The New Is

David Weinberger
Home Blog
Posted: September 12, 2005
Last updated: September 12, 2005

This was originally written as an introduction/overview of a talk I gave at Ars Electronica in Linz, Austria, September 1, 2005. The theme of the conference was "Hybrids: Living in paradox." I've modified this draft very slightly to make it clearer.

Aristotle uttered the most unparadoxical statement imaginable: A thing is what it is and it isn't what it isn't. It's the Law of Identity, and it's holding us back.

For Aristotle and the Greeks in general, knowledge and Being were one and the same: To know what a thing is meant to know what type of thing it is. If your knowledge doesn't accord with reality, then it's not really knowledge. Of the ten different questions one might ask about a thing (which Aristotle called “categories”), he singled out “What is it?” as the most important. And his answer was that to be something – a robin, for example – is to be subsumed with other similar things under a broader category: To be a robin is to be like other birds in being a feathered biped but unlike other birds in the particulars. This genus-species system of similarities and differences is an incredibly powerful way of comprehending large complex systems, such as the universe.

From Aristotle's way of thinking came a history of thought and politics that made certain assumptions: Because knowledge and being are fused, just as there is only one reality, there is only one  structure of knowledge. The best people to put this structure together are experts. Because of the economics of parchment and paper, experts filter what we need to know. They become gatekeepers, priests of knowledge.

The digital age undoes all of these assumptions, changing the nature of knowledge and even of meaning itself. We are entering the age where to understand something is to see how it isn't what it is.

Until now, the structure of knowledge has mirrored the way we've structured the physical world: We take a pile – think of your laundry – and split it into lumps, and then split those lumps into further lumps, until we have piles that are not worth splitting any more. So, we create a library classification system such as the Dewey Decimal System, or a Periodic Table of the Elements, a Tree of Life, or a business organizational chart. But when we're dividing up our laundry, we have to put our socks into one pile or another, but not both (the Law of Identity). Why should the same restriction hold when we're dealing with ideas? Why can't ideas go in many piles? Why can't a single intellectual leaf hang from many branches?

This is precisely what happens in the digital age. If you are trying to decide where to put a digital camera in your physical store, you're going to have to pick one or two areas. If you're listing it on your Web site, you'll put it in as many categories as you can because you want people to be able to find it. Is the digital camera photography equipment, a vacation accessory, a sports item, a featured sale item? The answer is: Yes. And if you can think of other categories in which to list it, you have an economic motive to do so.

This makes a mess of your site's organization. But that's a good thing. In the digital age, messiness is not a sign of disorder. It is a sign of a successful order. Messiness is a virtue.

In fact, a new strategy is emerging that leaves the assigning of what something is to the last moment. For example,  uBio is a tool for biologists that gives every species a unique identifier. To this ID it attaches as many different names as it can find, including scientific names, vernacular names, and names in many languages. Another database contains dozens of the most popular scientific taxonomies of species. uBio makes no decision about which is the right name or the right taxonomy. Instead it lets the particular scientists do that. Yet other scientists can know for a certainty which species are being referred to because they each have that unique ID.  Meaning is no longer attached firmly to being: It often works better to assign meaning after the fact.

This is nowhere more obvious than in the recent movement towards social tagging obvious at sites such as http://del.icio.us and www.flickr.com. Here any individual can assign a tag – a word or two – describing either the Web page address being noted (del.icio.us) or the photo being uploaded (flickr). This helps individuals organize their collections, but more important, anyone can see all the objects tagged with any particular tag. For example, at flickr you might want to see all the photos tagged as “Iraq” and at del.icio.us all the pages tagged as “relativity.” Thus a page that the author may have thought was about, say, a couscous recipe may to others be tagged as a page relevant to the study of the distribution of grains in Northern Africa. (In the tagging world, the author has no privileged position when it comes to saying what her own work is about. See? The French philosophers were right!).

The right way to tag might seem to be to wait for experts to figure out what the most useful set of tags are. Otherwise, you may tag a photo of the guards at Buckingham Palace as “Buckingham,” she might do it as “guards,” and I might do it as “London.” But we know from previous attempts to systematize all knowledge, there is no one right way to do it. So, taggers does precisely what the gatekeepers of knowledge have feared: They go ahead and do it any way that makes sense to them, and figure computers will sort it out afterwards.

Sort it out perfectly? Nope. Taggers are creating a huge mess. We will never find all the photos of London, if only because somewhere someone has tagged a photo as “My Vacation” or “Pretty View.” Did that person make a mistake? Not at all. We're just not aiming for perfection any more. We don't need perfect knowledge in an age of knowledge abundance. We just need pretty good knowledge, and that's something we don't need perfect gatekeepers for.

To the gatekeepers what looks like chaos and the degradation of learning to Netizens looks like an exponential increase in intelligence.

The difference in views occurs in part because the Net explodes the old view of intelligence as the containing of lots of knowledge. This container model is reflected in how we talk about documents: We say they have contents even though print is as 2-dimensional as a shadow. On the Net, documents – pages – get their value to a large degree not from what they contain but from what they point to.  Without links, there is no Web. This is an ecstatic model (to borrow Heidegger's term) rather than a container one.

(This means, by the way, that the Web is built on a spirit of generosity. If every site were as stingy with external links as most commercial sites, there would be no Web. In this way, the Web reflects our better, social nature.)

Links, not containers: A page is what it points to.

Multiple tags, not single meanings: A thing gains more meaning by having multiple local meanings.

Messiness, not clean order: The best definitions are ambiguous.

What are the sources of these breaks with the idea that knowledge is supreme when it sees just one, sharp-edged order?

It is the connectedness of the Net. We can see what the world is thinking. But that just leads to relativism, a form of disappointment.  Instead, the Net is filled with joy. That is why almost a billion people are using it and are finding it transformative. In fact, we are escaping from the old, dissatisfying clash between objectivity (the world as it looks when we're not looking at it) and subjectivity (the world as it matters to us). With the Internet, we get multi-subjectivity for the first time. Take blogs. They look like publications, but they're overwhelmingly conversations. We're linking to one another, disagreeing, amplifying, making fun, extending, sympathizing, laughing. We are talking with one another, thinking out loud across presumptions and continents. If you want to know about an idea, you could go to an encyclopedia and read what an expert says about it. Or you could find a blog that talks about it and start following the web of links. You'll not just see multiple points of view, you'll hear those points of view in conversation. That's new in the world.

The old dream of finding a single knowledge for the entire world – having knowledge be like reality, in other words – is dying rapidly. The connectedness of the Net has made it too clear that the world is not going to come to agreement and be able to write its single encyclopedia, covering everything we need to know without dissent. Cultures and languages are not going to go away. But we should not be left in despair because we now also know that for as long as we manage to not to destroy this blue pearl, we're going to be engaged in endless conversation. Conversations iterate differences upon a common ground. Conversations are themselves paradoxes. But because they happen, they're miracles.

So knowledge has become the continuousness of conversation. It has become a miracle. Knowledge can no longer fix the meaning of a thing with a single pin of meaning. To understand now means to hear the multiplicity of meaning talked about across the world.

The more of the world we get into the conversation, the more the world will mean.