Very like a network

Posted by on Jan 29, 2010 in Copying, Science, Web/Tech | 3 Comments

“Hamlet: Do
you see yonder cloud that ’s almost in shape of a camel?

Polonius: By
the mass, and ’t is like a camel indeed.

Methinks it is like a weasel.

Polonius: It
is backed like a weasel.

Hamlet: Or,
like a whale?

Very like a whale.”

Yesterday I had a repeat performance of familiar disagreement with someone who loves the Network Metaphor – someone who sees the big N as the most important lens through which to build understanding how things spread.

Now, what I don't want to do is rehearse the arguments at length again (go here or here or indeed read one leading network scientist's views on the misapplication of the central ideas of network theory to human behaviour) to get your head around the detail. Nor do you need reminding that it's a whole lot more interesting and complex than the influentials models suggest – human life is not largely constructed of hubs and spokes, our connections and relationships are more fluid and flexible and of course it's often not the stickiness or other characteristics of a thing (or meme or whatever) that determines how far that thing spreads: it's the people and how they choose to interact with each other.

Suffice to say our use of the Network idea is really a Metaphor: – a comparison highlighting certain similarities – human social structures and interactions often behave very much like what we call networks elsewhere. e.g. the physical sciences or in electronics

Unfortunately we often mistake the metaphor for identity; the map for the landscape. 

Wouldn't it be better to think about it as a simile (as Billy Shakes does in the quote above). No one imagines that Hamlet and Polonius are saying that the clouds are actually camels, weasels, whales or any other living thing, merely that for a moment, in certain ways, they resemble that kind of thing in certain ways.

"Like a network", indeed.


    February 4, 2010

    It’s interesting that you should post this to Twitter only hours after @theplayethic linked to this excellent discussion on network science between a physicist and a political scientist.
    I’m inclined to say that networks are real; as Fowler says, the atomistic view of human beings is flawed. WE need research methodologies that can take into account the connections between things. Would be interested to know your opinion.

  2. Mark Earls
    February 4, 2010

    Interesting. I missed this as am travelling at the moment. Pat and I were discussing this area recently but this looks a good one.
    FOr what it’s worth, my feeling is the “n” idea is useful but it is also misleading, denoting fixed and unchanging channels down which we/whoever can send stuff.
    I’ll have a think and repost

  3. Mark Earls
    February 7, 2010

    Ok Had a chance now to read the piece and ponder.
    I think my comment above might be at cross-purposes: absolutely agree that social is the proper and fundamental context for understanding, studying and looking to shape human behaviour – trying to do anything out of that context misses the point entirely. Other people and our interactions with them shape so many things in our lives – the good, the bad and the downright amazing. So in that sense, “networks” are very real.
    But the critical pov I was trying to articulate in the main body of the post is this: human social networks are nowhere near as fixed and solid and unchanging as the “n” word suggests. Our connections are multiple, multi-directional and constantly mutating.
    And as the Borgatti piece linked to above suggests, too many of our attempts to understand social contexts through applying “network science” get over-specific – over-focussed on the individual node and nodal outcomes – when the big picture is where much of the juice lies. E.g. in network effects or understanding the underlying rather than the surface network structure…
    Hope that clarifies