Erik Sandewall
Leaving Traces – Words on the Road
Speech at the Inauguration of a statue
Linkoping, August 20, 2014
original: https://www.ida.liu.se/statyn/statue-citations.pdf
Linkoping’s great son - Tage Danielsson - once explained, with his
characteristic irony, that accidents in nuclear reactors are so
extremely unlikely that they actually never happen. This is also how I
feel to day: the probability of one day being able to participate at
the inauguration of a statue of yourself is so small that I find it
difficult to believe that it has happened. However, if it has happened,
and if you make a visit to the statue, you should be able to see that it
contains some inscriptions on the ground or roadway that the statue
person is walking on. My topic now is to explain the meaning and the
background of those inscriptions.
They came about because the sculptor, Johan Malmstrom, invited me to
select some words or phrases that could be placed on the statue, and I
have then selected five dictums or expressions that have been
inspirational for me for a long time. I call them ”words on the road”,
”ord pa vagen” in Swedish, and for background I should tell our
foreign visitors here that in Swedish language, just as you can give
someone ”one for the road” meaning a drink before departure, you can
also give someone ”a word for the road” or ”word on the road” meaning
something to keep in mind during the journey. So the inscriptions on the
base of the statue, the road that the statue person is walking on, are
intended as such ”words on the road”, both for the occasional
passers-by, and for the students that start their scientific journey
here. The first one of these phrases is from the title of the
concluding work of Galileo Galilei,
Discorsi e dimostrazioni matematiche, intorno a due nuove scienze
which means ”Discourses and Mathematical Demonstrations Relating to Two
New Sciences.” This work is a very important one in the history of
science for a variety of reasons, but the reason for my interest is
maybe particular. As you know, I have been active during the early
development of computer science as a new science, and I have seen how it
is not merely a spin-off from an existing science (such spin-offs happen
all the time). Computer science is a fundamentally new science, and
then it’s interesting to study earlier situations where radically new
sciences were defined. I am preparing a written note about this topic,
but let me mention the two main points here. First: Galileo proposes
not one, but two new sciences, but in the article he does not specify
them by name or in any other clear way. Commentary in the history of
science usually just say that Galileo was a founder of modern physics as
an empirical and experimental science that uses mathematics to formulate
its findings. Some identify two specific areas that are covered in this
article, namely, kinematics and strength of materials. Another,
tantalizing, interpretation is that this article is fundamental for both
natural science and engineering science. This may or may not have been
on Galileo’s mind, but it is true that this article as well as Galileo’s
other works have been fundamental, not only for physics, but also for
engineering sciences.
Then what about computer science? It is of course an engineering
discipline, since it deals with design principles for a certain class of
structures and mechanisms, namely, data structures and computer
programs. But should it also be considered to coexist with an empirical
science, in the same way as Galileo introduces physics and engineering
concurrently?
I think the answer is yes, or at least it should be yes, and that the
empirical counterpart in our case is a field that ought to be called
information science, although with a different definition than the usual
one for that term. The empirical science that I have in mind is one that
includes the field of knowledge representation, as well as large parts
of cognitive science, and of computational linguistics, and of so-called
semantic web studies, and several others. Taken together, these may
constitute an empirical discipline or interdisciplinary area where we
study the structure of information on the level of concepts and
relationships, not merely as sequences of bits, as in information
theory.
My second point is that Information science in this nonstandard sense is
complementary to the empirical science that Galileo advanced, in an
interesting way. Galileo used a classical distinction between the
”primary” and the ”secondary” qualities of objects in the world, where
primary quali- ties are intrinsic ones such as velocity and temperature,
properties that the object has in itself, and secondary qualities are
the ones that an observer assigns to them. He wrote:
"I think that tastes, odors, colors, and so on are no more than mere
names so far as the object in which we locate them are concerned, and
that they reside in consciousness. Hence if the living creature were
removed, all these qualities would be wiped away and annihilated"
One may add that the observer-assigned secondary qualities will also
include the classification into different types of objects and the
relations between objects, so they are the concept-level information
that is important in many parts of computer science, and that is the
object of study in information science.
To summarize: Galileo and his contemporaries introduced the focus on
what they called the primary qualities of the world, and on the use of
mathematics for describing those qualities, and as well they introduced
the synergy between fields of science and fields of engineering. This
has been enormously important during the almost four hundred years since
then. In our epoch, we can begin to see the emergence of a similar
tandem between an engineering field and an empirical field, namely,
computer science and information science, but here it is the secondary
qualities of the world that are at the center of interest. To my mind
this is a developent of long-term historical importance.
In this perspective I think it was both appropriate and fortunate that
the name of our department, which we decided more than 30 years ago, was
chosen as the Department of Computer and Information Science. There were
also other reasons of a more temporary nature for the choice of this
name, but we can see now that it’s a name that is in line with long-term
perspectives.
The second ”word for the road” is a quotation from our great countryman
Carl von Linn´e, or Carolus Linnaeus, who wrote
Omnia mirari, etiam tritissima
which translates as ”Everything deserves attention, even the most
trivial and mundane”. This is a good advise when it comes to software
development and to software documentation, where the devil lurks in all
the details. However it is more profoundly relevant in the formally
based variety of information science, and here I must return once more
to the connection with Galileo. He characterized the scientific method
as consisting of three steps, which he called resolution, demonstration,
and experiment, successively. (The term demonstration here refers to
the demonstration of mathematical arguments). The first step, which he
called resolution, should be understood as organized observation.
Organized observation may apply to a small set of phenomena, but quite
often it requires a large set of observa- tions, as done for example by
Tycho Brahe, the astronomer, as well as by Carolus Linnaeus himself, and
by many others.
In information science, organized observation and structuring is
represented by the construction of large knowledgebases, and factbases,
and ontologies, and semantic dictionaries, all of which are needed for
bringing concept level knowledge closer to being formally understood.
The knowledge that is captured in these knowledgebases is trivial and
mundane for us. It is exactly the trivial and mundane information that
it is such a challenge to come to grips with, especially when it occurs
in very large quantities – and hence we need to remember ”Omnia mirari”,
do not disregard something merely because it seems to have a trivial
character.
I come then to the third ”word for the road” which is also a well-known
one, this time from Descartes, or Cartesius, who introduced the
cartesian coordinate system, but who also formulated the dictum
Cogito ergo sum
meaning I think, therefore I exist. The reason why Descartes was
interested in this composition of two obvious statements was that in his
philosophy, and following the principles laid down already by Aristotle,
he wanted to eliminate everything that could possibly be questioned, in
order to arrive at an irrefutable theory. To begin with he needed one
single statement that could not be doubted, and that would be used as a
solid point of departure for his continued investigation. This
enterprise reminds in a way of the one by Galileo, I think. The
Routledge Encyclopaedia of Philosophy makes the following statement:
In beginning with the Cogito, we build a philosophy detached from
history and tradition.
and this is of course analogous to what Galileo did. I like the phrase
’cogito ergo sum’, not only because it tries to build a conceptual
framework from first principles, but also because it represents the
ultimate in reflexive thin- king, that is, thinking about your own
thinking, which is arguably one of the necessary requirements for
intelligence in the machines.
One of the well-known objections against the Cogito argument is that the
existence of an ’I’ is assumed in the first part of the dictum. If you
say ”I think” then the existence is assumed right there, so the
statement assumes what it claims to prove. Maybe so, but I would like to
turn this argument around: the phrase Cogito ergo sum reminds us that
thinking as we know it requires an agent, an ’I’. This is what makes it
so fascinating from the point of view of artificial intelligence. I have
proposed that the concept of ’software individuals’ is going to be
crucial both for artificial intelligence and beyond. By this I mean
software artifacts that have an identity and that are able to
understand, and to reason about their own identity, as well as the
identity of their peers.
If Descartes’ argument is to be taken seriously, then these software
individuals should also be said to ’exist’. And this argument would be
valid even if the computation in them were to be distributed over
several pieces of hardware in the so-called cloud, and even if many
other individuals would execute in the same cloud at the same time. So
is this compatible with our notion of ’existence’?
My fourth quotation is from a work in a Nordic language, namely the
Icelandic Edda, which contains this well-known verse
Deyr fe,
deyja frændr,
deyr sjalfr et sama;
ek veit einn,
at aldri deyr:
domr um dauδan hvern.
The first three lines here remind the reader the his cattle will die,
meaning that his worldly assets will run out, and that his kinsmen will
die, and that he will die himself as well. These somber observations
would seem to be just as undoubtable as the Cogito that Descartes used
as his starting point, but of course they are a lot more concrete.
(Actually, religious systems of thought are often based on these
premises). Anyway, the unknown poet then states that he knows one thing
that is never going to die: ek veit einn at aldri deyr, namely, a
person’s posthumous reputation. If we translate this to our academic
reality, it suggests that your final CV will stay on forever, and maybe
the same will apply to the reports from the evaluation committees of
your research projects.
These thoughts may be encouraging, or scary, but I would like to add
another meaning to the phrase Ek veit einn at aldri deyr, namely, that
science itself will never die. Think of it: Empires and civilizations
have come and gone, since the dawn of science as we know it, but science
has persisted and it has been passed on from one generation to the next,
and from one culture to the next. It is driven by the natural curiosity
of human beings, and it survives because of our ability to transmit
scientific insights to those that come after us. This is the greatest
thing about being a scientist: the work that you immerse yourself in
contributes to an edifice that will live on and that has a permanent
value. And that is really worth one’s lifetime efforts. My fifth and
final quotation is not on the statue, since this might have been
misleading. It is a dictum that is used in some formal contexts by the
Royal Swedish Academy of Sciences, namely
For efterkommande
meaning For those who come after us. This simple phrase was adopted
during the early years of the Academy, since it was formed with the
explicit goal of harboring research that has practical utility, such as
Linnaeus’s expeditions for example.
This all reminds us that there are two different reasons for pursuing
science, namely to satisfy our natural curiosity, and to do something
useful. These two reasons apply both to individuals, and to the entire
society, and it is usually said that these reasons go hand in hand: if
science and scientists are just allowed to pursue their scientific
goals, which tend to be curiositydriven, then the useful practical
applications will be obtained sooner or later.
The curiosity argument applies to the empirical sciences, since we want
to find out more about the world, but it applies as well to engineering
sciences. The desire to construct things also seems to be an innate
characteristics of humans, as witnessed by the construction of hunting
traps and pyramids for example. And research in computer science
contains many examples of where people built software systems that did
things that no program had ever done before, and more or less for the
sake of trying.
But the results of research are not always beneficial, so one may argue
that the dictum for efterkommande should not only be seen as an
encouragement towards what we are anyway doing, but also as an
admonition to think about its long-term effects, although it is very
difficult to sort out what are the beneficial effects, and what are the
detrimental ones. In this context I think of a statement by the new
archbishop of the Church of Sweden, Antje Jackel´en, who has said about
religion, in her case, that its proper role is to ask questions, and not
to provide answers. For us as researchers, since it is virtually
impossible to properly assess the effects of research, I think the right
attitude for us also is to ask the questions, and to think hard about
what the answers may be, and then to act according to your conscience.
What kind of questions? Questions about anticipated negative effects of
science often concern the violent use of its results, for weapons and in
war. But when it comes to information technology we must also consider
its possible use for another type of invasion, namely, for the invasion
of privacy. We need another type of questions, another question schema
to use our jargon, that concern privacy invasion. What has been the
effect of e-mail technology on human privacy? What about the use of
electronic payment systems? What has been the effect of search engines,
such as Google? What about unmanned vehicles on the ground, and in the
air, so-called drones? What will be the effects of the so-called cloud
technology that is increasingly being used? What about automatic
language understanding and translation? What about knowledge
understanding systems? What about artificial intelligence?
No one has convincing final answers to these questions, and it is easy
to conjure up both optimistic and pessimistic predictions with respect
to these ongoing developments. That is why I would like to interpret the
dictum ”For efterkommande” as an admonition to continue asking these
questions and to discussing the possible answers. Or in short, let’s
just think of what we are doing.
Turning back now to my new statue – I am very happy with it. And I
really appreciate having had the opportunity to highlight these
inspirational quotations by having them inscribed on it. I hope that
many people will read them, that they will get their own ideas from
them, and maybe begin asking ”why are these quotations here, what are
they doing in this context?” I have added some speculation of my own in
my interpretations of these quotations, and the visitor may or may not
agree with my ideas, but hopefully he or she will also relate to the
inscriptions by asking questions and thinking about the answers. This
statue has a very interesting title; it is called Leaving traces. This
is shown physically by the footprints on the ground in the statue, which
you can see there together with most of the quotations that I have
talked about. Maybe the artist - Johan Malmstrom - wanted to suggest
that I have been leaving some traces, but this will be for others to
judge. What I would like to read into it, from my point of view, is that
the statue person is guided and inspired by the traces left by those
that walked the road before him. This is something that applies to
every scientist. There is this well-known phrase that we can do what we
do because we stand on the shoulders of giants, but actually I think the
observation that we as scientists are leaving traces is also very much
to the point. We are not standing still when we do science, we move
ahead, we search our way ahead, and while doing this we leave traces,
but we are also led by the traces of those that researched before us. I
want to thank Johan Malmstrom for this fascinating concept as well as
for its realization as a statue, and for our interactions when it was
prepared.
But most of all I want to thank IDA, as represented here by Mariam
Kamkar, for having given me the honor of being elected a statue person,
and for having acquired and mounted this very nice statue.
And finally I want to thank you all for having come here for this
inauguration event.
_g@@@@y_
_@@@@@@@@R*
_ @@@@5@FRE
. $@@y a$
=._ZW$$ ` FE, , ,
,zyKa@g~ _a@@@gg$~ag
a@g@@@$ $@@@@@@@$
4@@FE 4_ ~?@@@@@P~
~~@@, #_ w a@@@F~ .
#@L ,~.yy@@@@@E
`# gg@@@@@@E~
` ~4@@@@@@@yyy_
gr _gy@@@@@@@@@@'
_y@@_2@@@@@@@@F~` .
g@@@@@@@@@@@@@@@F
_ygg@@@@@@$__P@P@@$~ _*
__g@@@@@@@@@@@@@@@@gg_~ ag@~
_y@@@@@@@@@@@@@@@@@@@M@@@Fy7@~a_
*H@@@@@@@@@@@@@@@@@@@@@gR_=F= ~~\
F$@@@@@@@@@@@@@@@F~~@w@_ ' L
`-@@@@@@@@@g. `yd@ $
\ W@4@@@@@@$; "" $ Q
__ , ~@F`~~@@@@@y L `
$@@ $ `^agy__$@A4$, ," ,
4@@yw @$_ ~~``` d@_ ,' ' $
+ $@@_g@@@% _ga ~_ , | $
_ @FR@@@@F ~ _ "y '' | $L
a ~ g@@@@$ 1g@a@y ?g . y$ @
g a@@@@@@E 7F@F4@a* + `a_ 4@ _@
g@by@$@@@@@$ y ~+ 7g i 4@ g@
y@@@@@@@@@@@Ma." + y ?$ @ @@
4@@@@@@@@@@E F , "# | T _@@s
_y@@@@@@@@@@$ J _ , g, 4@F~
i y@@@@@@@@@@@~ = , $ y@' a@_
$@@@@@@@@@@@ _ ^ == 'a@$@@L y@@F .
J a@@@@@@@@@@@@ $L *- f@R@@$ @@@@y
l a=@@@@@@L`@@@ ~~ $@@@@F\ y@@@E2~~_