Deconstructing Agile Communication

“What threatens is indeed writing” Derrida

In “Agile Software Development”, Cockburn (2000) writes against writing. The heat of the communication channel, he argues, increases as we move away from text towards face-to-face communication. Marks on paper are a poor form of communication, he argues, which are open to misinterpretation and feedback delays. Improving upon this are: email exchange, audio-tape and continuing in order of “richness”; video-tape, phone calls and close physical proximity facilitated by a modelling apparatus.

The primacy of speech is expanded upon with the idea of osmostic communication; meaning that background information (i.e. conversation) is consciously accepted or positively filtered into the unconscious. In open areas, when someone speaks, “traces” are picked up by others which becomes a backdoor transference of knowledge.

A decade ago these ideas were seen as radical, but have now been widely accepted across Agile methodologies and are prevalent in current practice. These ideas are not radical, but have for centuries been an integral and common thread in Western thought.

Derrida (1976) in “Of Grammatology” demonstrates that since Plato, speech has been of central importance whilst writing has been secondary. He calls this bias logocentrism.

“Logos” is the Greek word for speech or reason. To be logocentric is to believe that truth is the word. The Gospel of St. John declares “In the beginning was the Word, And the Word was with God, and the Word was God.” “The Word” becomes a transcendental signified, which is a meaning that lies beyond everything in the whole universe and is thus centred.

The metaphysics of presence is based on the God-Word transcendental signified. Talking is seen as a direct presentation of thought and Cockburn adds that in someone’s physical presence, communication also involves body language and vocal intonation. Writing however is seen as an absence of presence which involves some kind of loss. A document that I write and you read when I am not present might be “mis-read”. So I would only write if you we not present. Speech is favoured over writing.

But isn’t true meaning conveyed when I am speaking in your presence? When I write, is that meaning not distant and untrue? Is writing something that is derived from speech?

Saussure in the “The Course in General Linguistics”, creates a binary opposition between speech and writing and comes down firmly on the side of speech.

Saussure believes language is system of signs. A sign is composed of a signifier and a signified. The signifier can be a word, or sound-image of a sign, whereas the signified is the concept or meaning. A linguistic sign of “cat”, is made up of the sound “c-a-t” and the concept of a “cat”.

Derrida believes that Saussure regards the signified or meaning as being more important than the signifier sound (“c-a-t”). Saussure believes that the sound (“c-a-t”) gives us access to the meaning. That the sound is outer and the meaning is inner. Using Christian referents God-the-Father is the essence while Christ is the mouthpiece.

Saussure argues that speech is a way of representing inner meaning whereas writing is a way of representing speech. If speech is a sign of inner meaning then writing is only a sign of speech and thus only a sign of a sign.

Saussure thus concludes the sounds of speech should be the object of linguistics and not writing.

Derrida then operates a deconstruction reversal on how writing can be seen as central in Saussare’s own text. Saussure on the one hand argues there is a natural bond between the signifier and the signified, but on the other believes that sound and meaning is arbitrary. “Cat” gets its identity from its sound difference with other signifiers. “C-a-t” is slightly different from “m-a-t”, which in turn is slightly different from “m-a-t-e”. A sound then, is what it is, because it differs from other sounds in the same language. Likewise conceptually, the signified has no meaning in-and-of-itself, and only receives meaning through difference. Meaning is thus, necessarily unstable, if it depends on difference.

How can Saussare still privilege speech, argue there’s a natural bond between sound and meaning whilst also arguing that both sound and meaning exist through difference?

Derrida says he can’t.

Instead of speech being more important than writing or vice-versa, Derrida puts both terms under erasure. Erasure is signified by writing the word and marking a black ‘X’ over it. The technique was borrowed from Martin Heidegger, meaning here that both speech and writing are inadequate to describe the more general play of differences common to both. But in this discourse we cannot do without them, so they are put under erasure.

Derrida then goes on to say that the play of difference that constitutes both speaking and writing is derived from what he calls “arche-writing”. Arche-writing is an original form of language that comprises all modes of communication. It is that which allows this free-play.

Cockburn has in fact reproduced that which was already in the cultural osmosis, and justified logocentrism within software development. Stakeholders now have a theoretical reason not to write. And the writing that is done, is of a very poor quality. If writing is so evil, why do we write software?

Logocentrism has become so pervasive that it has affected books spawned by software development. Like Maoist’s in the cultural revolution, Jason Fried and DHH (2010) in “REWORK” shout;

“When you get out of school, you have to unlearn so much of the way they teach you to write there. Some of the misguided lessons you learn in academia:

  • The longer a document is, the more it matters
  • Stiff formal tone is better than conversational
  • Using big words is impressive
  • You need to write a certain number of words or pages to make a point
  • The format matters as much (or more) than the content of what you write”

In poorly written and throwaway prose they put forward, that writing should be conversational and non-precise, and that any kind of formal structure is “drippy”. Would DHH write code in such a flippant manner? Is it really a wonder that specifications are the worst written artefacts in software development?

We need to return to the communication medium of writing as a skilled, and sometimes arduous craft. We should favour neither speech nor writing above each other, or as binary opposites. Both mechanisms should exist equally in our world and each be used correctly in their appropriate contexts.

Advertisement

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: