jeremy,

you hide behind ANT to give us a very impoverished description of the world
in which designers are like billiard balls acting on other billiard balls.
How does ANT help us be in our different worlds?

i read you posts and understand that you don't care for whether you deal
with billiard balls, humans or computers as long as they have an effect.
for you, action is a cause-effect relation between actors and actees.

incidentally i read a good deal in latour's recent reassembling the social
-- pretty unconvincing.

klaus

-----Original Message-----
From: PhD-Design - This list is for discussion of PhD studies and related
research in Design [mailto:[log in to unmask]] On Behalf Of jeremy
hunsinger
Sent: Wednesday, April 22, 2009 10:41 AM
To: [log in to unmask]
Subject: Re: actor-networks Re: Discourse on object level

an action is a relation between two things.  in theory one of those things
can be another relation, so i think you can see where that goes.

i think you will find many many people who will tell you they never wanted
it called actor-network theory, but the name stuck.  sort of like the
volkswagon 'thing' and 'beetle'

personally, i think it is the relations that are of interest, which
means the actions.  other people have other standpoints.   I've made
the argument before that i don't really care about the 'nodes' of the
network at all, i just want to see the relations and have them define the
nodes.  perhaps the node will end up as a person, perhaps the node will end
up as a machine, perhaps it will be a machine that acts like
a person perhaps it will be a person who acts like a machine.   but
again... standpoints vary on what is most important.  what is very important
is being clear about the various principles that guide actor- network
theory, and for that one should probably read latour's recent book.
On Apr 22, 2009, at 10:30 AM, Terence Love wrote:

> Dear Jeremy,
>
> Many thanks again - especially for your patience!
>
> Does this mean that an action always has an actant and an actee?
>
> Or can one have an actant with no actee?
>
> Or is it only the action that is of interst? In that case, would it
> imply
> ant as an 'action network theory', which one might expect to represent
> theories about networks of actions?
>
> Best regards,
>
> Terry
>
> -----Original Message-----
> From: jeremy hunsinger [mailto:[log in to unmask]]
> Sent: Wednesday, 22 April 2009 10:12 PM
> To: [log in to unmask]
> Cc: [log in to unmask]
> Subject: Re: actor-networks Re: Discourse on object level
>
> Usually sequence is part of the narrative of the description of
> events, which is part of the write up.  it is usually handled by the
> methods of analysis you use, such as  ethnography, which keeps track
> of its own time.
> however, you can see how that becomes a problem when you start talking
> about atemporal methods like semiotics, there you probably wouldn't
> use time, because it isn't really part of the method's normal
> analysis.  but in discourse analysis time is also present as one
> tracks the changes
> (actions)
> in the discourse over time.  now here's a trick that i've proposed and
> i blame terry pratchett for the idea, but time... can also be
> accounted for as an actor in actor-network if you plan on doing that.
> i could see how time could become an actor if you were analyzing say
> an emergency room where time acts in all kinds of relations to all
> kinds of things.
> however, for the most part, people don't seem to use time as an actor,
> and they just use the temporal relations common to the method they are
> using.
>
> the problem might be with your construction below (and my prior loose
> speaking)... which represents an actor/action divide.  some actor-
> network take the term actant from semiotics, Griemas i think, but
> maybe propp before.  they use actant to resolve the issue where people
> assume there is an actor without action.  There is no necessary divide
> between actor/ action; the 'actor' does not become apparent until the
> 'action'.  That is to say, that there are actions which are relations,
> and actants acting, but without the action, we have no relations, and
> thus no actor.   Actants are things acting, there is no actor, without
> the action, and thus no temporal divide.
>
> In my prior example below, i posited the existence of an actor without
> acting, the way we would do that would be to have actions ongoing
> (which almost always happens) from that actant.  usually any given
> actant is doing many things in the system and the problem is sorting
> it out.
>
> On Apr 22, 2009, at 9:52 AM, Terence Love wrote:
>
>> Hi Jeremy,
>>
>> Thank you. That is really helpful and clarifies a lot.
>>
>> Wondering how in ant you handle information about sequence and time
>> when an action happens?
>>
>> I can see how you correlate an action to an actor and identify
>> classes of relationship between actors but identifying the sequence
>> of actions and how they relate to identified actors, actor
>> relationships and actions I'm unclear on.
>>
>> Best wishes and thanks,
>>
>> Terry
>>
>> -----Original Message-----
>> From: jeremy hunsinger [mailto:[log in to unmask]]
>> Sent: Wednesday, 22 April 2009 8:52 PM
>> To: [log in to unmask]
>> Cc: [log in to unmask]
>> Subject: Re: actor-networks Re: Discourse on object level
>>
>> reasons and causes are described after the analysis is finished.
>> you look
>> at the system of relations and you can then describe what happened
>> and can infer whatever cause fits the described data.  let's keep in
>> mind that actor-network theory is not a method, it is a standpoint
>> about how to treat research and how to gather that data using
>> methods, such as semiotics, discourse analysis, or ethnography.
>> it's primary use is to mould the data collection and to provide
>> insights into data analysis..   it doesn't assign reasons so much as
>> track actions and relations in networks.  reasons and causes are
>> things to be very skeptical about because frequently we have less
>> than a cause and more of a conjunction or constant conjunction
>> according to hume, so... actor network would note that x did y, but
>> when y then immediately did things it would not note that x caused y,
>> because as you can imagine y may merely have been waiting until time
>> z to act, and action y was incidental.  one can only find out these
>> relationships through time.
>>
>> now after the analysis is over and you have your data and you see
>> that every time x is in proximity of y, y acts somehow, you may be
>> inclined to hypothesize a causal relation, and others over time may
>> support that or deny that.
>>
>> one thing to note here is that mental models, 'reasons' can be
>> 'actors' in actor-network.  a good idea can 'act', recruiting people
>> through people, etc.   latter theories might call this unification of
>> actors a mess or an assemblage.  but it is very useful to be able to
>> track an idea as an actor.
>>
>>
>>
>>
>> On Apr 22, 2009, at 8:23 AM, Terence Love wrote:
>>
>>> Hi Jeremy,
>>>
>>> How do you deal with the reasons and causes for actions in ant?
>>>
>>> Best wishes,
>>>
>>> Terry
>>>
>>
>