we just don’t need the kind of explicit reasoning and logical, rule-bound cogitation that the first paradigm, and the semantic web, appeared to presume were so important.
On the other hand, many LLMs are exquisite confabulators and bullshitters, able to generate vast amounts of prose that claim with apparent confidence things that are utterly untrue, rewriting history to suit the apparent whims and wishes of the user, referencing academic papers making impossible claims that don’t exist, and so on. Although the apparent reasonableness of most LLMs has increased, and continues to each year, the tendency for LLMs to veer off into unreason and mutual insanity when the flow of a conversation takes them that way, seems something inherent to the stream-based-next-token-generating paradigm they sit upon.
Perhaps, if the semantic web had been as popular in the 1990s and 2000s as the original web had been, the first paradigm for AI might have won the race instead, and we would now have AIs that are both terser and more rational?
There is one area in which the value proposition of meeting semantic web standards is relatively clear, however, and that relates to making the web more accessible for persons with, for example, visual impairments. A picture of an apple, for example, would need to be labelled as a picture of an apple in order for a screen reader to tell a user what the images contain. By making the contents of websites more accessible to machines, semantic web standards also help make contents more accessible to ever more people as well. Requirements to make more of the web more accessible have been growing in recent years, especially in the EU, and so perhaps it is through such mandates and requirements that the semantic web can still find fuel to grow.
Idea 6: Household Microcultures: A licence to be uninhibited in eccentric passion
If TBL appears eccentric, it’s likely partly because he is, and also because he was born and raised in a household environment where pursuits, preoccupations and interests that would strike most people as unusual were simply accepted. Both TBL’s parents were computer scientists and mathematicians, working in the nascent British computer industry of the 1950s. And both were highly creative systematisers, who fully supported TBL’s efforts as a child to build his own Heath Robinson-style computing device in the 1970s for the sheer sake of it, despite the risk of electrocution and exploding televisions.
The first thing I built was a switch for my model railway. Then I built an intercom that linked the upper and lower floors of the family house. (‘He was very useful as the engineer around the house’, Mum would later say.) I bought a ‘breadboard’, a simple physical platform for building circuits, and starting chaining together logic gates made from my cast-off transistors. You could make a circuit on the breadboard in minutes, and if it worked, you could solder it up on a printed circuit card to make it permanent. I made a train whistle circuit and some automation for the model trains. [pp. 19-20]
TBL remembers his parents, who lived into their nineties, very fondly, despite occasionally being a victim of the downside of obsessive preoccupation:
Dad was brilliant, but he could be a little absent-minded. One time, he took me to pick up his shirts at the dry-cleaner’s. He got the shirts, but left me behind in my pushchair. Another time, he parked our car by an embankment on the Thames. He returned to find the tide washing over it. When he was travelling back to London from Manchester will colleagues one day, he could not find the return half of his train ticket at the barrier at the station. His colleagues assured the ticket inspector that he had just lost it and was always forgetting things. So he got home, and then my mother asked, ‘Conway, where’s the car?’ [pp. 11-12]
For someone not similarly dispositioned to his parents, accounts like the above would have been grist for a misery memoir. For TBL, such incidents appear as fondly remembered as any others.
Idea 7: Who owns my data double?
At the end of the TRIP Leading interview, Stewart picked up that TBL had become slightly frustrated with the types of question he had been asked (including implicit invitations to express envy and enmity towards billionaires and politicians), and asked (something like) “Is there anything you wish we had asked you?”
Somewhat regretfully, TBL stated (something like): “Well, I do wish you’d have asked more about the SOLID principles I’ve been developing and promoting the last few years:”
Much as the Semantic Web was TBL’s passion project in the late 1990s and 2000s, so in the 2010s and 2020s the question of personal data ownership - cue bono? - appears to have become TBL’s contemporary concern.
And with good reason, I’d suggest. If you’re not paying for a service, so the saying goes, you’re not the customer, you’re the product. And for companies like Facebook/Meta and Google the value of harvesting personal data is clearly estimable in the billions, with dividends returned to shareholders, not the users whose data are harvested. Meanwhile, inefficiencies and disconnects in many public and private services, lack of comprehensive and shared data about individuals, frequently leads to deeply frustrating experiences for those trying to get, for example, good quality healthcare, housing support, and financial services.
The problem, from TBL’s perspective, isn’t necessarily that too much, or too little, data is held about individuals, but that the individuals don’t own their own data. If individuals really did own their own data, they would have a clearer idea about how much such data are worth, and be able to make more informed and engaged decisions about how such data should be shared.
TBL tries, with some success, to articulate his vision as follows:
Within the semantic web community, the idea of a ‘bit of the web of data which you own’ was becoming more and more desirable. So we built containers for our own data, and we called the containers Personal Online Data Stores, or ‘PODS’ [p. 245]
We struggled to think of what to call [the PODS specification]… settling on ‘Social Linked Data’, or Solid… [p. 245]
With Solid we had an ecosystem where your pod would sit like a blank canvas, or an empty Scrabble board, waiting for apps to write facts into it. [p. 245]
The Solid layer would accomplish two things simultaneously. First, it would restore the privacy of the individual on the web, who would no longer have to worry what data was being generated about them, or who was looking at it. Second, it would unlock all manner of new functionality, by connecting data that had previously been stored in separate containers. [p. 247]
For example, there were obviously huge benefits to be realized by attaching your smartwatch data to your medical records. If you were looking for new shows to watch, or new media to consume, you might attach your browsing history to a content recommendation algorithm. If you wanted to get a mortgage, you could attach your spending habits to your application. Maybe a travel agent would even be willing to pay you for access to information about the countries you’d visited and the restaurants you preferred. Of course, all of this was optional; if you were a privacy-oriented person, you would never have to share any of this information, and even if you did, you’d automatically be able to see who had access to anything you did share. [p. 247]
As with the semantic web, such a shift in data ownership standards might seem both highly technical and highly idealistic (two terms at whose intersection is often found TBL) and we might struggle to imagine how we could conceivably get from here - where warring territories of international corporations own and profit from our data in separate proprietary cages - to there - where Solid PODS empower the individual to choose what to share with third parties. It could come about, perhaps, if such a shift to Solid PODS were shown to enrich and empower all stakeholders - individuals, public services, private corporations alike - much as people tend to stick to the same rule of which side of the road to drive on for their benefit as much as others; and almost everyone agrees that some taxation is acceptable to pay for, at least, basic physical infrastructure and national security. Similarly, it could become adopted if a coalition of corporate underdogs - the second, third and forth largest companies, collectively owning 70% of personal data - adopts it to challenge the top dog. Or because a province, then a state, then a country, then a continent, adopts it first for public services, and at each stage the benefits in terms of reduced data friction and improved services are demonstrated.
Idea 8: Apps: The Walled Gardens of the Modern Web
To an extent, phone apps are just little wrappers around pieces of web content. But as TBL came to realise, in practice they’re something much more, and much worse:
I think the mobile experience would have been greatly improved if you didn’t have to download an app every time you wanted to engage with a new service. Interestingly, that’s what Steve Jobs seemed to first showcase with the iPhone.. If you watch his original presentation for the device, from 2007, he repeatedly demonstrates the unlimited functionality of the iPhone’s Safari browser. The app store wouldn’t launch for more than a year.
Why did Jobs change direction? The answer was suggested to me at a clandestine rendezvous I had in the late 2000s with a Google engineer who shall remain unnamed. We met at a restaurant at Half Moon Bay, a popular beach town across the mountains from Silicon Valley. At a quiet table overlooking the ocean, the engineer explained to me that, from what he could see from deep inside Google, Apple was deliberately throttling the functionality of both the Safari and Chrome mobile browsers. Apple, you see, got a 30 per cent commission of the apps it sold; plus, it received continuing income from subscriptions and in-app purchases. It was much more profitable for Apple to direct the developers to build an app than to build a mobile website. Apple got a cut from the former, and nothing from the latter.
I considered what he was saying as I stared out into the Pacific. He was right, of course …[p. 226]
TBL then concludes:
The Apple app store was an example of exactly the kind of for-profit gatekeeping I’d always hoped the web could navigate around. But the iPhone was extraordinarily popular - I owned one myself. The toll booths the web had managed to avoid on the PC had found their way into the mobile realm.
Quite.
Idea 9: Even ‘Nice Guy’ Millionaires ‘Trade Up’
My last observation may strike the reader as somewhat facetious or ad hominem, because it is. But it was something that struck me, as ‘interesting’ at least, while reading the book.
Part way through, I lost count of the number of wives TBL has had, and became somewhat bemused by how casually he discussed the ending of each relationship, as well as non-professional relationships in general. I am fairly the answer is three. But each coupling and uncoupling - except the last - is mentioned with such parenthetic insouciance as to give each status change a kind of ‘blink-and-you-miss-it’ quality.
On TBL’s first wife, whom he met at Oxford:
When my time at CERN was over, I saved the Enquire program to an 8-inch floppy disk, then entrusted it to Brian Carpenter. I included with the disk a five-page written explanation of what it was, and what it did. It wasn’t the last time I would see Brian Carpenter, but it was the last time I would see the disk. Our contracts were up and we returned to Poole. Sometime after, Jane and I decided to end our marriage and go our separate ways. [p. 40]
So, even decades later, TBL still appears bereft… about losing the disk containing a search program he’d written called enquire-within.
On meeting TBL’s second wife:
While living in Switzerland, I met Nancy Carlson, an American working as an analyst at the World Health Organisation headquarters in Geneva. We moved in together and, a few years later, we married. Nancy worked with computers, too, and at home we shared a 286 Toshiba laptop. The personal computing revolution of the 1980s was well underway, and it was no longer necessary to build your own computer. I left my beloved homebrew computer behind in Poole; and in the mad rush to leave for CERN, I ended up throwing it out. (Kevin, the photographer of the group, took some photos of it.) In my apartment in Switzerland, I was content with a PC clone. [p. 44]
So, in the paragraph describing his meeting and marrying his second wife, TBL does use the term ‘beloved’… to describe the homebrew computer he discarded in Poole.
Writing about attending a trip to Finland to receive the Innaugural Millenium Technology Prize, TBL writes:
The Finnish visit was one of the last official trips I took with Nancy. Our marriage in fact came to an end over the next few years, and I was hopeful for new horizons in my work and personal life [p. 192]
Two children, almost two decades of marriage, but time for Tim to move on. With whom is conveyed through some possibly inadvertent foreshadowing, in which Tim devotes more than a single sentence to a woman he knows (other than his mother):
I first met Rosemary Leith at a dinner in 2008. The hosts were my friends Nigel Shadbolt and Wendy Hall, two professors at the University of Southampton. Nigel, Wendy and I were all on the board of the Web Science Trust, a charitable trust which supports interdisciplinary research into the effects the web has on society at large. They were courting Rosemary to join our board.
Rosemary was a highly accomplished businesswoman with an enormous gift in her ability to connect with people and understand how they think. She holds others and herself to a very high standard. She had grown up in Canada, then moved to the UK via Switzerland in the late 1980s to work for the private equity firm Pallas. In the late 1990s, she co-founded an early content company called Flametree, a dot-com that took advantage of web technology to provide flexible work-life solutions. Rosemary and her co-founder were very far ahead of their time; both had relied on flex-work in order to succeed in business while simultaneously raising children. Flametree was a success, and in 2001 they sold it to PwC just three years after inception. Rosemary then worked as a venture investor, using her ability to recognize early shifts in society that impacted technology and to see their resulting commercial impact.
I found in Rosemary a strong-willed, highly intelligent woman. She had dark hair, dark eyes and a winning smile. I liked her right away, and I began inventing reasons to see her again. I had a TED talk comping up and, although I didn’t know her very well, I asked her to help me write it. I believe she was a bit skeptical of me at first - but why don’t I let her tell this part of the story? [pp. 204-205]
There then follows another page in which Tim quotes Rosemary lavishing praise on Tim.
Then, in the world’s most predictable plot development:
Years later, in 2014, when I asked Rosemary to marry me, the Queen gave us permission to use the Chapel Royal, dating from Henry VIII’s time, and the adjoining St James’s Palace, as our venue. [p. 228]
Many years ago, I remember listening to an episode of Radio 4’s public statistics programme, More or Less, discussing divorce rates within straight (or ‘mixed sex’) couples, as well as amongst gay men, and gay women. The key finding surprised me enough to have been burned in my consciousness ever since.
The key finding was this: Lesbian divorce rates are the substantially higher than divorce rates in mixed-sex couples, which are substantially higher than divorce rates amongst Gay Men. The researcher who crunched the numbers was interviewed, and matter-of-factly stated something like, “this was an expected result, as in mixed-sex couples it is more likely to be the female partner who initiates divorce proceedings, so in a sense we have in a Lesbian marriage a ‘double dose’ of partners with an increased proclivity to initiate divorce proceedings.” This finding is perhaps not we might expect if we assume divorce occurs mainly through something like a personality-mismatch pathway, in which inherent differences between male-typical and female-typical personality profiles would be expected to be greatest in mixed-sex couples, leading to the greatest level of conflict leading to divorce.
Now, though I think this higher-female-propensity-to-divorce tendency likely is true amongst most marriages, I think there may be an exception to this ‘rule’: amongst ‘high status’ (and rising status) males, I suspect it may be the male partner of a mixed-sex marriage who becomes more likely to divorce or separate (or at least to become increasingly inattentive, leading to the female partner to initiate divorce or separation proceedings). Why? Because someone who over the course of a marriage or other long-term relationship gains substantially in status and success also gains in partner options, i.e. the number of other people who may seek or be receptive to advances from ‘high status’ individuals. And, given polygyny (the marriage of one to many) is near-universally proscribed within modern societies, the caricatured response to these growing status-based opportunities for partners is for high status men (especially, but I suspect some women too) to ‘trade up’ through serial monogomy, multiple marriages, instead.
So, my suspicion is TBL’s multiple marriages may well fit this cliched pattern amongst ‘successful high status men’. So far, so disappointingly familiar. Three somewhat redeeming aspects to TBL’s own version of this pattern, however: firstly, that he never seems to have anything negative or critical to say about his previous wives (if in part because, at least in this autobiography, he has little to say about them at all); secondly, that he appears immensely proud and devoted to his children from previous marriages; and thirdly, that those traits he sought to ‘trade up’ in in his partners were those of professional success and ambition rather than youthfulness (see DiCaprio’s Immortal Girlfriend, for example), physical attractiveness, and subservience.
Summary
So, there’s nine somewhat-random-and-scattershot ideas prompted by TBL’s book, and the podcast by which I became aware of it. Near the start I mentioned that I’m somewhat jaundiced by works of industrialist hagiography, hence not taking to Bregman’s Moral Ambition, and there are certainly aspects of This is for Everyone that could fit within this mould. Like Helen Lewis, I also rate highly Steven Johnson’s arguments about the adjacent possible, the concept that new ideas have their time and place, and it can often be little more than accident who takes credit for their discovery, as an antidote to excess beliefs that, but for a scattering of great individuals throughout history, we would still be banging rocks together. But at the same time, in this reality at least, I think there are clear, obdurate, and positive ways in which the specific form of the digitally connected world in which we all live - such as those related to interoperability and net neutrality - continue to be shaped by the flavour of interconnectivity brought by TBL’s vision in the late 1980s, rather than some - perhaps more profit-minded - also-ran who would have stepped into the adjacent possible if, for example, TBL’s bosses at CERN were not as forgiving of his idiosyncracies and reveries as they turned out to be.
Footnotes
The tl;dr: like many breathless works of hagiography, Moral Ambition struck me as monotone, valourising a personality profile that, in certain lights, sure looks pathological. Whereas ‘great man’ accounts running from Galton’s 19th century proto-eugenics work Hereditary Genius, and through to ‘titans of industry’ books of the 20th century, and into ‘cyberindustrialist’ books of the 21st century, valourise cognitively gifted monomanaics’ pursuit of wealth and growth in the private sector, Bregman’s Moral Ambition seems to employ exactly the same formula in valourising the same gifted-obsessive profile in the public and third sector.↩︎