The engine of serendipity

This post, along with seventy-eight others, is collected in the book Utopia Is Creepy.

We are given the words we need when we need them. “Serendipity” snuck into the language just 250 years ago, in 1754, when Horace Walpole, the novelist, coined the word in a letter he sent to an acquaintance named Horace Mann. Walpole was inspired by a fairy tale called “The Three Princes of Serendip,” which tells the story of a group of traveling royals who, as Walpole puts it, “were always making discoveries … of things which they were not in quest of.” It took a long time, though, for the word to make the leap into common usage, as Robert Merton and Elinor Barber explain in their book The Travels and Adventures of Serendipity. They could find only 135 instances of the word appearing in print in the 200 years after Walpole coined it. But from 1960 on, its use exploded. As Richard Boyle writes in a review of Merton and Barber’s book,

Serendipity appears in the titles of 57 books between 1958 and 2000 … Furthermore, the word was used in newspapers 13,000 times during the 1990s and in 636,000 documents on the World Wide Web in 2001. The English-speaking world has gone overboard for the word.

Apparently, something happened to us 50 or so years ago that gave us a keener appreciation of – or need for – the serendipitous. Maybe it had something to do with the prosperous calm we were granted after two calamitous world wars. Or maybe it was an offshoot of the let-the-sunshine-in sensibility of the beats and their hippie offspring.

No sooner, though, did the word become popular than its debasement began. It drifted quickly toward vagueness and then, as Merton writes in an afterword to his book, vacuousness:

For many, it appears, the very sound of serendipity rather more than its metaphorical etymology takes hold so that at the extreme it is taken to mean little more than a Disney-like expression of pleasure, good feeling, joy, or happiness. For those who have consulted dictionaries for the word, its typical appearance between serenade and serene may bring a sense of tranquility and unruffled repose. In any case, no longer a niche-word filling a semantic gap, the vogue word became a vague word.

Boyle brings the story up to date:

So it is that in 1992 the word serendipity was emblazoned on the cover of a catalogue for women’s underwear without further explanation. That in 1999 a review of the autobiography of Sir Alec Guinness drew attention to the actor’s “serendipitous writing style (sly, witty, elegant).” That in 2001 the following was to be seen on the Internet: “Serendipity: When love feels like magic you call it destiny. When destiny has a sense of humour you call it serendipity.” And that in 2002, again on the Internet, we find “Serendipity Airedales, home of the top winning Best in Show Airedale in the history of the breed.”

And then, in the spring of 2006, Eric Schmidt preannounces a new Google service called Serendipity, which “tells you what to type.” Soon after hearing Schmidt’s speech, I serendipitously stumbled onto a blog post on the subject of – you guessed it – serendipity by Steven Johnson, author of the popular book Everything Bad Is Good for You (we are also given the books we need when we need them). Johnson was annoyed by a sentimental little op-ed in the St. Petersburg Times by a journalism professor named William McKeen, who argues that serendipity has become an “endangered joy.” Writes McKeen:

We have become such a directed people. We can target what we want, thanks to the Internet. Put a couple of key words into a search engine and you find – with an irritating hit or miss here and there – exactly what you’re looking for. It’s efficient, but dull … It’s all about time. So many inventions save us time – whether it’s looking for information, shopping for clothes or checking what’s on television. Time is saved, but quality is lost. When you know what you want – or think you do – you lose the adventure of discovery, of finding something for yourself.

Nonsense, replies Johnson:

I find these arguments completely infuriating. Do these people actually use the web? I find vastly more weird, unplanned stuff online than I ever did browsing the stacks as a grad student … Thanks to the connective nature of hypertext, and the blogosphere’s exploratory hunger for finding new stuff, the web is the greatest serendipity engine in the history of culture … It’s no accident that BoingBoing is the most popular blog online – it’s popular because it’s an incredible randomizer, sending you off on all these crazy and unpredictable paths.

No, says Alan Jacobs, in a comment on Johnson’s post, McKeen’s right and Johnson’s wrong:

My particular situation is that of a scholar, and I think Steven is – what’s the technical term? – nuts to think that I now have more serendipity than I did before. When I used to rely on print dictionaries or encyclopedias, I would very often forget what I was looking for because, in thumbing the pages, I would stumble across all sorts of interesting words or topics, which would lead me to look up other interesting words or topics, along the way to which I would be distracted by yet other words or topics that I had never seen before.

A fellow named Nick (no relation) also begs to differ:

I disagree, Steven. The kind of surfing you describe is still narrow, even if it feels broad. You can’t just say “I read lots of sites: Boingboing, Metafilter, Waxy, and Kottke!” The article said it best: you’re finding exactly what you’re looking for. BoingBoing’s interesting-link bounty, diverse as it might be, cannot be called “serendipity”. When you pointed your browser there you knew what you were gonna get: links to Wonderful Things. It’s a carefully curated randomness, as selected by Cory And The Gang. You don’t expect them to start posting things that lie outside their purview, like sports scores, financial analysis, or Oprah articles.

Personally, I come down right in the middle: I think Johnson’s absolutely right, and utterly wrong. I do think the web has expanded the sum total of serendipity in the world. I come across a heck of a lot more random stuff today than I did before I went online. So, yeah, the web is probably “the greatest serendipity engine in the history of culture.” Still, though, I find myself agreeing with McKeen when he talks about the lack of “surprise” in internet surfing and, even more so, with Nick when he talks about the web’s “carefully curated randomness.” Once you create an engine – a machine – to produce serendipity, you destroy the essence of serendipity. It becomes something expected rather than something unexpected. Looking for serendipity? Just follow these easy links! Serendipity becomes a game of trivial pursuit. It becomes an end in itself rather than an unanticipated surprise that leads you to some greater understanding. Serendipity should be a doorway, but on the web it’s a kitschy picture stuck on your refrigerator with a magnet.

What we’re seeing, I guess, is the final stage in the process of the debasement of the word “serendipity” and maybe even of the very idea of serendipity. The last act will be the unveiling of Google Serendipity. That will mark the end of serendipity’s brief travels and adventures, the moment when serendipity becomes systematized out of existence.

EPILOGUE: I’ve been testing out Yahoo Publisher Network ads on my RSS feed, and I just noticed that at the end of the feed for this post appears an ad for “Serendipity Bedding.” Who says algorithms don’t have a sense of humor?

UPDATE: Steven Johnson replies.

15 thoughts on “The engine of serendipity

  1. Anonymous

    This brings back fond memories of my favorite search engine feature ever: AltaVista’s random link. These days, I suppose, the chance of it sending you to a link farm would be too high to make it entertaining…

  2. Tim

    In my scholarly and non-scholarly use of the web and the library, I’ve found that serendipity is plentiful in both, but as inverses to one another.

    With internet and electronic searches, you might (at least initially) know what you’re looking for, but the serendipity is found in your sources, which could turn out to be anything, including books, journals, or sites you never would have sought out otherwise. Compared to book searches, the internet is much more source-neutral: a Google search will return hits for books, newspapers, and blogs, without differentiating between them. This can be frustrating, but when you make a discovery, it’s — well, serendipitous.

    In books, libraries, and other “traditional” information technologies, what you discover might more often be serendipitous, but the sources you start with, by necessity, just aren’t — at least to the same degree. Even when you’re browsing bookshelves or library stacks, you’re limited to the material immediately in front of your eyes — not just what’s been published in books, in languages you can readily understand, but what’s been categorized in nearby call numbers, what your library or bookstore has purchased, and what’s immediately available on the day you browse. More often, too, you stick with a single source — one journal, or one author, or one reel of microfilm.

  3. Josh K

    Thanks Nick, for putting your finger on the “creepy” factor of the internet so well. All this choice, and yet it boils down to a very homogeneous choice.

    A maze of twisty mirrors, all alike.

    Rubes like Stephen Johnson think the hyperlink is a shortcut to experience. I think what Dvorak and Orlowski have been banging on about for so long is the confusion between information and knowledge. Simply having a fact (or opinion) available instantly doesn’t mean a whole lot. You can’t do very much with this except generate more “facts”.

    We have an over-abundance of information, but a paucity of real knowledge. I don’t think the net kids will ever understand the difference – God help them.

  4. Soledad
      I agree with the fact the Serendipity should be a doorway and certainly technology can help us to build it. I used Blogger to create a web place aiming to learn, to research and to get in touch with people who want to share their expertise. And who knows… may be one day we would be able to create some knowledge. The place is called Serendipity

      So, I would go along with the fact that some people might be moving from “learned information to learning information. And curiosity will be how you establish your expertise”. In fact, that is more serendipitious than the idea of a service that is going to “tell me what should I be typying”.

      By the way, I have been noticed of “The engine of serendipity” post by a friend whom I met thanks to the Serendipity web place. So serendipity’s word, idea, and concept are alive. There is no final stage outhere :-)

  5. Scott Wilson

    I think Johnson is closer to the truth than his detractors, although the issue may be one of personality and the filter through which we see things than the actual function of the Internet. For one thing, I think anyone who says “…you’re finding exactly what you’re looking for” with a straight face vastly overestimates the accuracy of search and other automated engines; for another, if you’re the sort who only clicks on relevant results, then I suppose that yes, you do find exactly what you’re looking for. But then, I imagine that sort of person would brush right past unusual findings in their library research as well.

    I think it’s a mistake to say the serendipity can be taken out of anything by automating it, either. I guess you can argue the nature of the word is diminished by commonality, but how can you say non-electronic serendipity is any more serendipitous than the Internet variety, when in fact it’s still simply a product of a much larger, more poorly understood system: physics. I guess at that point it gets into philosophical digression, but whatever it is that produces serendipity for Nick in the natural world is just as systematic as Johnson’s BoingBoing (Johnson’s BoingBoing? How’s that for a serendipitous statement?), we just don’t understand it as well.

    And I suppose that’s where I think it’s more about individual perceptions than any concrete definition, because serendipity is ultimately about what Johnson didn’t expect to find, not what you thought Johnson shouldn’t. Who is anyone to tell anyone else what they were thinking at the moment they had that flash we call serendipity? If Johnson finds it on the Internet, sure, call him naive and a rube if you like, but serendipity is fundamentally experiential. Trying to judge it from outside that frame of reference is what is really naive.

  6. Liz Lawley

    Nick, on what do you base the assertion “the word appeared in print just 135 times in the 200 years after Walpole coined it”? To my knowledge, there isn’t yet a full concordance to the world’s published works over the past 250 years.

  7. Nick Carr

    Liz, The 135 number comes from Merton and Barber’s “laborious” scholarly search (as they describe it). Although Walpole coined the word in 1754 in a letter, his letters weren’t published until about 100 years later, so the word didn’t actually begin to appear in print at all until the late 1800s.

    I revised the post’s text slightly to make this clearer.

    Thanks, Nick

  8. David St Lawrence

    There seems to be some confusion as to how discovery is accomplished.

    Serendipity has far more to do with who the observer is than the mechanism that presents the information to be observed.

    The individual who is aware on many levels enjoys a great deal of serendipity in any environment. The dullard is happy to discover that which he is looking for.

    All this munching about on the subject of whether the Internet or dead tree media offer a better chance of enjoying serendipity is a waste of breath. It misses the mark entirely.

    Serendipity occurs when the observer is sufficiently aware to actually observe what he sees and is able to draw conclusions from it.

    The fact that the Internet provides an observer with far more data than the best library makes it obvious that there is more serendipity online for those who are aware.

  9. Ross Dawson

    My blog post as below appears at:

    A topic of great importance – serendipity – has suddenly surfaced in public debate. William McKeen, chairman of the University of Florida journalism department, recently wrote an article in the St Petersburg Times titled The endangered joy of serendipity, suggesting that in an online world we are less likely to stumble across the vital information you aren’t specifically looking for. Steven Johnson, author of among other titles Everything Bad is Good For You, responded with a blog post Can we please kill this meme now, strongly disagreeing that online information is worse for serendipitous discoveries than print, sparking substantial debate on the theme. With the mainstream press commonly taking their stories from discussions in the blogosphere, not surprisingly the BBC took up this issue of the importance of serendipity, with a piece Serendipity casts a very wide net.

    I’ve been speaking about serendipity for some years, and more specifically the concept of “enhanced serendipity”, that is, deliberately making fortuitous and valuable accidents more likely to happen. As part of the debate Nicholas Carr wrote a post expanding on the history of the word serendipity. However he missed out an important detail of the story. As Carr wrote, the word originates from Horace Walpole, who coined it from the story, The Three Princes of Serendip. The three princes, in their adventures, had the faculty of making happy and unexpected discoveries. However these didn’t just happen to them; the princes actually helped to create them. In the following tale, excerpted from a retelling of the The Princes of Serendip by Richard Boyle, the three princes are advisors to the great Emporer Beramo.

    Beramo has fallen in love with a beautiful slave girl called Diliramma, who one day questioned his honour in public. In a fit of rage, he had her bound and abandoned in a forest. The next day, Beramo was filled with remorse and ordered a search for his paramour. No trace of her was found, leaving Beramo ill with sorrow.

    Witnessing the emperor’s suffering, the princes advise him to build seven beautiful palaces and to reside in each one for a week. In addition, the best storyteller in each of the seven most important cities of the empire is to be brought into his royal presence to recount a marvellous story.

    Over the weeks, in his various palaces, Beramo listens with appreciation to six of the stories, his health steadily improving. While listening to the seventh story, about a ruler who spurns his lover, Beramo suddenly realizes that it concerns Diliramma and himself. On being questioned, the storyteller reveals that he knows Diliramma and that he is searching for her lord to tell him that she still loves him despite his act of cruelty. Overjoyed, Beramo sends for Diliramma and they are reunited.

    In this story, the princes have created a strategy for making a happy accident more likely to happen. This is a great example of enhancing serendipity, not just being subject to it. That is what we must seek to do, in creating links between ideas and people that would be enormously valuable if only they were made. So many of the emerging technologies of today, from blogs to collaborative filtering systems such Last.FM, absolutely facilitate happy accidents.

    The debate on the topic is very important. I believe that online search tools are currently at a very early stage of development, and so they are hardly likely to cut us off from accidental discoveries of relevant or interesting information any more than we have been in a print world. However we are moving closer to a time when we will be able to hone in on what we are seeking with great precision. I have previously envisaged a “serendipity dial” which we can situate either to give us great accuracy, or a greater possibility of accidents in our discoveries. I don’t share McKeen’s concerns. Most people are far more diversely informed than they were not long ago, except by choice. The tools we have are not at fault. As we move forward, we need to be highly aware of the degree of serendipity we are choosing. The new world of information gives us that choice.

  10. _oh

    The trouble with people nowadays, is that they expect technology to upgrade on a weekly basis. Whether it is a new feature on their phone, or a different icon for their Hard disk drive, or screen saver or ring tone. People seem to be addicted to this ‘change-crazy’ bit-based culture, of passing fads. What everyone forgets, about books, is that technology was universal and understandable to people all over the globe for many centuries. Compare that to a computer, which needs some upgrade or revision or something, nearly every time you turn it on. I am reminded of Charmlemagne, the great emperor spending a lifetime learning to read. But once he learned to read, that was it – a whole new territory opened up in front of him that was universal. I am also reminded of Doug Englebart’s idea, that people need to spend 6 months or so, learning to drive around the digital world – and like Charlemagne learning to read, suddenly a whole new world opens up in front of you. It is about the intelligence contained within the human being, not the machine. But instead, from the Parc Alto concept of ‘office automation’ onwards, it was dumbed down for the user – a mouse, is a close approximation to a child’s first experience of seeing shapes and colours, of learning to point to things and utter baby words, to convey different definitions of things. This is the place where digital worlds have been stuck for the last couple of decades. All we do is point to new icons and symbols – but the user per se, is not encouraged to become more intelligent.

    Brian O’ Hanlon.

  11. Bhuwan


    This topic interests me a lot.

    One thing which i do not understand here is, why are we beating about a phenomena which by definition is uncertain?Please check my blog regarding this. “Let it happen”. If it has to, it merry will.


  12. Anonymous

    In my humble opinion, the debate needs to draw some distinctions. We can’t accuse the web as a whole of being bad for serendipity; and we need to distinguish between different types of serendipity.

    On the first point: while it is true that the better search engines are the quicker we get to what we want (without having time to glance at the unexpected), search engines are not the extent of the web. Portals, blogs, and even RSS (all forms of aggregates) feeds provide much of the serendipity found outside the web.

    On the second point: it is one thing to stumble upon the unexpected, and another to stumble upon something someone put where we are, next to what we are looking for (if indeed we are looking for something, which is not always the case). Machines can provide the first type of serendipity, which I agree is of a poor kind. But editorially-curated aggregates are also pervasive on the web, and these have much of the richness found in real life.

    For a more elaborate discussion of these issues see my recent post on the subject.

  13. kochikvp

    I recommend the following to anybody who thinks the internet has consigned serendipity to the dustbin: next time you visit any website, just look around at the various links on whichever page you are on. You are certain to see at least one or two links that, while being unrelated to your current search, look like they may offer something interesting enough to check out. Use the “Open link in new window” feature to open the page to which that link leads without distracting you from whatever topic you are currently trying to get information on. Later on, when you have some leisure on your hands, you can go back to these windows lying open on your desktop and read those pages. I do this all the time, and because of this, end up getting information on far more topics from each surfing session than I ever set out to get. Long live Serendipity.

    Nevertheless, the perception that serendipity has lost out in the internet age has, in my opinion, some basis. The reasons have to do with both technology and mindset: 1. Low bandwidth ensures that the cost of “internet digression” such as the above is often prohibitive. 2. Newer web access devices such as handhelds, mobile phones, etc. suffer from limitations that make it difficult to digress in the above manner. 3. Above all, the attention deficit nature of our society ensures that we just do not want to digress as above. In the old days when we used to visit libraries to dig for information, we were just more willing to allow ourselves to digress. And that is the real reason for the perceived loss of serendipity. As with many other developments that we regard as the bane of society, it is not so much that things around us have changed – it is that *we* have changed!

Comments are closed.