Tacit vs. explicit knowledge (take II)

["Hart, Geoff" <Geoff-H -at- MTL -dot- FERIC -dot- CA>]

Mark Baker continued the debate: <<1. Tacit knowledge: the inexpressible
knowledge that informs performance and which cannot be learned by theory but
only by practice. (The existence of which you may or may not believe in.)>>

I've never heard a definition of tacit knowledge that includes the word
"inexpressible", and defining something this way is a tautology and thus not
particularly useful. Pretty much all knowledge can be "expressed", whether
in words, pictures, or by means of hands-on demonstrations. One thing the
literature on knowledge management systems shows is that tacit knowledge is
_rarely_ inexpressible; most often, it just requires someone to ask the
right questions ("why are you doing that?", "can you show me that again so I
can see what you're doing?").

It's far too mystical for my taste to postulate knowledge that can only be
experienced, not explained or communicated. Granted, some knowledge can be
particularly difficult to pin down and communicate, but that's a far
different issue. And it's why people hire us: we're good as figuring out how
to solve this kind of difficult problem.

<<2. The unexpressed knowledge of how to do design which is capable of being
expressed. Capturing and expressing this knowledge is the ambition of
knowledge engineers. Their aim is to assist it training new experts and/or
the programming of computers to replace experts.>>

This is exactly the kind of knowledge that is eminently transferrable,
assuming the student can be trained. Not all can; I'll never sink a 50-foot
jump shot no matter how many times a star basketball player shows me. I
think you're unnecessarily narrowing your definitions by restricting this to
knowledge engineers; it's a component of pretty much any form of
apprenticeship (e.g., carpentry), not just something for theoreticians.

<<3. The created knowledge that is the result of actually designing
something. This knowledge is not tacit in either of the senses above. It has
been expressed in specifications, designs, and product. Technical writers
take this knowledge and use it to create the fourth kind.>>

Good writers also use this information to establish context, such as
recommending that you not use a laptop at -40C or stating that good backups
are necessary because (e.g., with handheld field computers) the environment
inherently jeopardizes the data. Much of this information can indeed inform
the creation of procedural knowledge, but it can also do so much more if you
use it to improve your understanding of context.

<<4. Procedural knowledge. How to actually use a specific widget to perform
a specific task.>>

I think we agree that this is the easiest to make explicit, and is primarily
what we do in our day jobs. But much of what we do goes beyond this into
understanding how users think, providing a framing context in which they'll
perform tasks, and so on.

<<The origin of this debate concerned the value of a PhD program for
training technical writers and it turns on the distinction between 1 and 2
above. You and Jim have both confused the issue by bringing types 3 and 4
into the debate.>>

Point 3 is clearly the sort of knowledge that an experienced writer can
teach to someone else. "I've built this and it works. Here's why it works,
and here's how you can apply that understanding elsewhere." A PhD program
can teach you exactly this kind of theory (things that other designers of
information have found to work well). An experienced writer can also learn
this from years spent writing.

The sensible middle course lies in understanding that you can learn some
things faster from those who already do them well than from being forced to
learn it on your own. That's the advantage of book learning: it saves you
having to reinvent the wheel. You can achieve exactly the same goal
(possibly more effectively) through an apprenticeship or mentoring program,
but that option isn't open to most of us.

<<The question is, are writing skills *primarily* developed by the actual
practice of reading and writing, or by the study of communication theory.>>

I'd argue that you're making an artificial distinction. We certainly agree
that you learn skills by actually using them, not by reading books. But
where the theory comes in is suggesting which skills you should use, and
teaching you better ways to use those skills.

<<If you presume that type 1. tacit knowledge does not exist and that we
have actually mastered and can successfully teach all the relevant type 2
knowledge, then you would presumable opt for the study of communication
theory.>>

See above. That's precisely what I presume. But that doesn't cause me to
reject your next assertion either, because doing so would be overly
simplistic:

<<If you accept the wisdom or the ages, the advice of just about every
writer who has ever written on the subject, and the manifest evidence of
actual performance, you will opt for the practice of reading and writing.
(And spend you education dollars studying the arts or sciences.)>>

The majority isn't always right, nor are they universally right, nor are
they always working as efficiently as they could be. "We've always done it
that way" often reflects an evolved best practice, but also often reflects
inertia and an unwillingness to try something newer and possibly better.
Moreover, by framing the statement this way you convert this into a zero-sum
game in which one practice must come at the expense of the other. That's not
the case. Study theory to learn better ways to do things, but don't stop
there: practice that theory until you're good at it.

--Geoff Hart, geoff-h -at- mtl -dot- feric -dot- ca
(try ghart -at- videotron -dot- ca if you get no response)
Forest Engineering Research Institute of Canada
580 boul. St-Jean
Pointe-Claire, Que., H9R 3J9 Canada 

"Wisdom is one of the few things that look bigger the further away it
is."--Terry Pratchett