Plato and The Internet PDF

POSTMODERN ENCOUNTERS
Plato and
the Internet
Kieron O'Hara
Series editor: Richard Appignanesi
ICON B O O K S UK
T O T E M B O O K S USA
Published in the UK in 2002 Published in the USA in 2002
by Icon Books Ltd., Grange Road, by Totem Books
Duxford, Cambridge CB2 4QF Inquiries to: Icon Books Ltd.,
E-mail: info@iconbooks.co.uk Grange Road, Duxford,
www.iconbooks.co.uk Cambridge CB2 4QF, UK
Sold in the UK, Europe, South Africa Distributed to the trade in the USA by
and Asia by Faber and Faber Ltd., National Book Network Inc.,
3 Queen Square, London WC1N 3AU 4720 Boston Way, Lanham,
or their agents Maryland 20706
Distributed in the UK, Europe, Distributed in Canada by

South Africa and Asia by Penguin Books Canada,
Macmillan Distribution Ltd., 10 Alcorn Avenue, Suite 300,
Houndmills, Basingstoke RG21 6XS Toronto, Ontario M4V 3B2
Published in Australia in 2002

by Allen & Unwin Pty. Ltd.,
83 Alexander Street,
Crows Nest, NSW 2065
Text copyright © 2002 Kieron O'Hara

The author has asserted his moral rights.
Series editor: Richard Appignanesi
No part of this book may be reproduced in any form, or by any
means, without prior permission in writing from the publisher.
ISBN 1 84046 346 5
Typesetting by Wayzgoose
Printed and bound in the UK by
Cox & Wyman Ltd., Reading
The Knowledge Economy
We are now living, we are often told, in a know-
ledge economy. In the past, the key to wealth was
a plentiful supply of labour and control of natural
resources. Nowadays, though, developments in
technology - particularly computer and tele-
communications technology - have resulted in
undreamt-of ease in transferring data around the
world. A scientific paper written yesterday in
New York can be running off the printers today
in researchers' offices in San Francisco, London,
Sydney, Tokyo or Beijing. Scientists can collabor-
ate without ever meeting face to face. Desktop
computers can alert their operators to the latest
news items, stock prices or cricket scores as they
change.
In this new world, the important source of
competitive advantage for a developed economy
is no longer raw materials or labour, but know-
ledge: bright ideas, smart designs, clever organi-
sation. It can't compete in brawn, so it must use
brain. The new economy demands knowledge
workers who create value and wealth, not with
traditional money capital, but with intellectual
PLATO AND THE I N T E R N E T
capital.1 Indeed, many hold out the hope that the

exploitation of knowledge could be a key to
enriching the developing world too, in that know-
ledge capital is easier to find and cheaper to get
hold of than money capital, and the World Bank's
Global Development Network has been set up to
explore that possibility.2
All this is independent of the dotcom boom and
bust. Knowledge is big business in any industry.
A huge amount is invested annually in knowledge
via research and development programmes, staff
training, and so on; but because knowledge is
intangible, it can't be measured or accounted for.
As consultant Alan Burton-Jones puts it, 'know-
ledge is transforming the nature of production
and thus work, jobs, the firm, the market, and
every aspect of economic activity. Yet knowledge
is currently a poorly understood and thus under-
valued economic resource.'3
What is Knowledge? The Discipline

of Epistemology
So it makes sense, in this new world, to get to
know about knowledge itself: what it is, how you
THE D I S C I P L I N E OF E P I S T E M O L O G Y
spot it, how you make sure you have enough of

it. It turns out that there are people who investi-
gate these issues - the study of knowledge is a
branch of philosophy called epistemology.
Epistemologists investigate such questions as:
What is knowledge? Are there different types of
knowledge? Are there procedures that you can
follow to make sure you acquire knowledge?
How do you know that something is knowledge?
How do you discover whether you are mistaken?
Can you know what you don't know?
Epistemology is not the oldest branch of philo-
sophy, but damn near. The first philosopher
whose epistemological work has come down to
us in any quantity is Plato (c. 428-c. 347 BC). Such
was the power of his influence on later philo-
sophers that much epistemology since has contin-
ued along more or less the same lines. Indeed,
even in the present day, epistemologists are
wrestling with problems (and providing answers)
that bear a strong resemblance to Plato's.
In this twenty-first-century context, there are
many dilemmas that Plato would recognise. For
instance, how do we sort out the knowledge from
PLATO AND THE INTERNET
the dross on the World Wide Web, the true from

the false? Even so, can it be right that epistemo-
logy should still operate from essentially Ancient
Greek assumptions? Does the emergence of
massive and thoroughgoing technological change
make no difference to the epistemological land-
scape? Or alternatively, do we need to look
afresh at the phenomenon of knowledge in the
Internet age?
This is the question I will examine in this
book. I will sketch the traditional philosophical
arguments, to give a rough idea of the general
Platonic position that still holds sway today.
Then I'll talk about the technological, commer-
cial and political context which has made the
study of knowledge suddenly so pressing. To
close, I'll bring the ideas of traditional epistemo-
logy together with the new technology, to see
how far, if at all, the discipline has to change.
Scepticism
In order to understand the historical roots of
modern epistemology, let us begin by asking why
Plato studied knowledge in the first place. Plato's
SCEPTICISM
opponents in his writings on knowledge were

philosophers who were sceptical about know-
ledge, who didn't believe, for some reason, that it
was possible to gain knowledge at all.
Such sceptics4 included the Sophists of the fifth
century BC, fiercely critical philosophers who
tried to explain the universe in terms of the way it
appears, rather than by making up theories about
what underlies its appearance. Unfortunately, by
attracting the ire not only of Plato but Aristotle
as well, they garnered a reputation as bloody-
minded quibblers more concerned with showing
their verbal dexterity than finding the truth, keen
to show wrong right and right wrong. They, no
doubt, saw themselves as anti-theorists, under-
mining spurious attempts to divine the occult
phenomena that supposedly lay behind experi-
ence. They were much more interested in the use
of argument for practical purposes (e.g., in the
law-courts, or in the lively Athenian democracy),
rather than for 'pure' metaphysical speculation.
Be that as it may, the Sophists' attack on
theories, and on the practical deployment of
argument, made them very potent and worthy of
P L A T O A N D THE I N T E R N E T
Plato's opposition. There is a lot of power in a

theory - think of the status of science and scien-
tists today. In general, sceptics do not try to argue
on the theorists' terms (like one scientist arguing
that another's experiment was poorly designed in
some way), but instead try to suggest that the
whole theory itself is unfounded. Nowadays, a
space-age Sophist would probably say that if you
download a page from the World Wide Web, you
might as well take it at face value because there's
no way you can reliably find out whether it's true
or false. It's an anti-authoritarian gesture; it says
that all these expensive scientists and journalists
and commentators have no special access to
truth. And it has a surface plausibility, in that it
is very difficult to ground a theory securely on
unimpeachable foundations. If you keep asking,
'but how do you know that?', eventually you will
get the testy answer, 'I just do, that's all!' And
when the expensive establishment thinkers get it
wrong - in Plato's day, with the decline of
Athenian power; in recent times, with the BSE
crisis - there is further circumstantial evidence
against rooted knowledge/power structures.
THE B I R T H OF E P I S T E M O L O G Y
In general, scepticism is found attractive as a

doctrine by thinkers who worry about glib answers
to deep questions, and as a rhetorical device (a
stiff test) by those who would hope their theories
stand up to the most rigorous scrutiny. Ancient
sceptics included Pyrrho, Sextus Empiricus and
Cicero; in more modern times scepticism has
been used by thinkers as diverse as Montaigne
and Descartes, Hume and Hayek. Little wonder
that Plato, theorist extraordinaire, deeply opposed
the sceptical strain of thought and devoted such a
large amount of time to trying to refute it.
Plato and Knowledge: The Birth of

Epistemology
Even though we have much of Plato's writing,
and even though Plato is a very lucid writer, the
actual substance of his thought is not as clear
as it might be.5 For one thing, rather than pro-
ducing work in the standard form of a treatise or
monograph, in which the author sets forward
his own view and defends it against counter-
argument, Plato wrote dialogues. Such dialogues
were generally led by Socrates, for whom Plato
PLATO AND THE I N T E R N E T
had great affection,6 and included other promi-

nent philosophers, Sophists and public figures of
the day (typically, the dialogue would be named
after one of the participants - Theaetetus, Meno,
Gorgias, etc.). Plato - who does not appear -
usually, but not always, can be taken as endors-
ing the views expressed by Socrates. But the dia-
logues are rarely conclusive, and at the end the
participants often agree to differ. In the Parmenides,
Socrates puts forward arguments that we can
confidently attribute to Plato, yet is overwhelmed
by Parmenides' criticisms, and the reader is left in
little doubt that Parmenides is the moral victor in
that one.
Plato's philosophy was not an ivory-tower pur-
suit without application to the real world. His
home city of Athens had recently suffered a terri-
ble reverse in war, amid scandal, corruption and
military and naval incompetence. He believed
that a properly run city needed a trained class of
rulers. Athenian boys were taught how to wrestle
or swim; why could they riot be taught the virtues
appropriate for a good leader? This is a riddle
which has always intrigued humanity; indeed, the
10
THE BIRTH OF E P I S T E M O L O G Y
English public-school system was one attempt at

an answer.
As Plato observed, there are great disputes over
the nature of virtue. But if we don't know what
virtue is - and these disputes imply that we don't
- how can we teach it? Many of the early dia-
logues consist in Socrates demonstrating that his
interlocutors do not really know what virtue is,
and so the question of knowledge - what know-
ledge of virtue, or anything else for that matter,
consists in - loomed large in Plato's work. His
ideas changed somewhat over his career, and I
don't want to go into the particular details of his
epistemological views in this book. But the main
point is that he drew an interesting and com-
pelling distinction between knowledge and true
belief.
SOCRATES: Do you think that knowing and

believing are the same, or is there a difference
between knowledge and belief?
GORGIAS: / should say that there is a difference.
SOCRATES: Quite right; and you can prove it
like this. If you were asked whether there are
11
such things as true and false beliefs, you would

say that there are, no doubt.
GORGIAS: Yes.
SOCRATES: But are there such things as true or
false knowledge?
GORGIAS: Certainly not.
SOCRATES: Then knowledge and belief are
clearly not the same thing.
(Gorgias, 454cd)7
True beliefs are useful, and will not lead you

astray. But they are not reliable, in that you could
never be sure about them.
SOCRATES: True opinions are a fine thing, and

do all sorts of good so long as they stay in their
place; but they will not stay long. They run away
from a man's mind, so they are not worth much
until you tether them by working out the reason
[why they are true].
(Meno, 97e-98a)
Plato spent much effort trying to establish the

essential difference between knowledge and true
12
THE B I R T H O F E P I S T E M O L O G Y
belief, particularly in his great epistemological

work, the Theaetetus. In that work, Socrates and
Theaetetus spend a large part of the middle
section comparing knowledge and true belief
(187b-201c). The close of the dialogue discusses
a theory that knowledge is true belief plus some-
thing called a logos (201c-210b).
The nature of this logos has proved trouble-
some for philosophers to interpret8 - it seems to
be a sort of explanation or rational account - and
anyway Plato does not seem enthused by the
idea. The dialogue comes out against the 'know-
ledge = belief + logos' theory. Nevertheless, these
discussions and suggestions have been very
influential in later philosophy. Knowledge has a
reliability that true belief doesn't; to have confi-
dence in your true beliefs, you must give some
sort of justification of them. This justification
process, whatever it may be, is what turns a belief
into knowledge; it makes it reliable.
Let us call this analysis of knowledge as 'justi-
fied true belief the 'JTB analysis'.
13
PLATO A N D T H E I N T E R N E T
Other Types of Knowledge

But is all knowledge like this? Using the idea of a
belief that happens to be true to point up the
essential reliability of knowledge is a nice rhetori-
cal trick, but it doesn't follow that all knowledge
is of this form. For surely, some types of know-
ledge don't really contrast with true belief at all.
Here are two examples that, at least at first sight,
seem to be somewhat different in essence.
• Know-how. In Plato's distinction, the differ-

ence between knowledge and true belief is that
the knowledge can be brought to account. But
where know-how is concerned, the contrast is
different; the difference between the man who,
for example, knows how to drive a car and the
man who merely believes he can is that the for-
mer can drive a car. If the latter can drive a car
as well, then surely he knows how to drive a
car; he may believe (and for some reason not
know) that he can drive a car, but that is a dif-
ferent question. It is not clear, at first sight,
that driving a car is a matter of belief at all.
• Bodies of knowledge. In common idiom, an
14
OTHER TYPES OF KNOWLEDGE
encyclopaedia contains knowledge. If there is

an agreed interpretation of the words in the
book, the interpretation should entail that the
words express things that are true and justifi-
able. But the encyclopaedia has no psycho-
logical states at all, and doesn't believe or
disbelieve its contents. Beliefs or other psycho-
logical states need not come into it.
Why does Plato ignore these types of knowledge?

The simple response is that he was chiefly con-
cerned with establishing a distinction between
true belief and knowledge, because his diagnosis
of the ills of Athens had led him to worry that
people were relying complacently on their beliefs
about virtue and not seeking knowledge. Those
whose beliefs about virtue happened to be true
could teach virtue to the next generation; true
beliefs don't lead one astray. However, how could
it be known whose beliefs were true and whose
false? Socrates showed that many of the most
confident thinkers were mistaken. For Plato, it all
came down to beliefs; other kinds of knowledge,
if such there be, weren't the point.
15
There are more complex reasons, too. For

example, with regard to know-how, Plato does
discuss expertise in a number of places (e.g.,
Theaetetus 146ce), but he seems to have wanted a
definition of knowledge as a single, unified phen-
omenon, of which different types of expertise
would be species. This made it very difficult for
him to do full justice to the heterogeneous nature
of knowledge.
With regard to bodies of knowledge, Plato was
very chary about written-down ideas. He believed
that writing was a degenerate form of communi-
cation, and this no doubt influenced his adoption
of the dialogue form. In the Phaedrus, Socrates
says:
Writing shares a strange feature with painting.

The offspring of painting stand there as if they
were alive, but if anyone asks them anything,
they are solemnly silent. The same is true of writ-
ten words. You'd think they were speaking as if
they had some understanding, but if you question
anything that has been said because you want to
learn more, it gives just the same message over
16
OTHER TYPES OF K N O W L E D G E
and over. Once it has been written down, every

discourse rolls about everywhere, reaching just as
much those with understanding as those who
have no business with it, and it does not know to
whom it should speak and to whom not. And
when it is faulted and attacked unfairly, it always
needs its father's support; alone, it cannot defend
itself or come to its own support.
(Phaedrus, 275d)
The written is infinitely less interesting to Plato

than live discussion. But nowadays technology
has blurred the distinction. Some speech - as
recorded on TV or radio - is as unchanging as a
piece of text. One cannot interact with a record-
ing; it will not change if one debates with it (it
'cannot come to its own support'). One can, of
course, debate with the person who made the
recording, but the recording will still stand. Like
a piece of writing, a recorded utterance will
remain in existence even if it has been exposed as
a lie, or if its author has changed his mind. Like
writing, it can be broadcast to a wide and undis-
criminating audience (it 'rolls about everywhere').
17
None of these things seems particularly contro-

versial to us, but to Plato they were the mark of a
degenerate use of language that traded integrity
for power. On the other hand, much written lan-
guage today has speech-like features that Plato
would have welcomed. E-mails and text messages
allow a certain immediacy of interaction, while
dynamically constructed Webpages present infor-
mation to readers that is customised to their
requirements and won't be circulated further.
Of course, there was no reason why Plato
should be expected to anticipate such develop-
ments. As far as he was concerned, written com-
munication is dead; its author may change his
opinions, or the words may be misinterpreted
with no way of putting the reader right. But it
seems an odd view to hold in the present day, and
therefore shouldn't count as a reason why a
modern epistemologist should be opposed to
treating bodies of knowledge as knowledge.9
Justified True Belief

I don't want to go into a detailed analysis of
Plato's epistemological views (see Further Reading).
18
JUSTIFIED TRUE BELIEF
The point I want to make is that already, in the

first few years of the discipline of epistemology,
its general features were in place. In the red cor-
ner, a sceptic who denies the possibility of know-
ledge; in the blue corner, JTB - an idea of
knowledge as true belief supplemented by some
sort of justification or holding to account. Very
few well-known figures in epistemology since
have attempted to break the hold of this vision of
the debate over knowledge, even though the
vision itself was the particular product of Plato's
political reaction to a domestic reversal in a
Greek city-state nearly two and a half millennia
ago. I'll refer to this tradition in epistemology as
the 'JTB tradition'.
I will include within the JTB tradition a number
of philosophers who do not endorse as such the
idea of knowledge as equivalent to justified true
belief (as we have noted, this group includes Plato
himself - see the Theaetetus, 201c-210d), but
who follow Plato's general account that knowl-
edge is true belief plus something, the 'something'
turning 'mere' belief into knowledge. The JTB
tradition is therefore firm that knowledge is a
19
psychological state, which involves the person

with knowledge assenting to or endorsing a
proposition.
There is huge dispute amongst philosophers
about what a proposition actually is, but for our
purposes we can ignore this issue, because all are
agreed that a proposition is expressed by a declar-
ative sentence (i.e., one that says something about
the world, not a question, command etc.). So pro-
positional knowledge is knowledge that some-
thing or other is the case. And this does look very
like a belief - an attitude towards a proposition.
For example, Sir A.J. Ayer devotes a whole
chapter of his theory of knowledge to examining
various different types, only to decide in the end -
on the basis of what appears to be not very much
- that knowledge is JTB after all.10 Susan Haack
moves straight into an impressive examination of
the justification of beliefs (though in fairness it
should be said that this is an interesting philo-
sophical issue in its own right).11
Away from the mainstream, the American
philosopher Fred Dretske tries to avoid the JTB
definition, and instead analyses knowledge in
20
JUSTIFIED TRUE BELIEF
terms of the information that caused the belief.12

This is potentially a helpful move (as we shall see
later on) but in this particular formulation doesn't
get us very far. It is still the case that (a) some-
thing is known only when it is believed, (b) some-
one only knows something when they believe it,
(c) what is known is true, and (d) it has a justifi-
cation (in terms of information theory). In an
important new work, Timothy Williamson breaks
radically with the J I B characterisation, but stays
within the JTB tradition by setting out from the
assumption that knowing is a state of mind, and
that knowing something in general entails believ-
ing it.13 Even Wittgenstein continues to think pri-
marily of beliefs, and sees his chief opponent as
an epistemological sceptic.14
Nicholas Everitt and Alec Fisher (see Further
Reading) produce a thorough survey of the efforts
of philosophers to define the 'something extra',15
but are forced to conclude that there is still, after
two and a half millennia, no final consensus:
[W]e have learned that there is universal agree-

ment that knowledge requires truth. Secondly,
21
there is an almost equally widespread agreement

that knowledge requires belief. The disagreement
arises on what more is required for knowledge
than true belief.16
Surely another option open to us is to reject the

analysis as a whole and step outside the JTB
tradition. This is a radical step in the face of
universal agreement. Why make it? Is there any
reason for us to do it?
The Sceptic Bites Back

First of all, the animus of the JTB tradition has
usually been focused on the sceptic. As we have
seen, Plato was at least in part responding to scep-
tics, but scepticism can be much more thorough-
going than that of Plato's opponents, Sophists
such as Protagoras or quasi-Sophists like Gorgias.
The earliest, really dedicated sceptic in intellect-
ual history was Arcesilaus (c. 315-c. 240 BC),
the first man who was positively in favour of
withdrawing belief - and once you start question-
ing absolutely everything, it is not difficult to
win arguments by destroying your opponent's
22
THE S C E P T I C B I T E S B A C K
position without advancing any positive thesis

yourself.
At the basic level, everyone has wondered how
they know they are not currently dreaming. Rene
Descartes (1596-1650) famously tried to build
philosophy on unassailable foundations by con-
ducting a fictitious argument with a very bloody-
minded sceptic who doubted everything; the
attempt, sadly, foundered on the fact that most
observers agree that the sceptic rather won the
day. Indeed - I blush - even I once managed to
prove, on the basis of some fairly innocuous
premises, that everybody was dreaming all of the
time!17
It made sense for Plato to aim his arguments at
sceptics, as they were prominent in Athenian
political life at the time. It is less so for modern-
day epistemologists, for no one occupies the scep-
tical position, except within academia. There are
other purposes for epistemology, and other
people who are interested in the arguments.
Given the changing intellectual background,
surely it would be better to reject scepticism as a
type of philosophical disease (with Wittgenstein -
23
PLATO A N D T H E INTERNET
see Note 14), and focus epistemological efforts

on the areas where they can do some good and
make a difference in the world.
Information Overload
When Plato was writing, scepticism was a serious
political problem. Nowadays it is not. But his
genius and lucidity have combined to cast a spell
over the JTB tradition, and the same issues that
preoccupied Plato are still being investigated,
even though conditions have changed. Techno-
logical, economic and political developments
that Plato could never have foreseen mean that
epistemological theory needs to deal with different
phenomena, and many epistemological assump-
tions could usefully be amended. 'The nature of
knowledge cannot survive unchanged within this
context of general transformation.'18
The new world to which epistemology needs to
be applied is characterised by a dramatic increase
in the quantity of data that are being created and
stored, as storage capacity has multiplied and
cost diminished. When Plato was writing, the
space required to store a work written on a scroll
24
INFORMATION OVERLOAD
was large, the cost of reproducing it huge. Bound

books reduced that space, and, once printing
allowed the production of millions of volumes,
libraries could hold them. Now, a floppy disk
will hold a decent-sized book, a CD-ROM even
more, and a laptop computer still more. More
books are being written, and more periodicals
published. More photographs are being taken,
X-rays shot, videos made.
And the cost of storing all this material has
shrunk rapidly. The new British Library in
London, designed to hold 12 million volumes,
cost £500 million. But magnetic storage is a
different matter. A byte is the amount of storage
required for a single character, so a book is 1
million bytes, or a megabyte. A gigabyte is the
standard measure of data, a thousand megabytes
(i.e., a thousand books). And the cost of storing a
gigabyte of data magnetically was not much
more than £5 in late 2000, and will be well under
£1 by 2005, when the virtual equivalent of the
British Library will cost a mere £12,000.
The opportunities this new technology has
brought are already being exploited. To begin
25
with, very little of the new material recorded in

the world appears in print. As a consequence, the
production of data has been 'democratised', in
that most new stuff is created by individuals for a
small audience: for example, office documents
(80 per cent of all new paper documents), photo-
graphs (95 per cent of film documents) and cam-
corder tapes (20 per cent of magnetic tape
storage). And it is the magnetically stored docu-
ments that are increasing in number; the use of
paper and film for original content is remaining
more or less constant.
Hence, whereas even 20 years ago the amount
of data accessed (and created) by the ordinary
citizen was relatively small, nowadays anyone
living in a developed economy, with com-
puters, electricity and reliable telephone or wire-
less connections, can get hold of colossal quanti-
ties of material, and can author a pretty huge
amount as well. In all, as estimated by Hal
Varian, humankind creates round about 1
exabyte of data annually. That is 1 billion giga-
bytes, or 1,000,000,000,000 books.
Which is not too far off 200 books' worth of
26
THE INTERNET A N D T H E WORLD WIDE WEB
data for every man, woman and child on the

planet! Every single year!19
This is not a world with which we, as yet, feel
comfortable. We can all remember when people
were starved of information; now we are drown-
ing in it. If people had little or no access to the
data they needed in the past, now we all know
the feeling of having to trawl through pages and
pages, documents and documents, to find just the
exact thing we need. Having more data has not
solved all our information requirements. It has
merely given us information overload.
Will it be possible to deal more efficiently and
intelligently with all these data? There are
grounds for saying that it will. To see why, let's
take a look at one of the influential technologies
of the information revolution, the Internet.
The Internet and the World Wide Web

We all know the outer forms of the Internet:
e-mails, the Web, newsgroups etc. We know we
can contact our friends, do our banking through
it, buy books or CDs. Some people say it will
change our lives; others that it will make no
27
difference at all. It is extraordinary, though,

how few commentators are actually aware of the
technology, and how prosaic it is.
The Internet is simply a network of computer
networks, connected using the Internet Protocol
(IP). IP enables a computer to take a file, break
it up into slices called packets, and then send
the packets to a destination (another computer)
down a phone line using dynamic routing (i.e.,
making up routes to the destination, via various
intermediaries, while the packets are in transit).
Dynamic routing was a Cold War idea - if the
intermediate computers were knocked out by
enemy action, another route could be created. In
practice, dynamic routing is useful in peacetime
as it can steer packets around more standard
equipment failures. The Transmission Control
Protocol (TCP) tracks the packets, and the com-
bination of protocols - the Internet standard - is
called TCP/IP.
The public face of the Internet, most responsi-
ble for the Internet's take-off, is that fraction of
the net called the World Wide Web (WWW or the
Web). This is, in effect, a vast collection of files in
28
THE I N T E R N E T AND THE W O R L D W I D E WEB
multiple media - text, pictures, sounds, video,

and so on. The Web works by storing the files on
many different computers, called servers. It 'pre-
tends' that the files on it are stored at a single
source, by having a consistent system of addresses
for the files, called Uniform Resource Locators
(URL), which is what you type in the address box
of your browser (the software that gives you a
view of the Web, such as Internet Explorer, or Net-
scape Navigator). For example, the URL of Icon
Books' Website is http://www.iconbooks.co.uk.
Your computer gets at the file held at that URL
by using (usually) the Hypertext Transfer Protocol,
which is what the 'http' stands for. The use of
HTTP makes it possible to access files stored any-
where in the world as if they were all held on a
single giant host computer.
What makes the Web possible is its universal
language, the Hypertext Markup Language
(HTML), which structures a file so that your
computer can lay out its data in the browser
window, while also allowing it to contain
links to other files - these are the hyperlinks
generally highlighted in blue in a piece of text.
29
Clicking on such links allows your computer to

get at another file at another URL. Well-designed
hyperlinks let you navigate through the Web
looking at only the particular files you are inter-
ested in, unlike a standard text or video, where
the structure is linear and you cannot easily hop
about as you wish. HTML is an international
standard, not owned by a particular company,
and administered by the World Wide Web Con-
sortium (W3C), which helps the Web to be open
both to readers and writers.
The openness of HTML allows pretty well any-
one to create content and post it on the Web. The
result has been an explosion of data. There are
about 2.5 billion fixed documents on the Web,
and when you add in all the databases that users
can access through forms to create customised
Webpages (e.g., the files that allow you to log on
to your bank and see your own bank statements),
that figure rises to 550 billion documents.
This is a huge amount of material (7.5 million
gigabytes of data = 7,500,000,000 books). And
this is just the Web; the Internet is far more than
that! Needless to say, it is an unprecedented
30
THE I N T E R N E T A N D THE W O R L D W I D E W E B
amount of information for people to have access

to instantaneously. And it is this that has made
the Internet the engine for a colossal shift in
human capabilities.
Some commentators have refused to see the
Internet as a significant enough change in tech-
nology to make it philosophically interesting;
others have worried that the Internet will only
bring problems. It is certainly important not to
overstate the case for the defence, particularly in
a world where half the population has never
made a telephone call.20 However, the sudden
explosion of available content could be as signifi-
cant to human development as the invention of
movable type. As we have seen in this section, the
actual technological developments aren't earth-
shattering; a few transfer protocols here, a
markup language there, and a system for manag-
ing changes and addresses. But, as with printing,
a number of small changes, taken cumulatively,
could have major effects.21
31
Knowledge, Technology and

Organisations
Many of the Internet's small changes will be cen-
tral to the knowledge economy. They will enable
organisations to disperse knowledge across their
members (employees, representatives, etc). This
is essential - if knowledge is the key to competi-
tive advantage, as we saw at the beginning of this
book, then it has to be used well and efficiently
by organisations.
Can an organisation possess knowledge? Does
it make sense to say that? Recall that the JTB
tradition in epistemology says that knowing is a
psychological state, that it is something done
only by people.
Well, there is certainly one uncontroversial way
in which an organisation can be said to possess
knowledge (possibly metaphorically), and that is
if it employs someone who unambiguously has
that knowledge. Let us imagine a fictitious firm,
Worldwide Manufacturing; suppose it makes a
particular type of widget, type X, and it employs
Smith, an expert in X, in its London office. Then
we can say that Worldwide Manufacturing con-
32
KNOWLEDGE, TECHNOLOGY, ORGANISATIONS
tains or possesses knowledge about X, in this

case in the person of Smith. If Smith goes,
Worldwide Manufacturing would no longer con-
tain that knowledge. This is pretty straight-
forward at this stage - nothing about the
situation contradicts the JTB theory.
Now consider a second stage. WM executives
realise that Smith is a scarce resource - he goes on
holiday, he sleeps at night, he has coffee breaks,
all of which mean that his expertise is not always
available. Furthermore, he is based in London,
whereas WM is a multinational; if something
goes wrong with their X-factory in Sao Paolo, it
will at best be difficult to employ Smith's exper-
tise, with time differences, logistical problems of
getting the knowledge from London to Sao Paolo
in time, and a language problem - Smith speaks
no Portuguese.
As a result, WM undertakes a knowledge
acquisition exercise. This involves specialists
interviewing Smith and creating a system that
apes his expertise in X. This could be a computer
system, but we will suppose it is a set of inter-
linked Webpages designed to mimic Smith's
33
problem-solving behaviour. If something goes

wrong on the X-assembly line, the manager can
look up the appropriate page, which will tell him
or her what to do. When there are different possi-
bilities, there will be a choice of links to take the
user to the appropriate page. The way the user
navigates through the set of Webpages will track
the way that Smith diagnoses problems in X and
sorts them out. These Webpages can then be
posted on WM's corporate Intranet (i.e., a pri-
vate section of the Internet, accessible only with a
password), and so - via an automatic translation
system - are immediately accessible to the non-
experts in Sao Paolo and elsewhere who need to
know about X. They will also be accessible to
people in London who might need Smith's exper-
tise after office hours or during his holidays.
After this second stage, we can still say that
WM contains Smith's knowledge. Smith works
for WM, as at the first stage. It is just that WM is
more efficient about sending Smith's knowledge
around the firm.
But now consider a third stage, when Smith
retires, or is poached by WM's rivals. WM does
34
K N O W L E D G E , TECHNOLOGY, O R G A N I S A T I O N S
not have to stop making X-type widgets; neither

does it have to employ a new expert. Smith's
expertise is still available to engineers 24 hours a
day, seven days a week, as long as they have a
laptop and a phone line. The Intranet is still run-
ning the Webpages about X-type widgets, and
their navigation structure causes the user to go
through the same problem-solving process that
Smith would have gone through.
Now, earlier on, we used the phrase, 'WM
contains knowledge about X', loosely, meta-
phorically, knowing there was an easy rational-
isation of it, since Smith was an employee of WM.
Now, however, we cannot be so loose, because
Smith has left. No person within the organisation
has Smith's problem-solving capability. The people
who drew up the Webpages were most likely
independent consultants who have long since
gone; let us further suppose that nothing has ever
gone wrong with an X-type widget, and that
therefore no one in WM has ever even read the
Webpages. No one in WM has any justified true
beliefs about problems with X-type widgets.
And yet, it is very intuitive to say that WM still
35
contains, or possesses the knowledge about, X. If

there is any difficulty with an X-type widget in
Brazil or England or Russia or India or Nigeria,
the WM engineer on the spot will be able to find
out from WM's corporate Intranet exactly what
to do. The knowledge available for the engineer
is so much more than he or she would get from a
book about X, in that the engineer, together with
the links that organise the Intranet, will solve the
problem. The engineer need not read all the
Webpages; instead it is the organisation of the
Website that provides the expertise that previously
Smith used to provide.
It looks like we would have to say that any
epistemology that is to be applicable in this
context should allow for organisations as well
as people possessing knowledge in a non-
metaphorical way. It would certainly be possible
for an alternative analysis to be developed, or the
JTB tradition defended (see the final section); but
epistemology will be that much more valuable in
the knowledge economy if epistemologists take
its concerns seriously, and don't try to analyse
them away. As Lyotard puts it:
36
KNOW L E D G E , T E C H N O L O G Y , O R G A N I S A T I O N S
We may thus expect a thorough exteriorization

of knowledge with respect to the 'knower,' at
whatever point he or she may occupy in the
knowledge process. The old principle that the
acquisition of knowledge is indissociable from
the training of minds, or even of individuals, is
becoming obsolete ... The relationship of the
suppliers and users of knowledge to the know-
ledge they supply and use is now tending ...to
assume the form already taken by the relation-
ship of commodity producers and consumers to
the commodities they produce and consume -
that is, the form of value.22
And indeed, we shouldn't end this section with-

out noting that in the social sciences, treating an
organisation as an agent in its own right, greater
than the sum of the people who make up that
organisation, with its own goals, methods and
values, would not be unusual. The JTB tradition
would need an update to retain consistency with
such thinking.23
37
Managing Knowledge
The knowledge possessed by an organisation is
an asset of that organisation. I mean by this that
the knowledge is a claim on future benefits. The
organisation invests money in the acquisition of a
knowledge asset (e.g., by beginning a programme
of research and development) in the hope that
this investment will pay dividends by increasing
future effectiveness (or profits, in the case of a
firm).
Like any asset, the knowledge an organisation
possesses needs to be managed properly in order
to realise the anticipated benefits. This is where
the insights of epistemology will be of enormous
value, as an understanding of the properties and
dynamics of knowledge is essential to knowing
how best to use knowledge for an organisation's
general benefit.
An example: on 23 August 2001, the famous
pottery company Royal Doulton announced a
half-yearly loss of £9 million and placed its fact-
ories on a four-day week; its shares had fallen to
a fifth of their 1998 value. One major factor in its
decline was that it had failed to spot trends in the
38
MANAGING KNOWLEDGE
ceramics market. For instance, over the last few

years, British diners have abandoned the big
family dinner in favour of TV dinners, meals out,
meals al fresco, and generally more casual dining.
However, Doulton had continued to produce sets
of matching china, rather than individual pieces
available at supermarkets and other convenient
outlets. It did not know what its customers wanted;
its half-yearly report admitted to 'weaknesses in
its information systems'.
What is going on? Doulton employs about
3,000 people in Stoke-on-Trent, almost all of
whom will have been well aware of - indeed part
of - the trend away from set-piece meals. Poor
management of that knowledge, possessed by
nearly all of its employees, meant that the
changes in dining patterns were never taken
account of by management. As a result, the com-
pany made losses, and people's jobs are at risk.24
What sorts of issue are involved in managing
knowledge? A recent UK scientific research pro-
ject, Advanced Knowledge Technologies (AKT),
has characterised knowledge management as six
challenges. These are as follows - and to show
39
that they are not merely management issues, I

have annotated them with meaty and classic
epistemological questions, given in italics.
• Acquisition. The initial problem with know-

ledge is when and how to acquire it. We can
distinguish three levels of acquisition. First,
there is knowledge that currently does not
exist and would have to be discovered by a
programme of research (e.g., a cure for AIDS).
Second, there is knowledge that exists, but is
not possessed by the organisation. Third, there
is knowledge that is possessed by the organisa-
tion, but in the wrong form (e.g., in our
example from the previous section, Smith's
knowledge of X-type widgets was held incon-
veniently in his head, and a decision to acquire
the knowledge from him and place it in a set of
Webpages was taken). The management deci-
sions here depend on the difficulty of acquir-
ing the knowledge (does a suitable code
exist?), the costs of the.acquisition, and the
projected benefits. How can we know what we
don't know? How can we know what we need
40
MANAGING KNOWLEDGE
to know? How can different types of know-

ledge be justified?
• Modelling. Once acquired, knowledge has to

be stored in a useful way. The representation
must be such that it is easy to write the know-
ledge down as it comes in, and equally easy to
read the knowledge once it is acquired. How
should different types of knowledge best be
represented? How far does changing a repre-
sentation change the knowledge?
• Retrieval. When an organisation has a large

repository of knowledge, it must be possible to
get at the knowledge quickly enough for it to
be of use. This means that the repository must
be structured to allow a quick, reliable and
efficient search of its contents. If the know-
ledge disappears in a pile of other stuff - like
the Ark of the Covenant in the final scene of
Raiders of the Lost Ark - it is as good as for-
gotten. How best to organise connected pieces
of knowledge? How do they relate to each
other?
41
PLATOANDTHE INTERNET
• Re-use. If some knowledge is present in a firm,

people who need that knowledge need to
know how to get hold of it, so they don't go
about re-acquiring knowledge expensively (or
incompetently, like Royal Doulton). How do
we recognise knowledge that we need and
don't have? How do pieces of knowledge
relate to each other?
• Publishing. It is essential that the people who

need knowledge get it at the right time. Too
soon, and it contributes to their information
overload; too late, and ... well, it's too late.
They also need it in the right form. Someone
who needs one little piece of technical infor-
mation does not need either a whole set of
explanations for beginners or a highly rarefied
piece of theory. Does changing the way know-
ledge is visualised change the knowledge?
How do we know what we need to know?
• Maintenance. Having developed a repository

of knowledge, it has to be kept running and up
to date. This involves a series of disparate
issues, such as: verifying that the knowledge is
42
MANAGING KNOWLEDGE
correct, and that the representation correctly

represents the knowledge; updating the
knowledge to keep track of changes in the
domain; 'forgetting' knowledge that is mis-
leading or out of date; altering formats as the
organisation's needs change; maybe even pack-
aging and selling the knowledge if it would be
valuable for outsiders. How do we know that
an account of the world is correct? How do we
know that a sentence correctly represents a
proposition? What is the logical relation
between statements, and how can the effects of
the removal of a statement be tracked? How
can different types of knowledge be justified?25
These challenges all involve epistemological

problems, many of which have long been the
object of study in traditional epistemology,
others of which are new, still others old problems
with newly acquired significance. But this is a
context where the application of epistemology
would be extremely timely. The problems of
scepticism, which have driven so mucli epistemo-
logical research, are very much less pressing.
43
Data, Information and Knowledge

Is this to say that everything that appears on the
Internet, or in corporate disk space, is to be called
knowledge? Is every item in a symbolic code
knowledge - even those items that are unreadable
or useless?
Of course, that is not the idea. What we want
to sort out is which items are worthy of epistemo-
logical investigation. We will focus on computer-
based items, although the same distinctions that
we are about to draw will apply to paper-based
materials, and even to thoughts and beliefs.
Let us begin by examining the basic content in a
computer - a set of symbols; we can call this data.
Data can be any old garbage, and so may be useful
or may not. A subclass of data we can call informa-
tion, which is those data that have an interpreta-
tion, or which make sense. Of course, the
attribution of meaning to something is a deep and
complex philosophical problem in itself, but let's
assume that we can draw some sort of a distinction
between meaningless stuff and meaningful stuff. In
the case of Webpages, meaningful stuff can of
course include pictures and video, as well as text.
44
DATA, INFORMATION AND K N O W L E D G E
How might information contrast with know-

ledge? Let's try to suggest a distinction by the use
of an example.
For several years now, supermarkets have
recorded the details of each purchase made in
them by using electronic tills, each of which
updates a central database. As a result, super-
markets with nationwide chains of branches have
enormous masses of data about billions and bil-
lions of purchases. These data are clearly infor-
mation, as the big databanks will have some
interpretation which enables the reader to work
out which purchase is referred to, what was pur-
chased, how much it cost, etc.
Is this vast quantity of information useful in
itself? Hardly. No human reader could scan the
printouts in a lifetime, even if he wanted to, still
less draw any conclusions from it. If such stuff is
knowledge, then there doesn't seem to be much
point in having it.
However, one can imagine converting such
data-sets into knowledge (this would be a know-
ledge acquisition process, in the terms of the pre-
vious section). Using special computer programs
45
called machine learning programs, or induction

programs, a computer can trawl through enor-
mous data-sets and extract statistical relation-
ships between some of the data. These relationships
can then be published as sets of rules, which
might look something like:
• People in Hertfordshire spend more per visit to

the supermarket than people in Berkshire.
• People in Inner London who buy beer are

more likely to buy crisps simultaneously.
• People in the South-West have become keener

on branded tinned food over the last five years.
What I would like to suggest now is that these

probabilistic inductions are not only information
but knowledge. They are knowledge because they
are usable, because they can be translated into
effective action. They are not, let's face it, terribly
exciting, but they will help a supermarket save
money on deliveries, warehousing and marketing
by sending goods where they are most likely to be
46
DATA, INFORMATION AND K N O W L E D G E
sold quickly. It is not an exact science, by any

means, but there is some payoff for this sort of
activity, and it would not be possible without the
collection of the giant data-sets through the elec-
tronic tills.
Note further that none of this need ever appear
as anyone's belief. The content of the data-sets is
obviously collected independently of any human
operator. The machine learning program will
extract the knowledge without consulting any-
one. And the knowledge thus extracted might
easily be translated into managerial actions
(e.g., changes in orders of wholesale goods)
automatically, without any human managerial
interference.
And it is knowledge of this sort that organisa-
tions such as supermarkets have to manage. There
is an investment in the collection of those data and
in running the hardware and software that makes
it possible, and organisations need to know
whether the investment is worth it. The reader can
satisfy himself or herself that the six challenges of
knowledge management will apply here. Dealing
effectively with bodies of knowledge of this sort
47
will make a big difference to the profits of

commercial firms and to the success of non-
commercial organisations.
So, let's recap our characterisation. Data are
sets of symbols, while information is meaningful
data. Knowledge is that information which is
usable for the purposes of supporting or suggest-
ing action; it is a stepping stone to some end.
Data, information, knowledge: they sit as a
pyramid. Perhaps, with tongue only slightly in
cheek, one might even imagine wisdom sitting at
the apex, characterised as the ability to select
appropriate important goals for one's useful
knowledge to be applied to.
Scepticism is out of place here. The super-
market of our example wants help with the six
challenges of knowledge management. A proof
that, say, the external world exists and that the
five senses can deliver knowledge, reassuring and
philosophically interesting though this may be, is
not going to be of any help in deciding how many
tins of beans to send to Wiltshire. The manager
who doubted the existence of Wiltshire would be
given a long rest, not a proof.
48
THE S E M A N T I C W E B
The Semantic Web

This characterisation of knowledge - as usable
information - has been generally adopted by
those researching knowledge management in the
Internet age. By creating technologies that help in
the handling of information, more information
becomes usable, and therefore by our criteria,
turns from information to knowledge. Hence
technologists, such as W3C, or the AKT project
mentioned above, who have taken the 'usable
information' characterisation on board, can pro-
duce real systems and tools that actively help
organisations and individuals become more effec-
tive by controlling the information they have
access to. Information overload is alleviated, not
by reducing the amount of information available,
but by increasing the amount of knowledge.
W3C has developed a language for writing
Webpages to supersede HTML, a language called
XML (extensible Markup Language - actually a
language for representing other languages, but
let's steer clear of details!). XML allows designers
to place 'tags' in Webpages that are invisible to
the reader (i.e., they don't appear in the browser
49
when the Webpage is being read), but which can

be read by a computer. These tags can then 'tell'
the user's computer what the number or the word
is about. Contrast this with HTML, which sim-
ply tells the computer where to place the data on-
screen.
A browser that is XML-enabled (i.e., can cope
with XML) will be able to search not just for key
words on the page but also taking into account
their underlying meaning, as represented by the
tags. So, whereas currently a search engine will
come up with, say, 10,000 hits, with XML it
might come up with only 20 hits, but they will be
the 20 you want, because they will be tagged with
the subject area of your interest. If your interest is
in Kate Bush, then typing in 'Bush' together with
some specification of your musical requirements,
the engine will ignore all the George Bushes,
Shepherd's Bushes, rose bushes, African bushes,
etc. The new search will be much more sensitive
to the context of the query.
That in itself would give the user of the Internet
much more control of the information received.
As it stands, XML is a comparative infant, but
50
THE S E M A N T I C W E B
telling computers what Webpages are about is an

idea that will surely be exploited in a large num-
ber of currently undreamt-of ways. The interpre-
tation of Web-based text and images, currently a
human responsibility, could at least partly be done
automatically. This will allow projects like AKT
to write software that can provide information
(e.g., flight information) and world-altering ser-
vices (e.g., ticket-buying agents), ushering in
context-sensitive knowledge services - ways of
providing people and organisations with the
technology to solve their particular knowledge
problems as they happen, and as they need
solutions.
Most of these services would ordinarily be
done by a person - the difference is not that new
work gets done, but that it gets done quickly,
automatically, in a standard way, and exhaustively
(i.e., an agent can search the entire Web for the
best price for an air ticket, in a fraction of a sec-
ond). Information overload would be alleviated
by the reduction of routine tasks, and the stan-
dardisation of non-routine tasks.
This conception of the World Wide Web
51
augmented by XML is called the Semantic Web,

and is based entirely on the new characterisation
of knowledge. Without the recognition of the
epistemological importance of usable informa-
tion, such a conception might never have arisen.
It certainly has nothing to do with JTB. Many of
these developments will seem small-scale and
technical, but they are powerful and cumulatively
important, and they would not have come about
without the change in approach to the problem
of knowledge.
Knowledge: Justified?
So, we have established a rival conception of
knowledge, and furthermore one that has been
the driver of important developments in techno-
logy. What is the relationship between the 'usable
information' conception of knowledge and the
JTB tradition?
One thing that has often been remarked upon
in the history of epistemology is that the nature
of the justification of a belief sufficient to con-
vert it to knowledge (or the 'something extra'
required in the stead of justification) is highly
52
KNOWLEDGE: JUSTIFIED?
problematic. Even leaving aside some fiendish

thought experiments developed by Edmund
Gettier in the 1960s that cast doubt on any rea-
sonable notion of justification as being sufficient
- experiments which no one has satisfactorily
explained away - there has never been any con-
sensus about what justification would do.26
Even under the 'usable information' characteri-
sation, some sort of justification of the knowl-
edge's efficacy would be required. The quality of
a stock-control system will affect a firm's profits
and its employees' jobs; the quality of a heart-
monitoring system will affect a person's chances
of life. In such sensitive contexts, the knowledge
that has been put into the computer had better be
reliable.
But there is a difference in the role of the
justification under the two characterisations. In
the JTB tradition, the justification is central - a
true belief is knowledge if and only if the justifi-
cation of the belief is adequate. Philosophers have
suggested a number of possible types of justifica-
tion. Examples include: the knowledge can be
derived from some privileged, deep foundational
53
knowledge from which all knowledge must

spring; the knowledge was discovered using reli-
able techniques; the knowledge is part of a coher-
ent picture of the world. There is no consensus
about which, if any, is correct.
Under the 'usable information' view, on the
other hand, a justification is a kind of certificate
that the knowledge could be the basis for action
of some type. It would be the proof that the infor-
mation being discussed was usable. However, the
existence of the proof would be required, not to
turn the information into knowledge, but to
demonstrate that the information was usable and
reliable. Often, of course, usability requires a
demonstration of reliability - as with computers
in sensitive contexts. But the existence of a just-
ification, or proof, or anything extra over and
above the usefulness of the information, is not
conceptually bound up with the identification of
information with knowledge.
Interestingly, Plato discusses the possibility of
such a pragmatic definition of knowledge in the
Theaetetus 167ab. The idea that epistemologists
might be concerned with which knowledge is
54
KNOWLEDGE:TRUE?
worth knowing - as opposed to which know-

ledge is legitimised by various different technical
accounts of knowledge - has not been pursued
much since, but Plato does give the issue pro-
found thought in the Theaetetus 172b-177c. (He
would no doubt disagree with our conclusions
here, though.)
Knowledge: True?
Usable information that is prepositional will
(usually if not all the time) be true. In our super-
market example, if the propositions induced by
the machine-learning program (e.g., that people
in Hertfordshire spend more, etc.) actually turn
out to be false, the supermarket will make poor
decisions based on this misinformation. Its acting
on these false propositions will be likely, all else
being equal, to reduce its profits. Hence, the
information will not be usable in the way
required by our characterisation; it will not be
able to underlie effective action.
Does this mean that all knowledge is true acc-
ording to our characterisation? Recall that this is
universally agreed in traditional epistemology
55
(Everitt and Fisher, pp. 48-9). Does the new episte-

mology that we are proposing here fall in with
the consensus?
The answer, interestingly, is 'no': one can act
effectively on information that is not strictly true.
To take an initial and obvious example: much
of the knowledge that is required in industry is
know-how rather than propositional knowledge.
It is knowledge that can help achieve something.
Hence it is of an imperative form (first do this,
then this, then this, if you want to achieve that),
rather than the propositional form (this is the
case), which is more usually associated with truth
and falsity. The imperative form can be useful or
not useful, appropriate or inappropriate, but not
strictly true or false.
A second example of knowledge that is not true
is what is called default reasoning. We rely on
default reasoning to an enormous degree in our
problem-solving. The idea of default reasoning is
that you can treat generalisations as universally
true, even when they are not. So, for example,
when told that Tweety is a bird, we deduce that
Tweety flies, based on our knowledge that all
56
KNOWLEDGE:TRUE?
birds fly. The reasoning is sound. Only one prob-

lem: it is not true that all birds fly (penguins and
ostriches don't, for a start), and if we later find
out that Tweety happens to be a penguin, then we
will withdraw the conclusion that he flies. This is
default reasoning.
In general, default reasoning has been ignored
by philosophers (with honourable exceptions). It
was first properly examined by computer scien-
tists trying to develop machines with artificial
intelligence, who were faced by the practical real-
isation that default reasoning is used all the
time.27 We blithely assume that all elephants are
grey, all rooms have floors, all cars have four
wheels, all tigers have four legs, all chairs can be
sat on, all restaurants charge customers for food,
etc. These propositions are all false (it is not true
that all rooms have floors, because when my
house was being renovated, the kitchen had no
floor; and so on with the other examples). But
surely, if we know anything, it is that rooms have
floors. The information provided by these propo-
sitions - the essential inferences they allow us to
make - justifies this conclusion.
57
A third example is that of heuristic reasoning.

This is reasoning by rule of thumb, and is a vital
part of the knowledge of experts in particular
fields (i.e., of expertise). Much expertise consists
of getting the right answer quickly with the mini-
mum of investigation. For example, given a med-
ical dictionary and a functioning laboratory most
of us could diagnose an illness eventually. The
problem would be that it would take so long, the
patient would probably be dead by the time we
knew what he had. The expert diagnostician, on
the other hand, has to pinpoint the disease quickly
enough to treat it. In order to achieve that speed,
the diagnostician will bundle together a lump of
medical knowledge into a quick-and-dirty rule,
which will tell him that, for example, if a patient
has a white-blood-cell count of x, and a certain
set of antibodies present, he probably has a dis-
ease from a particular class. This is a heuristic,
and is literally false, in that the patient could have
that particular white-blood-cell count and that
particular antibody and yet not have the disease.
Clearly, heuristic reasoning is related to default
reasoning; the distinction is that the heuristics are
58
KNOWLEDGE: BELIEF?
less likely to be borne out by events. They are not

deep-down assumptions but clever shortcuts
developed professionally as part of the execution
of expertise.
Each of these types of information, I am claim-
ing, should be seen as knowledge, in that they are
all extremely usable for effective action. Every-
one uses know-how and default reasoning; exp-
erts rely on heuristics. Can an epistemologist
possibly ignore these?
Knowledge: Belief?
Perhaps the most radical departure I am propos-
ing is that knowledge need not be a psychological
state. There are philosophers who are prepared
to take a smaller step, that knowledge isn't a
belief; for example, Timothy Williamson (see
Note 13). The requirements of the Internet age,
however, are stronger than that. Knowledge does
not have to be anything psychological at all.
As we have seen from the widget example earlier,
from the point of view of an organisation and its
need to manage its knowledge effectively to get
the maximum benefit from it, there is no essential
59
difference between the organisation possessing

knowledge via its employees, and its possessing it
via some artificial means (e.g., an Intranet, manu-
als, organisational procedures, etc.).
Seen from the JTB tradition, the widget exam-
ple changes radically as we go through the vari-
ous stages. When there is an expert on X-type
widgets in the firm, there is a (merely) metaphor-
ical sense in which the firm can be said to possess
the knowledge; when the expert leaves, then
there is no person within the firm with any
knowledge; the two situations are completely dif-
ferent. This surely renders the JTB tradition use-
less from the point of view of those who need to
understand and manage their knowledge in an
organisational context. Epistemologically, the
organisation is interested only in its capabilities,
not in how those capabilities are manifested. In
other words, it is the 'belief element of JTB,
rather than the 'justified' or 'true' bit, that has
most to do with rendering the JTB tradition
ineffective as a tool for understanding the
requirements of the knowledge economy which
we set out at the beginning of this book.
60
TWENTY-FIRST CENTURY ALCHEMY?
Conclusion: Twenty-first Century

Alchemy?
To review our analysis, we have examined the
JTB tradition in the context of the knowledge
economy, and found it wanting. An epistemology
suitable for this technological climate can be fairly
relaxed about justification; knowledge should be
justified, but different types of knowledge will
need different types of justification. The require-
ment for truth needs to be replaced with a
requirement for effectiveness (which will, one
imagines, coincide with truth to a large extent).
The analysis of knowledge as a belief, or as any
psychological state, seriously undermines the JTB
tradition in the modern context.
The proposal is that the JTB characterisation
be replaced by the characterisation of knowledge
as 'usable information'. We have illustrated the
ideas here, although of course what has trans-
pired in this book falls a long way short of a tight
definition. No doubt, giving such a definition to
the new conception of knowledge will be as diffi-
cult as the task of refining the JTB conception has
been. However, technologists such as W3C and
61
AKT have been able to use 'usable information' as

a practical guide to understanding the problems of
knowledge management, and developing techno-
logies that can help alleviate those problems.
There are responses that a philosopher of the
JTB tradition can make. A heroic response asserts
that the traditional analysis of knowledge defines
what knowledge is, and therefore usable infor-
mation cannot be knowledge (however interest-
ing or important it may be). This, in effect, says
that whatever problems there are in the know-
ledge economy, we're not going to help you with
them.
There is no argument against this response, of
course; it is perfectly self-consistent. But we have
seen from the account of the knowledge manage-
ment challenges that there are serious and inter-
esting epistemological problems associated with
them, and it would be disappointing were our cry
for help refused.
More constructive is an analytic response
which would review our examples and try to
show that they 'really' are examples of JTB in
some way or other. For example, this response
62
would attempt to show that some of our 'false'

knowledge is really true when you understand it
properly, or that what seem to be non-beliefs are
actually reducible to psychological states in some
way.
Though it takes our concerns more seriously,
this isn't a great improvement on the heroic
response. Its underlying premise, that epistemo-
logy is more or less okay once we realise that
we have misconstrued the significance of our
'counter-examples', fails to take into account the
essential need for experts to reach out to their
constituency, rather than the other way round.
An expert, whether an epistemologist, physicist,
astrologer or car mechanic, must tailor his or her
expertise to the people who need it, not vice
versa. It is not for the busy corporate executive or
NGO director to try to fit their knowledge prob-
lems into the JTB framework; rather, it is the job
of the epistemologist to show how the JTB theory
applies to the problems as they appear on the
ground. Any devious analysis is, in a sense, the
'hidden wiring' that the user of the theory must
not be bothered with.
63
But, if we are to face up to the facts, the

Internet and related technological innovations
have changed the way knowledge is viewed. In
Plato's day, when the JTB tradition began,
knowledge was a personal accomplishment, and
so it continued pretty well up to the present. But
there has been a gradual change, so that know-
ledge is now also a commodity, to be bought,
sold, managed, invested in, leveraged, deployed,
etc. We need an epistemology appropriate for this
context, and the JTB tradition isn't it.
Of course, I have been describing a world of new
technology and buzzwords that may be a passing
fad. It may be thought that if JTB sits tight and
ignores present circumstances, it will regain its
applicability. This is certainly possible, but doesn't
detract from the need for an information-based
epistemology now, and the mere fact that the
knowledge economy may disappear next week
shows neither that it has no pressing epistemo-
logical requirements, nor that any epistemological
discoveries made in that context will have no
wider interest or application.
In short, there is a need for an epistemology
64
tailored for a specific audience which the JTB

tradition has failed to deliver, and it is in danger
of missing a real opportunity to move from
academe into a constructive engagement with the
world of commerce and technology.
It may be that epistemology now is in the posi-
tion of the philosophy of matter in the late
medieval period. Philosophers had always had
theories of what the world was made of, from the
times of the ancient Greeks to the alchemists, and
these theories came under the heading of 'natural
philosophy'. However, in the seventeenth century,
particularly thanks to Newton, natural philo-
sophy gradually acquired an empirical basis, and
found itself debating, not antiquated thought
experiments about how far you could divide
space or substance, but problems such as the
correct description of the orbits of the planets. It
is certainly not the case that the alchemists made
no contribution to our understanding of matter,
but it cannot be denied that if you want to find
out about what makes up our world, you would
be wiser to ask a physicist or a chemist. Natural
philosophy gradually transmuted into hard science.
65
There are resemblances between epistemology

now and natural philosophy then. After millen-
nia of relatively inconclusive debate, there is for
the first time a basis and a requirement for
empirical application of ideas. Some of these
ideas will be of great use in this new world, oth-
ers will turn out to be useless. My guess, and I
can't say I know this, is that the JTB tradition
will turn out to be a dead end - the alchemy of
our day.
66
NOTES
Notes
1. There are many works defending this switch to the
'new' economy. For instance, Thomas A. Stewart,
Intellectual Capital: The New Wealth of Organizations
(London: Nicholas Brealey Publishing, 1997), Leif
Edvinsson and Michael S. Malone, Intellectual Capital
(London: Piatkus, 1997), Thomas H. Davenport and
Laurence Prusak, Working Knowledge: How Organiza-
tions Manage What They Know (Boston: Harvard Busi-
ness School Press, 1997). A more philosophical analysis
is Jean-Franc,ois Lyotard, The Postmodern Condition:
A Report on Knowledge (Manchester: Manchester
University Press, 1984).
2. See Joseph Stiglitz, 'Scan Globally, Reinvent Locally:
Knowledge Infrastructure and the Localisation of Know-
ledge', in Diane Stone (ed.), Banking on Knowledge: The
Genesis of the Global Development Network (London:
Routledge, 2000), pp. 24—43, and other papers in that
volume.
3. Alan Burton-Jones, Knowledge Capitalism: Business,
Work and Learning in the New Economy (Oxford:
Oxford University Press, 1999), pp. 8-9.
4. I am using the term 'sceptic' loosely here to describe
anyone who, for whatever reason, doubted the veracity of
objective knowledge. 'Sceptic' can also denote, in a more
narrow sense, particular schools of philosophy such as
the Pyrrhonists and the Academic sceptics (of whom
Cicero was one).
67
5. I am not, in this book, going to supply the bio-

graphical context of Plato's work, interesting though that
is. A small and neat summary of Plato's life is given in
R.M. Hare, Plato (Oxford: Oxford University Press,
1982, Past Masters series), pp. 1-8.
6. Socrates left no writings. His execution, on conviction
for the corruption of youth (i.e. spreading dissension,
rather than any more lurid crime), appalled Plato, and
many of the early works are defences of and justifications
for Socrates' philosophical method.
7. It is standard academic practice to refer to passages in
Plato via the numbering of the early edition by Stephanus.
These are given in the margin of most editions of Plato's
works.
8. For a discussion of the interpretative difficulties, see
Robin Waterfield's essay in his edition of the Theaetetus
(Harmondsworth: Penguin, 1987), pp. 218-25, which
includes the classic references.
9. For an interesting discussion of this ancient prejudice
against writing see Jorge Luis Borges, 'On the Cult of
Books', in The Total Library: Non-Fiction 1922-1986,
ed. Eliot Weinberger, trans. Esther Allen and Suzanne Jill
Levine (London: Penguin, 2000), pp. 358-62.
10. A.J. Ayer, The Problem of Knowledge (Harmonds-
worth: Penguin, 1956), pp. 7-35.
11. Susan Haack, Evidence and Inquiry: Towards
Reconstruction in Epistetnology (Oxford: Blackwell,
1993).
68
NOTES
12. Fred I. Dretske, Knowledge and the Flow of

Information (Oxford: Blackwell, 1981), pp. 85-106.
13. Timothy Williamson, Knowledge and its Limits
(Oxford: Clarendon Press, 2000).
14. Ludwig Wittgenstein, On Certainty (Oxford:
Blackwell, 1969), ed. Gertrude E.M. Anscombe, G.H.
von Wright and Denis Paul.
15. Nicholas Everitt and Alec Fisher, Modern Epistemo-
logy: A New Introduction (New York: McGraw-Hill,
1995), pp. 17-50.
16. Ibid., pp. 48-9.
17. This is not, perhaps, a major moment in intellectual
history, but for the record the paper in question is: Kieron
O'Hara, 'Sceptical Overkill: On Two Recent Arguments
Against Scepticism', Mind, vol. 102 (1993), pp. 315-27.
For Descartes, see the Meditations on First Philosophy,
particularly the first meditation, which is in a number of
editions, including John Cottingham, Robert Stoothoff
and Dugald Murdoch (eds.), The Philosophical Writings
of Descartes: Volume II (Cambridge: Cambridge Univer-
sity Press, 1985), pp. 1-62.
18. Lyotard, op. cit., p. 4.
19. Most of the figures in this section have been taken
from Varian's heroic - and entertaining - study, available
at http://www.sims.berkeley.edu/research/projects/how-
much-info/summary.html.
20. Statistic from Matthew Symonds, 'Haves and Have
Nots', in 'Government and the Internet' survey, The
Economist (24 June 2000), pp. 19-23.

21. For a dismissive account of the Internet, see Gordon
Graham, The Internet: A Philosophical Inquiry (London:
Routledge, 1999). I have criticised this book in a review
(written with Louise Crow) in International Studies in the
Philosophy of Science, vol. 15 (2001), and in an article on
'Democracy and the Internet', in the Web journal Ends
and Means, at http://www.abdn.ac.uk/philosophy/cpts/
ohara.hti. For a worry about the effects on democracy
and free speech, see Cass Sunstein, Republic.com (Prince-
ton: Princeton University Press, 2001). The classic
account of the effect of printing on Western civilisation is
Marshall McLuhan, The Gutenberg Galaxy (Toronto:
University of Toronto Press, 1962).
22. Lyotard, op. cit., p. 4.
23. Examples of important and interesting works on
organisations include Kenneth J. Arrow, The Limits of
Organization (New York: Norton, 1974) and Mary
Douglas, How Institutions Think (London: Routledge,
1987). Charles Jonscher writes very well about the
importance of retaining the human element in such dis-
cussions in Wiredlife: Who Are We in the Digital Age?
(London: Anchor, 1999).
24. 'The China Syndrome', The Economist (25 August
2001), p. 33.
25. Longer discussions of .these six challenges can be
found at the AKT Website at http://www.aktors.org.
26. The Gettier experiments appear in Edmund Gettier,
70
NOTES
'Is Justified True Belief Knowledge?', Analysis, vol. 23

(1963), pp. 121-3. This paper is reprinted in a couple of
more accessible collections of philosophical essays, and
Everitt and Fisher discuss the arguments in Modern
Epistemology, op. cit., pp. 21-9.
27. Raymond Reiter, 'A Logic for Default Reasoning',
Artificial Intelligence, vol. 13 (1980), pp. 81-132.
71
Further Reading
A good general survey of and introduction to epistemo-

logy is Nicholas Everitt and Alec Fisher, Modern
Epistemology: A New Introduction (New York:
McGraw-Hill, 1995).
In this book I have touched upon only a few of the
issues raised by Plato's philosophy. Dave Robinson
and Judy Groves, Introducing Plato (Cambridge: Icon
Books, 2000), gives a general overview of his work
and influence. One of the many virtues of Plato is that
he is one of the few philosophers who can be read for
pleasure. To read Plato's epistemological works, start
with W.K.C. Guthrie (trans.), Protagoras and Meno
(London: Penguin, 1956). Both are useful, but the
Meno is particularly important. More difficult, but
central to understanding Plato's epistemology, is the
Theaetetus: see Robin Waterfield (trans.), Theaetetus
(London: Penguin, 1987), or Myles Burnyeat (ed.),
The Theaetetus of Plato (Indianapolis: Hackett,
1990), which contains the text translated by M.J.
Levett. The important book-length essays by Water-
field and Burnyeat are invaluable for the student of
Plato, and give a good guide to the issues and the liter-
ature. Another useful early piece is Walter Hamilton
(trans.), Gorgias (London: Penguin, 1960).
72
FURTHER READING
The history and general pattern of scepticism is given

in the essays in Myles Burnyeat (ed.), The Skeptical
Tradition (Berkeley: University of California Press,
1983).
A nice little reference book for the Internet is John

Cowpertwait and Simon Flynn, The Internet from
A to Z (Cambridge: Icon Books, 2001). I particularly
like the appendix on emoticons or smileys. An alto-
gether heftier tome is Margaret Levine Young, Doug
Muder, Dave Kay, Kathy Warfel and Alison Burrows,
Internet: Millenium Edition: The Complete Reference
(Berkeley: Osborne McGraw-Hill, 1999, but regularly
updated).
Charles Jonscher, Wiredlife: Who Are We in the

Digital Age? (London: Anchor, 1999), is an excellent
history of information technology, with an important
focus on the human element.
The knowledge economy, and the way it rewards

creativity, is discussed in John Howkins, The Creative
Economy: How People Make Money From Ideas
(London: Penguin, 2001).
A perceptive analysis, or prediction, of the effect of

technology on knowledge is in the classic Jean-
Franc,ois Lyotard, The Postmodern Condition: A
Report on Knowledge, translated by Geoff
73
Bennington and Brian Massumi, with a foreword by

Fredric Jameson (Manchester: Manchester University
Press, 1984).
The World Wide Web Consortium (W3C) administers
HTML; its website is at http://www.w3c.org. A brief,
clear statement of the importance of XML can be
found in Tim Berners-Lee, James Hendler and Ora
Lassila, The Semantic Web', Scientific American, May
2001, available at http://www.scientificamerican
.com/2001/0501issue/0501berners-lee.html. The man-
ifesto of the AKT project is available at http://www.
aktors.org/publications/Manifesto.doc.
Perhaps the most influential book on knowledge man-
agement of late is Ikujiro Nonaka and Hirotaka
Takeuchi, The Knowledge-Creating Company: How
Japanese Companies Create the Dynamics of
Innovation (Oxford: Oxford University Press, 1995).
74
KEY IDEAS
Key Ideas
Data are collections of symbols.
Epistemology is the philosophical study of know-

ledge. It addresses questions such as: what knowledge
is; what kinds of knowledge there are; where know-
ledge comes from; whether there are things that we
can't know; how we can know what we don't know,
and look for it. Clearly such issues are strongly related,
for example, to the study of science and how science
progresses. A great deal of epistemological research
has been aimed at confounding scepticism. Plato was
the first epistemologist of any note; a list of greats
would include Aristotle, Descartes, Spinoza, Leibniz,
Locke, Berkeley, Hume, Kant, J.S. Mill, Bertrand
Russell and W.V.O. Quine. Important writers who
have made contributions working currently include
Donald Davidson, Alvin Goldman, Susan Haack,
Jurgen Habermas, Robert Nozick, Richard Rorty and
Timothy Williamson.
Information is understandable data, data with a

meaning.
Information overload is the situation, common in

modern society, where someone has everything he or
75
she needs to perform a task, but it is buried in a pile of

irrelevant information; the person then may not have
the time or resources to find the important stuff.
Hence possession of the information is not sufficient
for it to be of any use.
The Internet is the technological advance that has

allowed anyone who can get onto the telephone net-
work and has a suitable receiving system (e.g., a com-
puter, a mobile phone) to access any of billions of
computer data files. The relevant technologies are very
simple: TCP/IP and dynamic routing. The net effect -
no pun intended - is to bring the largest quantity of
information ever gathered together within everyone's
reach.
Know-how is a type of knowledge often ignored by

epistemologists, yet is (a) a very common type of
knowledge, and (b) extremely important as far as the
knowledge economy goes. Representing know-how,
rather than prepositional knowledge, is one of the
most imperative tasks of the version of epistemology
outlined in this book. In the knowledge-management
literature, the terms used for know-how and preposi-
tional knowledge are procedural knowledge and
declarative knowledge respectively.
76
KEY I D E A S
Knowledge is a type of information that aids the per-

formance of effective actions. Traditionally, epistemo-
logists have focused on the psychological aspects of
knowledge; however, modern conditions - the know-
ledge economy - have shown the need to expand the
definition of knowledge to include non-psychological
states.
Knowledge acquisition is the process of adding

knowledge to an agent's stock (the agent can be
human, or non-human such as an organisation). Tasks
for a reconstructed epistemology include: outlining
reliable methods for acquiring knowledge from people,
computers and economic markets; and finding ways of
discovering knowledge that exists only implicitly
within an organisation and making it explicit.
The knowledge economy is one in which know-

ledge is a force of production and a source of competi-
tive advantage. The exploitation of raw materials and
labour characterised the so-called old economy; the
knowledge economy involves a general switch to
services from manufacturing, and places a premium on
an educated and productive workforce.
Knowledge management is the task of making

sure that an organisation's consumption, acquisition
77
and use of knowledge is cost-effective. In other words,

the task of: ensuring that any knowledge that the
organisation possesses is used as often and efficiently
as possible; ensuring that any knowledge that the
organisation needs but does not possess is acquired;
ensuring that all the knowledge the organisation pos-
sesses adds as much as possible to the organisation's
effectiveness.
Knowledge sharing is a tricky part of knowledge

management. Within an organisation, particularly a
big or widely spread one, it is often the case that the
knowledge that organisation possesses does not get to
the people who need it. Employee X does not know
that employee Y knows how to solve X's current prob-
lem. The aim of knowledge sharing is to ensure that X
can find out who in the organisation can help him or
her, without contributing too much to information
overload.
Scepticism is a type of philosophical objection to a

theory, objecting to the basic assumptions of that theory.
Sceptical arguments are usually very pernicious, and
the net result of epistemological scepticism has been
that, over the millennia, epistemologists have spent
much too much effort trying to confound the sceptic
and not enough finding out about knowledge, and
78
KEY IDEAS
therefore modern-day epistemology is not adequate

for the serious epistemological problems of knowledge
management and the knowledge economy.
The Semantic Web is a development of the World

Wide Web, involving the use of the markup language
XML which allows a Webpage to include within it a
specification of the knowledge it contains. This will
allow much more intelligent searching and navigation
of the Web.
The World Wide Web is a portion of the Internet

which provides a common format for multimedia
computer files to be displayed on-screen through a
browser. It exploits the markup language HTML,
which governs the display of data on-screen, and is not
tied to any one operating system (such as Windows or
Linux). This gives the Web a common look and feel,
wherever you are accessing it from.
79
POSTMODERN ENCOUNTERS
Plato
and the Internet
In the new knowledge economy, competitive advantage no
longer derives from labour and raw materials, but from
knowledge and creativity. A staggering amount of
knowledge is now available through the Internet. But
amidst billions of documents, we find ourselves drowning
in information, while being starved of the knowledge we
need. When the economic Imperative is to get clever
things done in smarter ways, it is clear that we need to
know more about knowledge.
And yet - what is knowledge? Though the arguments have

changed little since Plato's writings of 2,500 years ago,
Kieron O'Hara contends that the essential contrast Is not
between knowledge and belief, but between knowledge
and information. In Plato and the Internet, he argues that
what is important is not 'what facts you know', but 'what
you know how to do', and that we should expect to find
knowledge not only in our heads, but also in our
surroundings.
Kieron O'Hara is a senior research fellow in the

Intelligence, Agents and Multimedia Group at the
University of Southampton. He co-wrote the script of the
computer game Tomb Raider 4, and is the author of the only
scholarly paper about Carry On Cabby.
icon Books
UK £3 OB
Uff\ Xiv>IW ' KtJM-I+044-J44-5
^tm^m^^f -tt«*M
Canada $8.99
Totem Books
USA $7.96 .
lit ttw United StBtMt
dMrttwlwltalli.trMl.tar
JtatiOfMR Book rtatwork inc. , , _. «
Printed (nth* UK WWW.ICOIWOOlt8.ce.IW

Plato and The Internet PDF

Uploaded by

Copyright:

Available Formats

Plato and The Internet PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Plato and The Internet PDF

Uploaded by

Copyright:

Available Formats

POSTMODERN ENCOUNTERS

Series editor: Richard Appignanesi

Distributed in the UK, Europe, Distributed in Canada by

Published in Australia in 2002

Text copyright © 2002 Kieron O'Hara

capital.1 Indeed, many hold out the hope that the

What is Knowledge? The Discipline

spot it, how you make sure you have enough of

the dross on the World Wide Web, the true from

opponents in his writings on knowledge were

Plato's opposition. There is a lot of power in a

In general, scepticism is found attractive as a

Plato and Knowledge: The Birth of

had great affection,6 and included other promi-

English public-school system was one attempt at

SOCRATES: Do you think that knowing and

such things as true and false beliefs, you would

True beliefs are useful, and will not lead you

SOCRATES: True opinions are a fine thing, and

Plato spent much effort trying to establish the

belief, particularly in his great epistemological

Other Types of Knowledge

• Know-how. In Plato's distinction, the differ-

encyclopaedia contains knowledge. If there is

Why does Plato ignore these types of knowledge?

There are more complex reasons, too. For

Writing shares a strange feature with painting.

and over. Once it has been written down, every

The written is infinitely less interesting to Plato

None of these things seems particularly contro-

Justified True Belief

The point I want to make is that already, in the

psychological state, which involves the person

terms of the information that caused the belief.12

[W]e have learned that there is universal agree-

there is an almost equally widespread agreement

Surely another option open to us is to reject the

The Sceptic Bites Back

position without advancing any positive thesis

see Note 14), and focus epistemological efforts

was large, the cost of reproducing it huge. Bound

with, very little of the new material recorded in

data for every man, woman and child on the

The Internet and the World Wide Web

difference at all. It is extraordinary, though,

multiple media - text, pictures, sounds, video,

Clicking on such links allows your computer to

amount of information for people to have access

Knowledge, Technology and

tains or possesses knowledge about X, in this

problem-solving behaviour. If something goes

not have to stop making X-type widgets; neither

contains, or possesses the knowledge about, X. If

We may thus expect a thorough exteriorization

And indeed, we shouldn't end this section with-

ceramics market. For instance, over the last few

that they are not merely management issues, I

• Acquisition. The initial problem with know-

to know? How can different types of know-