Knowing without understanding: A reflection upon different ways of knowing, solving problems and computational approaches

Abstract

In this text I set out to reflect on different ways of knowing and problem solving by using examples from design experience, the scientific method, indigenous knowledge, and also address the role of intuition and computational methods (‘big data’). I suggest that the scientific method is only one particular way of enquiry into creating (and sharing) knowledge among many, and that at the root of every original enquiry lies an intuitive insight.
Further I propose that ‘knowing without understanding’ is the common state for much of our practical and applicable knowledge. We have acquired much know-how but little accurate know-why. From here i try to demonstrate parallels between quantitative algorithmic methods and what we call human ingenuity or intuition. However, this text is not about truth, creativity or problem solving, although they all are relevant. It is about different ways of coming to know.
The conclusion I arrive at is, that for most of us much of our knowledge consists of know-how only, and not of a true understanding, in the sense of deeper insights into the true causes of processes or reactions, (if such a thing was objectively possible from a constructivist perspective). If that is so, the alleged radical break with the tradition of knowing which results from computational methods, - as Chris Anderson suggests in the quote above - through which researchers may arrive at solutions and answers, without understanding why these are true, may in fact be nothing new, but merely the continuation of a long tradition of knowing without understanding, although with larger implications.
Keywords: experience, intuition, theory, practice, scientific method, knowing, episteme

Full text

Knowing without understanding: A reflection upon dif-
ferent ways of knowing, solving problems and compu-
tational approaches
Figure 1: “You can find out how to do something and then do it, or do something and then find out
what you did.” Isamu Noguchi, artist (Seen at Storm King Art Center, NY, after returning from
C:ADM 2010)
“The new availability of huge amounts of data, along with the statistical tools to crunch these
numbers, offers a whole new way of understanding the world. Correlation supersedes causa-
tion, and science can advance even without coherent models, unified theories, or really any
mechanistic explanation at all.”Chris Anderson, writer and editor-in-chief of Wired Magazine: The
end of theory, will the data deluge make the scientific method obsolete?, Edge.org, June 2008
“The view that scientific knowledge is not structured for use in design has a parallel in the use of
science in everyday activity, [...].” Nigel Cross in M. J. de Vries, N. Cross and D. P. Grant 1993,
p.272
‘‘Your dead child. Prepare him for new life. Fill him with the earth. Be careful! He should not over-
eat. Put on his golden coat. You bathe him. Warm him but be careful! A child dies from too much
sun. Put on his jewels. This is my recipe.’’1 Richard Sennett, 2005, The Craftsmen, p. 190

Abstract
In this text I set out to reflect on different ways of knowing and problem solving by using examples
from design experience, the scientific method, indigenous knowledge, and also address the role of
intuition and computational methods (‘big data’). I suggest that the scientific method is only one
particular way of enquiry into creating (and sharing) knowledge among many, and that at the root
of every original enquiry lies an intuitive insight.
Further I propose that ‘knowing without understanding’ is the common state for much of our prac-
tical and applicable knowledge. We have acquired much know-how but little accurate know-why.
From here i try to demonstrate parallels between quantitative algorithmic methods and what we
call human ingenuity or intuition. However, this text is not about truth, creativity or problem solving,
although they all are relevant. It is about different ways of coming to know.
The conclusion I arrive at is, that for most of us much of our knowledge consists of know-how only,
and not of a true understanding, in the sense of deeper insights into the true causes of processes
or reactions, (if such a thing was objectively possible from a constructivist perspective). If that is
so, the alleged radical break with the tradition of knowing which results from computational meth-
ods, - as Chris Anderson suggests in the quote above - through which researchers may arrive at
solutions and answers, without understanding why these are true, may in fact be nothing new, but
merely the continuation of a long tradition of knowing without understanding, although with larger
implications.
Keywords: experience, intuition, theory, practice, scientific method, knowing, episteme
Introduction
Attending the American Society for Cybernetics’ (ASC) conference “Cybernetics: Art, Design and
Mathematics” (C:ADM) in 2010 at Rensselaer Polytechnic Institute in Troy as a designer and
design researcher, and later as a member of the ‘methods group’ editing the proceedings and this
publication, was a profound learning experience for me. During the conversations however there
appeared to be much emphasis on the ‘scientific method’, rigourous application of methods in-
formed by reasoned, rational thinking.
I experienced all this somewhat frustrating as from my experience the general design activity is not
very rational but much more intuitive, building upon experience, trial & error, reflection and tacit
knowledge to explore a design idea in different iterations towards a certain goal. As such develop-
ing a design solution is more a hands-on process developing from the bottom-up then following a
theoretical procedure top-down. Rational explanations often are ‘invented’ later in this process and
woven into a rationale in order to make the implicit explicit, and also to provide stakeholders with a
coherent conceptual model. Some explanations could be seen as having the function of demon-
strating competence by adding retrospectively a certain ‘logic’ and rigour to an otherwise inherently
intractable, qualitative and intuitive process. And while intuition is essential for framing and solving
design problems at the same time it is considered a vague and unsystematic practice, in need of
support by a rational explanation. But can intuition be regarded a ‘method’ in the sense of a partic-
ular form of procedure for accomplishing or approaching a problem? Is intuition a skill? And if so,
how could we be able to consciously make use of it? That some of these questions seem unreas-
onable shows how unclear our relationship with intuition is. This could indicate that we understand
very little about our creative sources and how to consciously make use of them. Among the few
things we know is that, when one gets stuck on a certain problem it may help to go for a walk and
let the mind wander or switch to another, unrelated activity, and somehow a necessary step to-
wards a solution may come to mind. However such a strategy may not be very reliable and does
not guarantee a consistent outcome. As such, ‘intuitive thinking’ hardly can be considered a meth-
od as we do not know how we do it.
When conversing about methods at C:ADM, quantitative methods dominated the discussions,
which I found frustrating. For a designer quantitative methods appear seductive for they often
provide clear evidence for a successful design, such as Gordon Murray’s designs for racing cars
which have proven their effectiveness by winning many races (Cross, 2007, 92). However such

competitive situations are rare examples of design problems. In the sciences quantitative methods
have a long tradition and are part of an established discourse. As such it is easier to communicate
about them, while there appears to be less consent regarding evaluating and analysing qualitative
‘data’. Among all the conference’s strong advocates for scientific reasoning, i missed the support-
ers of ‘other ways of knowing’. Of those ways of thinking required in solving problems which often
do not have an immediately obvious technical solution but involve exploration, improvisation, play,
material properties or unclear situations, in which intuitive ways of thinking may be helpful in fram-
ing the problem.
Clearly intuitive insights and creative leaps of mind must pre-/historically have been crucial in the
long history of human problem solving, compared to which the scientific method is a rather recent
invention intended to supersede unsystematic and less reasonable approaches for particular types
of problems. Which way of thinking is more helpful for inventing? Before the invention of the bow
and arrow it was impossible to think of a bow and arrow. Yet still it was invented. Problem solving
involves a sense for possibility and of thinking beyond conventions and accepted limitations, with a
particular goal in mind. Both scientists and designers make use of this way of thinking.
“Intuitive understanding, in short, is not contrary to science or ethics, nor does it appeal to instinct
rather than reason, or to supposedly ‘hardwired’ imperatives of human nature. On the contrary, it
rests in perceptual skills that emerge, for each and every being, through a process of development
in a historically specific environment. These skills, I maintain, provide a necessary grounding for
any system of science or ethics that would treat the environment as an object of its concern.” In-
gold, 2000
Context: Intuition and method in design
From my experience as a designer and design researcher I often approach problem solving from a
less conscious and rational direction. Together with a design problem in mind arises a susceptibil-
ity, openness and way of looking at materials, principles or processes of the world around us that
might inspire to assist the solving of a particular design problem.
As an example I will describe how I approached the ‘standard’ that we were asked to bring with us
to the conference, as a design problem. The constraints: The standard should be a mobile display,
bear our name, be at least 1 Meter off the ground, help generate conversation and fit into our lug-
gage. The description did not explicitly state that it should be self-made. My immediate idea was to
create a self-supporting structure after the principle of a Buckminster Fuller/Kenneth Snelson
‘Tensegrity’ carrying a poster (Figure 2).
Figure 2: Kenneth Snelson, 1974, Free Ride Home, Tensegrity
Such a self-supporting structure of string and rods would elegantly combine principles of mathem-

atics, architecture and design and thus, conceptually, fit the conference theme. However, I also
was aware that this undertaking would require a substantial amount of work. Additionally it would
require time to assemble the structure at the destination. My solution should function with a simil-
arly elegant self-supporting principle but require less labour in construction and assembly. Some
days after this assignment I visited the Royal Society’s 2010 “Festival of Science and the Arts” at
the Southbank Centre in London. Here about 40 stalls exhibited scientific research ranging from
research into mathematics, astronomy to health and biology. A group of mathematicians presented
mathematical functions and used, as visual and tangible aids, bi-stable rings or hoops made of
rolled steel, about 40 cm in diameter. These rigid hoops were ‘O’ shaped, yet when twisted into an
‘8’ shape and released would ‘snap’ into a smaller, double-layered ‘o’ shape. Such a structure can
come to rest in two different states and is therefor described as bi-stable. After my visit I realised
that this principle of bi-stability might provide a solution to my ‘standard’ problem. My standard
could be made of multiple sections of hooped steel, which could be coiled up to fit into the luggage
and yet, once released, would pop-up into a standard, no assembly required. I started drawing
sketches (Figure 3).
Figure 3: Sketch for standard made from sections of
a large rolled steel hoop. Standing on four legs and
joined together with a pipe-clamp. Not knowing the
material properties, I anticipated that the individual
rods would tend to coil up. To prevent this, the tips
were held together with string.
Figure 4: Three-legged standard after as-
sembly. The steel rods being extracted
from a pop-up tent and sawed into three
sections, fastened together in their mid-
sections by foam and cable ties.
I required about 5.5 Meters of rolled steel. Friends recommended to buy a pop-up tent and extract
a rolled steel hoop from there. A pop-up tent was bought and in the workshop I sawed the extrac-
ted rods into three sections, each of my own height and used a discarded piece of foam and cable-
binders to join the three sections at their centres (Figure 4). Other then anticipated the three sec-
tions did not automatically coil up and could be gently bent into the required shape, without having
to connect their tips with string to prevent them from coiling up. The circumference of the structure
should not exceed 84 cm which equals that of a cylinder made of two A3 sheets (29,7cm x 42cm)
of paper.
During this process of handling the material I made a number of observations and decisions. Once
the material was at hand, the sketch very much lost its authoritative function. I mainly had to impro-
vise, going in my mind through a number of alternative and radically different solutions (a helix? a
zig-zag structure?), while gaining hands-on experience with the material. As time was running out I
knew I had to find a solution within a single two-hour workshop visit. The first insight gained was
that three legs would be better then four legs to provide a solid stand, to hold the cylindrical poster
and save material. The second insight was that cable binders instead of pipe-clamps would
provide a more elegant solution to join the rods together as their use required no tools to carry and

that a piece of modelling foam would not only provide a solid structure to fix the angles of the rods
but also a comfortable handle to transport the standard. In this process I also realised that a coun-
terweight in the lower half of the standard was not necessary as the structure was slightly heavier,
more solid and sturdy as anticipated. In the end the most difficult process was rolling the rods into
a tight coil which would fit into my luggage. This required over one hour. The final result looked
very much like the initial sketch, but more importantly worked according to the imagined principles
of a self supporting structure.
What was the knowledge, that ‘hunch’, that triggered a particular approach to solve the open prob-
lem within the constraints provided? With the ‘Tensegrity’ principle in mind i realised the potential
of a somewhat related structure at the science event. (I would describe the tensegrity principle as
a structure obtaining its stability through a controlled balance of flexible parts (string) and inflexible
parts (rods) which support each other forming a tense and integer structure.) The scientific method
is hardly applicable in this design process as the solution did not develop through an experiment or
an articulate theory or hypothesis, but was based on a concrete problem, an idea for a solution,
and an openness to make a mental connection when observing a similar principle.
I think this way of looking at the world is not the privilege of designers, but inherently and intrinsic-
ally human and developed during the evolution of tool making, cooking, breeding livestock and
crops, or building shelter. However somehow this knowledge to conceive and apply these tech-
niques of survival today appear to be worth less, vague and less effective compared to the out-
comes of scientific thinking. Why is that so? Although, for example, while our growing knowledge
of the genome and dna are modern, crops and lifestock have been selectively bred for at least the
past 10,000 years, without any knowledge regarding genetic code and laws of inheritance.
Through observation and experience though ancient humans acquired sophisticated techniques in
selective breeding in practice, e.g. which we see in the differences between domesticated animals
and their wild counterparts, e.g. dogs and wolves. Motivation to solve a problem, observational
skills and intuitive thinking will have played a role in this process. Scientists, such as Albert Ein-
stein and Werner Heisenberg, have repeatedly pointed out the crucial importance of ‘intuitive’
knowing or thinking to their research and were strong advocates of its cultivation. Albert Einstein
allegedly said: “The intuitive mind is a sacred gift and the rational mind is a faithful servant. We
have created a society that honours the servant and has forgotten the gift.” (Davis, 2000)
Instead though, not only within the research culture but our culture as a whole it appears that the
scientific method is considered the most effective and thus dominant epistemology. Other ways of
knowing, such as intuitive thinking, appear ambiguous, flawed even, and less valid compared to
scientific reasoning. Even a contemporary cooking recipe with its quantitative and clear numerical
listing reminds more of the instructions for a scientific experiment then the sensual and social
activity of preparing food.
After these reflections I would regard the scientific method as merely one particular epistemology,
one way of looking at and understanding the world among many other possible ways of knowing,
some of which we make use of on a daily basis. Such a particular epistemology is shaped by our
culture, tradition and also academic conventions.
Why does it matter how we know? How do practical thinking and theoretical thinking relate? How
do action and reflection relate?
Related appears the cultural distinction between physical activity and mental activity. The latter be-
ing regarded as of higher value. Perhaps our society favours theory over practice? Practical know-
ledge is to know how to do things. Is theoretical knowledge to be able to explain how to do things
and why things are done in a particular manner?
Different ways of scientific thinking
A quote kept coming to my mind during the conference from a book by anthropologist Wade Davis’
‘Shadows in the Sun’ from 1998. There he describes how the indigenous Siona people of the
Amazon region in Brazil find and prepare a ritual drug, its preparation requiring a complex set of
procedures. In his description there are two noteworthy points. The first being that the liana, Banis-
teriopsis caapi, which is used in preparation of the drug, is regarded by modern science as one
species, while indigenous people can distinguish eighteen different varieties on sight, even from a

considerable distance. The second interesting point is the complexity of the preparation of the drug
ayabuasca. The bark of the plant is bitter and an infusion causes vomiting and severe diarrhea.
Various intricate steps are necessary to isolate Tryptamines (the hallucinogenic ingredient) and
add enzyme inhibitors, from another plant, so that the former are not denatured in the human gut
and loose their effect. The question was how, so Davis, among the tens of thousands of plants in
the Amazon region, the Siona developed such an intricate method that permits them to produce di-
fferent types of the drug with different hallucinogenic effects. (Davis, 1998, p. 162)
It was this account of indigenous ‘chemistry’ i kept thinking about. Here we had a case of ad-
vanced knowledge of chemistry, probably passed on through practice and oral tradition, likely em-
bedded within a larger explanatory mythological framework - but without what western science
would regard as a true understanding from a scientific perspective. Is this another way to practice
chemistry? However, how does this relate to the scientific method and rational thinking? What kind
of ‘research process’ was involved here?
I make two observations: The first being, that the Linnean classification system apparently fails to
capture clearly visible botanical characteristics which can easily be distinguished. Where the Siona
can distinguish between eighteen different types of liana the Linnean system seems limited as it
can only classify one. Secondly, that deep knowledge of chemistry and pharmacology apparently
can be empirically gained and passed on over generations, however in this process a type of intuit-
ive thinking initially must have played a substantial role as success through trial and error would
not lead to many satisfying results. It is not a ‘method’ that can be copied, but a ritualistic ‘way of
living’ emerging within a specific environment and growing knowledge within an organism-niche
unity.
The knowledge acquired appears to be practical know-how. What it is not, is a Western-style sci-
entific theory of the precise underlying chemical processes. The Siona may not be aware that they
isolate, what we describe as, Tryptamines and enzyme inhibitors. They may have a mythical ex-
planatory narrative for the successive steps necessary in the complicated preparation. A type of
know-how which, from a scientific perspective, would be seen as not acceptable. It is another type
of knowledge, arising from a different culture, different goals and different questions.
The shamans leading the preparation may not know how exactly the chemicals complement each
other the way they do. Yet they are experts in their preparation and will likely have an explanation,
although not one that we would consider as being rational or scientific. Yet one that makes sense
within their own culture and epistemology. To me their work reminds more of a skill, holistically em-
bedded in a way of life, rich in ritual and mythology. In this context i am reminded that the word
‘skill’ originally derived from the Old Norse ‘skil’ which meant ‘knowledge, discernment’. To do
something well implied to have knowledge.
Although their motivation, culture and worldview may be different we may still describe their activity
as being scientific in the sense of an “intellectual and practical activity encompassing the systemat-
ic study of the structure and behavior of the physical and natural world through observation and
experiment” (New Oxford American Dictionary, 2005). The activity originates from a feeling of need
or recognising a problem, relies on perception and intuition, acute observational skills, certain de-
grees of rigour, systematic study, detail orientedness, experience and experimentation. From this
process it does not appear very different after all. For both the doing is adequate behaviour within
their integrated lived experience. In a sense the shamans are scientists and the scientists have to
have something shamanistic in order to succeed.
If i compare our western approach of science or problem solving to this, I perceive the latter as im-
personal, rational and utilitarian. Although most scientists I know are passionate and emotional
about their research it is in their professional communication and research papers where it is ex-
pected of them to be objective, rational and detached.
I think it his here where the two approaches begin to differ: The scientific approach attempts to
omit the subjective qualitative experience and transform it into objective, detached language and
theory in order to make the experiment independent from the person that is conducting it. This is
part of the scientific worldview and episteme. For practice (of science) to be replicable, the meth-
ods and instructions have to be communicated in the form of text, forming a theory or hypothesis.
As pure action and lived experience it is of no scientific value. It is the steps of reflection and critic-

al thinking that transforms the, initially intuitive, experience into the recorded one of (objective) the-
ory. It is with this particular scientific epistemology where the purposes differ widely. One is sub-
jective and emotional, the other aims at being objective and rational.
What happens over time when qualitative perception is becoming secondary to those quantities
that can be measured? Lewis Mumford addressed this in ‘Technic and Civilization’:
“[W]ith this gain in accuracy [of sensory perception], went a deformation of experience as a whole.
The instruments of science were helpless in the realm of qualities. The qualitative was reduced to
the subjective: the subjective was dismissed as unreal, and the unseen and unmeasurable non-ex-
istent. Intuition and feeling did not affect mechanical process or mechanical explanations. Much
could be accomplished by the new science the new technics because much that was associated
with life and work in the past - art, poetry, organic rhythm, fantasy - was deliberately eliminated. As
the outer world of perception grew in importance, the inner world of feeling became more and
more impotent.” (Mumford, 1934, 49)
A scale ranging from objective verifiable to emotional methods was developed by Arthur Koestler
in ‘The act of creation” published in 1964. He describes his method as a continuum reaching from
the ‘objective verifiability’ of chemistry which gradually diminished as we moved to the representat-
ive arts. He emphasised that there were no sharp breaks and it was a continuum in which intuitive
and aesthetic dimensions increased.
Figure 5: “[T]here certainly is a considerable difference, in precision and objectivity, between the
methods of judging a theorem in physics and a work of art. But I wish to stress once more that
there are continuous transitions between the two.” Arthur Koestler, 1964, The Act of Creation,
p.331, 332
From the above i realise that my first two questions are related:
It appears that intuition together with empirical observation could be seen as a foundation for sci-
entific thinking and they may be intricately linked and, metaphorically speaking, perhaps different
sides of the same coin. Different ways of thinking are required for different phases of recognising a
need, framing and solving a problem and knowing when it is solved.
Phenomenologist Merleau-Ponty has written extensively about this subject: “The whole universe of
science is built upon the world as directly experienced, and if we want to subject science itself to
rigorous scrutiny and arrive at a precise assessment of its meaning and scope, we must begin by
reawakening the basic experience of the world of which science is the second-order expression.
Science has not and never will have, by its nature, the same significance qua form of being as the
world which we perceive, for the simple reason that it is a rationale or explanation of that world.”
(Ponty, 1958, ix)
Anthropologist Jared Diamond concludes his text “Guns, Germs and Steel” with a comprehensive
discussion of the methods he used in the gathering and interpreting of his data. He relates these to
general scientific approaches and compares them in detail: “People's image of science is unfortu-
nately often based on physics and a few other fields with similar methodologies. Scientists in those
fields tend to be ignorantly disdainful of fields to which those methodologies are inappropriate and

which must therefore seek other methodologies—such as my own research areas of ecology and
evolutionary biology. But recall that the word “science” means “knowledge.” (Diamond, 1998)
Psychoanalyst Carl Jung writes in this context: “Scientific education is based in the main on statist-
ical truths and abstract knowledge and therefore imparts an unrealistic, rational picture of the
world, in which the individual, as a merely marginal phenomenon, plays no role. The individual,
however, as an irrational datum, is the true and authentic carrier of reality, the concrete man as op-
posed to the unreal ideal or normal man to whom the scientific statements refer. What is more,
most of the natural sciences try to represent the results of their investigations as though these had
come into existence without man’s intervention, in such a way that the collaboration of the
psyche – an indispensable factor – remains invisible. (An exception to this is modern physics,
which recognizes that the observed is not independent of the observer.) So in this respect, too, sci-
ence conveys a picture of the world from which a real human psyche appears to be excluded—the
very antithesis of the “humanities.” Under the influence of scientific assumptions, not only the
psyche but the individual man and, indeed, all individual events whatsoever suffer a leveling down
and a process of blurring that distorts the picture of reality into a conceptual average. We ought not
to underestimate the psychological effect of the statistical world picture: it displaces the individual
in favor of anonymous units that pile up into mass formations. “ (Jung, 1958, p7-8)
Let me summarise the main ideas from above before i proceed with another more speculative ar-
gument. Above I have discussed that intuitive thinking and feeling appear to be essential parts of
the thinking required for problem solving. I further tried to demonstrate that scientific thinking re-
quires intuition, reflection, and observational skills and is not necessarily only rational and abstract.
Also that scientific practice is not necessarily conducted wearing white lab coats but its methods
can appear in different disguises, being shaped by different epistemologies, worldviews, motiva-
tions, purposes and the questions asked. Moreover I have tried to show that such knowledge of
properties, tools or practices are intrinsically human and constitute a type of ‘know how’ know-
ledge. Such knowledge will include theoretical ideas inferred from practical experience or observa-
tion which may or may not be ‘true’ or accepted by other epistemologies. However such know-
ledge or theory may serve as a tool as long as it is of value. Once we begin to reflect upon our
knowledge about our doing, and less about the doing itself, we step from the immediate experi-
ence to the mediated, from practice to reflection and theory. With the need to communicate we be-
gin to develop abstract concepts and explanatory models that we share with others through
language.
An important consideration is that a strong emphasis upon rational, objective, detached knowledge
and quantitative methods may limit the potentially diverse experience and knowledge of our world.
In my view this would result in a poorer and more constrained cultural epistemology then
necessary.
Data mining, computational methods and knowledge
The increase of accuracy of perception that Lewis Mumford was pointing to in 1934 is today some-
thing not reserved for scientists’ but something most of us will experience on a daily basis, often in
the form of satellite imagery, scientific visualisations in the media, of animations of weather fore-
casts, fMRI images of neural processes in vivo or animations of genetic processes within biological
cells. Such highly realistic animations are modelled on scientific theories, measurements and data.
In their concrete emanation they also involve imagination and aesthetic decisions in their making,
although as recipients we may not be consciously aware of those and do not question their validity
as a result of their convincing photorealism. While we believe to perceive reality, what we actually
perceive is the result of deliberate human decision making. Information, from already selective
data, is omitted while selected information is emphasised. Another aspect is that increasingly
these visualisations are the result of computing data from very large computer databases which
would have been impossible to operate with only a few years ago. These allow scientists to visual-
ise their theories in the form of highly convincing (thus indisputable?) visual facts about the world.
With such computational methods begins the discussion of my third way of knowing.
Since many years computational modelling of data and data-mining are gaining importance in the
sciences. Although the tools for handling data still are limited, the term eScience has been coined,
where “IT meets scientists”. (Gray, Jim in Hey, 2009, xviii) These new methods are software based

and gaining popularity as a result of increasing processing speed, distributed computing, through
advances in data-visualisation methods and increasing accessibility of large amounts of digital
data (‘open access’). The purpose of this type of ‘data mining’ is the automatic or semi-automatic
discovery of new patterns, anomalies and dependencies within large amounts of data. The goal of
these computational approaches is to create new knowledge in a ‘human readable format’. Anoth-
er branch of this field is about creating dynamic models from data and compute models of future
scenarios, following certain algorithmic rules. A well understood example for such a modelling pro-
cess are fractals or strange attractors in which an apparently simple yet recursive and iterative for-
mula is executed, producing unforeseeably complex results. Other approaches with more variables
are used to create meteorological modelling of weather, spreading of epidemics or the realistic ad-
aptive movement of characters in a computer game. Let me introduce some examples in more
detail.
One such digital method, algorithmic processing of data, is used by Google to offer instant transla-
tions of texts. Initially copies of identical texts in different translations are collected in a growing
database. For each language the software creates a ‘language model’ by analysing its structure.
Then various translations of the same texts are analysed by the system which develops implicit al-
gorithms that determine how one language correlates to another. “There are specific algorithms
that learn how words and sentences correspond, that detect nuances in text and produce transla-
tion. The key thing is that the more data you have, the better the quality of the system,” (Steven
Levy, 2011, p.64)
Figure 6: Google’s Translate translation service based upon heuristics. Available at http:/
/translate.google.com
The surprising fact is that there is no dictionary involved. The software does not ‘know’ the mean-
ing of the words and texts it translates. The whole translation process is based upon correlational
models between languages, especially from words which tend to appear together. The accuracy is
determined by the quantities of available data. In that sense quantity is becoming a quality in its
own right.
Google also is monitoring epidemics by analysing search queries for certain keywords. When flu
seasonally increases, there are more flu symptom related searches. The same is true for allergy-
related searches during allergy season. Not every person looking for information is actually sick,
however a pattern emerges when all flu related searches are added together. As Google already

roughly knows our geographical location this can then be related to geography (Schmidt, WSJ,
14.08.2010). Results have shown that Google Flu trends can predict surges in flu rates one week
earlier then the Center for Disease Control (CDC) (Dugas, et.al.).
Other diseases, such as Malaria, Leprosy or Tuberculosis, also are approached from a computa-
tional direction, especially by bioinformatics, computational biology and evolutionary genetics.
Here statistical methods from millions of datasets of individuals are used to identify particular sec-
tions of a genome and how these vary within a population. An algorithm allows to show how a
pathogen historically effected natural selection across the human genome and how that genome
adapted, for example in developing sickle cell anaemia. This reveals which particular genes adapt
to pathogens and allows to develop new vaccines and therapies. This very clear insights are won
by determining patterns within genomes from a very large number of people. Again it is a quantitat-
ive approach that helps to create new insights.
A variation of this process is looking not at human DNA but instead at the genetics of the patho-
gens and determining how those adapted over time as they have a comparably accelerated rate of
reproduction and mutation. This allows to find possible vulnerabilities of pathogens and directions
to target them. Another effect is that the ability of mining such massively large datasets and look-
ing at large populations through their genomes is revolutionising the understanding of evolution. It
has become much easier now to create many different hypotheses for evolutionary mechanisms
and examine relatively quickly how single genetic units have adapted over time. (Sabeti, 2007)
This computational approach allows to get deep and unprecedented insights into evolutionary pro-
cesses from a genetical perspective through mining very large amounts of data.
Yet another approach has been used to discover previously unknown chameleon habitats on the
island of Madagascar. These datasets however are not as large, in this case it is the variety of the
data from several different sources that produced new knowledge. Researchers developed a mod-
el that combined NASA satellite imagery with information of locations where biologists and natural-
ists had spotted chameleons in the past. The satellite data involved environmental characteristics
such as rainfall, cloud cover, average and seasonal temperatures and topography. GARP (Genetic
Algorithm for Rule-set Prediction), a software application for biodiversity and ecology research,
generated one hundred species-range models for each of eleven chameleon species. This model
also, unintentionally, predicted chameleons in previously unknown areas in which none had ever
been spotted. The scientists considered these results to be erroneous. Eventually though it was
discovered that these were no false-positives when seven new chameleon species were dis-
covered at the predicted locations (Raxworthy, 2003). Interestingly the scientists themselves
thought the results to be negative and did not entirely trust the accuracy of their own computational
methodology.
A similar ‘computational turn’ has been taking place in the humanities where departments for ‘digit-
al humanities’ and ‘cultural analytics’ have been founded. Here software is used as a method for a
variety of purposes. Most often this involves large scale visualisations of visual artefacts, such as
magazine covers along a timeline (Manovich, 2008). An advantage of this method is to get an im-
mediate overview of patterns in a visual form such as the use of colour printing or amount of text
used. Another approach is the analysis of large amounts of digital text posted online and investig-
ating particular phrases, for example ‘I feel’ or ‘I am feeling’ (Kamvar, et.al., 2011). Scanning a
large number of blogs and microblogs this resulted, among others, in a visual mapping of the rela-
tionships between emotional states such as anger and depression, disgust and shame, happiness
and gratitude. While some of these connections have already been studied well, the resulting map-
ping presented a number of links that have not been studied, such as the relationship between
pride and shame or the close link between beauty and ugliness. This mapping also reveals that
feelings of emptiness and frustration often appear without other associations and can be seen as
emotional ‘islands’.
An example of adaptive computational modelling is used in computer games and computer anim-
ated films. Here biomechanical models are used to animate characters which gives their move-
ments a less mechanic and more realistic appearance and which does not require human animat-
ors. The result is that virtual characters respond to their environment in adaptive and more natural
ways without requiring time-consuming manual animation. This complex behaviour arises from
simple rules. The approach involves neural networks mimicking physics, artificial evolution and

biomechanics (Reil, 2011). As the motion of a character becomes more natural over multiple simu-
lated generations of the motion-formula, the results become more and more convincing. “Miracu-
lously”, so Torsten Reil, this process actually was successful, in that the ‘formula’ for movement
resulted in the character’s movement adapting to the environment and moving in a realistic way.
The “slightly scary thing” though, so Reil, was, that it was unclear why or how the model worked
(BBC, The Secret Life of Chaos, 2010). The desired result of this process of character animation is
a short ‘formula’ for character motion which is implemented into the software of the game.
The examples above have emerged from different disciplines and use diverse approaches, some
more transparent and open to scrutiny then others. What they all share is that new knowledge is
created through computational methods. Even if results are unexpected or ‘miraculous’ the work-
ings of the software applications and the data sets are well understood. From that perspective they
belong to the history of technological devices that helped to solve problems such as steam en-
gines, pulleys and gears. I think in this case the software is not thinking ‘better’ then a human but it
can operate faster. It is following rules that have been thought through by humans and will only
work as well as its program and data sets permit. Occasionally results may surprise their makers,
but most likely within a clear range of possibility. Software may discover unusual stars from satell-
ite data or it may animate a game character highly realistically. Yet it will not surprise us beyond
this framework by a unique creative act such as writing a musical symphony. All possible results
will be within a predefined range, specified in clear and well known instructions. Software applica-
tions cannot be independently creative even though the results that are discovered within the data
are unexpected. Is it not a sign of true research and experiment to make unexpected findings fol-
lowing a particular methodology?
If correlations imply to observe ‘a mutual relationship or connection, in which one thing affects or
depends on another’ (Oxford American Dictionary, 2005) it could be seen as a know-how ap-
proach to knowledge, in which patterns emerging together appear to be related to each other,
even if the causal relationship is not immediately obvious.
If know-how involves getting something done without having to understand why it works, then
these computational methods that allow us new applicable knowledge remind more of practical
know-how then the know-why which results from the traditional methods and principles of science,
which aim at understanding underlying deeper causes.
Knowing without understanding
Know-how in the sense that we naturally acquire and apply knowledge without a rational scientific
model for understanding this knowledge. Some examples include the use or invention of medicine,
yeast, baking bread, cement, pottery, selective breeding of livestock and crops, brewing alcohol,
among many others. Humans have made use of these practices for thousands of years without re-
quiring a scientific explanation or model to understand their underlying causes. Yet they recogn-
ised patterns over time and knew how-to make use of them. Our culture however appears to re-
gard the scientific approach as the only valid approach and dismisses other/older approaches as
unreliable.
I think that there are similarities to this ‘know how’ approach to be found in our engagement with
some of today’s technologies. The concept of knowing-without-understanding extends to everyday
appliances which we can use without necessarily understanding how they function, such as tele-
phones, computers, microwave ovens or combustion engines. Their ease of use does not require
understanding their inner workings. In case they do not function we invent explanatory models and
intuitive actions, which may turn into skilled behaviour and knowledge to remedy the situation. We
intuitively think of an action which may solve a computer problem. A friend of mine carries a heavy
rock in her (very) old car which she uses to hit on the carburettor in case the ignition does not start.
This method may sound inappropriate and unreasonable for a car mechanic, however it proves
highly effective and reliable in everyday circumstances.
For example, the engineer Otto Lilienthal (1848-1896) made sketches of bird’s wings and used the
principles he observed for the design of a hang-glider in 1891. He undertook over 2000 successful
flights without understanding that it was ‘lift’ caused by air flowing at different speeds that allowed
his glider to fly (Nature Tech: The Magic of Motion, 2006). Earlier examples would include the use

of yeast, the baking of bread or brewing of beer. How did the ancients discover that grain ferment-
ing in water would create alcohol over time? Who would dare to drink this concoction to discover
its intoxicating qualities? The same must be true for the ancient romans who could not have known
how the hydraulic properties of cement components chemically functioned, yet they were experts
in using it to maintain crucial parts of their infrastructure such as viaducts. Today we understand
the facts behind yeast, cement and fermentation in great detail, but we see that know-how is useful
without knowing why.
Today we use USB memory sticks to save digital data without being aware that these make use of
‘quantum tunneling’ effects to store information. Here a minority of experts are able to apply theor-
etical knowledge into practical devices, which can be produced and used without much skill or
know-how by anybody. The same will probably be true for many of the technological artefacts we
use on a daily basis. We know how to use them without the need to understand how they function.
We rarely even reflect on this lack of knowledge, except if the device fails to function as expected
and we are trying to solve this problem. It is then that we intuitively develop explanations and
strategies for lack of true understanding.
Another area in which unconscious knowledge or intuition are crucial is in protein folding. Proteins
are compounds of polypeptides, chains of amino acids, that facilitate particular biological functions,
for example enzymes. They are part of our body’s metabolic functions, in spider-silk and in the
memory of our brain. The correct shape, or ‘folding’, of a protein determines its functionality. Com-
puter software is notoriously unsuccessful in determining the most successful structure for pro-
teins. Often this requires supercomputers or distributed computing efforts, run on multiple com-
puters. In 2008 Biochemist Firas Khatib from the University of Washington developed the software
‘Foldit’ which turned the folding of the structure of a virtual enzyme on the computer screen into a
game. Players can download the software from the website and compete against each other by
folding new structures of known proteins. The goal was to see if “human intuition could succeed
where automated methods had failed” (Moskvitch, 2011). The results were very surprising in that
people were more successful in this tasks then computational approaches, perhaps, so Khatib, as
a result of highly developed human spatial thinking.
Conclusions
In this text i have reflected on a number of different ways of thinking and approaching problems. I
used examples from design thinking, traditional scientific thinking, indigenous sciences and com-
putational problem solving to reflect on the roles of experience, unmediated perception, subjectivity
and objectivity in the problem solving process. With diverse examples i reflected upon that there
are different ways of thinking and that our culture over-emphasises rational reasoning and scientif-
ic thinking without us being fully aware that this merely is one perspective.
Initially the subject of reflection was designerly thinking vs. scientific thinking and intuitive thinking
vs. scientific thinking. During the reflective process i understood that these are not opposed to
each other but intrinsic phases of an inquisitive creative process which requires intuition as well as
more rigourous and systematic modes. While science is a discipline that aims at creating know-
ledge about the world as it is, designing is an application of principles to create something original
and thus change the world as it is. Both fields, i tried to demonstrate, require intuitive thinking as
well as practical action and observation.
Design, like science, are both grounded in perception as both observe in order to understand. Sci-
entists developed technologies and instruments to perceive with ever higher precision, ‘looking’
deeper into their matter of investigation to experiment and construct theories (with perhaps practic-
al application). Designers observe with a sense for opportunity and practical application. They may
study shape, motion and material properties in order to apply the learned principles to a design.
Scientists observe with scrutiny in order to understand the underlying rules and mechanisms. De-
signers observe with a sense of possibility in order to apply the learned to other areas.
Nevertheless our society values scientific thinking higher then intuitive thinking. Perhaps the reas-
on for this is that we do not properly understand the crucial importance of intuitive thinking and
how it works?
Intuition is a particular kind of knowledge, however just not very verbal and articulate. For example

the tacit knowledge we gain through experience and immediate engagement. It is reflection that
mediates some of this experience into teachable knowledge and perhaps a theory.
Another thought was ‘what if we did not really understand that which we know’ and arrived at the
insight that this perhaps is the natural human state of mind: We observe and experience and de-
velop satisfying explanatory models that make sense to us and in this process we believe to ‘un-
derstand’. However there is not one single ‘truth’ to understand and different models are valid only
for as long as they are valid. Although there are clear distinctions to make in the world these may
result in many different theories and explanatory models. An example of this were Linnean classi-
fication and Siona distinctions. Both schemes serve different purposes and solve different prob-
lems, resulting from different environmental perspectives upon the integrated lived experience.
These experiences of the world are mediated into knowledge and theory, and considered valid as
long as they apply and are repeatable. In that sense we can apply know-how without a deep un-
derstanding from a scientific perspective, or a know-why. Much of our knowledge of the world
could be seen as know-how knowledge. We know how to get things done, yet we do not have ac-
curate knowledge of inner workings.
In that sense the evolving data-mining methods, based on clearly understood software operations
and structured databases, could be seen as nothing radically new but instead as a phenomenon
and a return to our natural creation of knowledge: Software applications mining selective data
about the world in order to distinguish patterns where we could not (at least within reasonable
time-scales), and visualising these patterns and relationships in visual formats. Now it could be up
to human ingenuity to construct theories and explanations from these new empirical natural
processes.
I think that at the heart of this enquiry lies an epistemological question. What it is to ‘know’ and
what it means to ‘understand’, if we do not know why our knowledge is true? Perhaps we cannot
truly and fully understand the basic laws and processes of our world but merely use ‘models’ of in-
ference which function and provide satisfying explanations as long as these are useful?
Note:
The writing of this text has been inspired by the ideas of several other authors. David Weinberger’s
“Too big to know”, Chris Anderson’s “The End of Theory: Will the Data Deluge Makes the Scientific
Method Obsolete?” published by Wired Magazine in June 2008. The text included detailed docu-
mentation of 18 examples. It was subsequently republished by John Brockman’s EDGE blog and
there discussed by a number of public figures. Also relevant were a talk about the knowledge of in-
digenous cultures by anthropologist Wade Davis at the Long Now foundation in 2010, and a key-
note talk by Terry Irwin titled “The seduction of certainty” given at the New Views 2 conference in
July 2008 in London.
Acknowledgements:
I wish to thank my reviewers Bernice Goldmark and Phillip Guddemi for their thoughtful and con-
structive comments upon reading a draft of this text. They suggested rephrasing to make ideas
clearer and made me aware of several misconceptions. Their input gave the paper a new direction
and I am very much indebted to both of them.
To be published in Autumn 02012 at Edition Echoraum, Vienna
http://www.echoraum.at/edition/neuersch.htm
References
Anderson, Chris (2007), The End of Science: Will the Data Deluge make the Scientific Method
Obsolete?, Wired Magazine No 16, 2007
Cross, Nigel (2007), Designerly ways of knowing, Birkhäuser, Basel, Switzerland
Sabeti PC, Varilly P, Fry B, Lohmueller J, et.al. (2007), Genome-wide detection and characteriza-
tion of positive selection in human populations, Nature. 2007 Oct 18;449(7164):913-8.

Davis, Wade (1998), Shadows in the Sun: Travels to Landscapes of Spirit and Desire,
Dugas AF, Hsieh Y-H, Levin SR, Pines JM, Mareiniss DP, Mohareb A, et al. Google Flu Trends:
Correlation With Emergency Department Influenza Rates and Crowding Metrics. Clin Infect Dis [In-
ternet]. 2012 Jan 8; Online first. Available from: http://cid.oxfordjournals.org/content/early/
2012/01/02/cid.cir883 accessed
Hey, Tony, Tansely, Stewart, Tolle, Kristin Eds. (2009), The Fourth Paradigm: Data-Intensive Sci-
entific Discovery, Publisher: Microsoft Research 2009
Ingold, Tim (2000), The Perception of the Environment : Essays on livelihood, dwelling and skill,
Routledge, London
Kamvar, Sepandar, Harris, Jonathan (2011), We feel fine and searching the emotional web, in
WSDM '11 Proceedings of the fourth ACM international conference on Web search and data min-
ing, ACM New York, NY, USA available at http://kamvar.org/assets/papers/wefeelfine.pdf
Manovich, Lev (2008), Cultural Analytics, Analysis and Visualization of large cultural data sets,
available at http://lab.softwarestudies.com/2008/09/cultural-analytics.html
Moskvitch, Katia, 2011, Online game Foldit helps anti-Aids drug quest http://www.bbc.co.uk/news/
technology-14986013
Mumford, Lewis (1934), Technics and Civilization, Routledge & Kegan, London
Natures Tech: The Magic of Motion (2006), [TV program], ARTE/ORF, 3.-5. March 2008
New Oxford American Dictionary (2005), 2nd Edition, Version 1.0.2, Apple Computer, Inc
Raxworthy, Christopher J., Enrique Martinez-Meyer, Ned Horning, Ronald A. Nussbaum, Gregory
E. Schneider, Miguel A. Ortega-Huerta & A. Townsend Peterson (2003), Predicting distributions of
known and unknown reptile species in Madagascar, Nature 426, 837-841 at http:/
/www.nature.com/nature/journal/v426/n6968/full/nature02205.html accessed March 2012
Reil, Torsten 2011, Slide presentation at ‘Casual Connect 2011’ conference, at http://www.slide-
share.net/naturalmotion/torsten-reil-casual-connect-keynote-2011 accessed March 17 2012
Russell, K. (2002) Why the culture of academic rigour matters to design research: or, putting your
foot into the same mouth twice. Working Papers in Art and Design 2, Accessed accessed January
27th 2011 from http://sitem.herts.ac.uk/artdes_research/papers/wpades/vol2/russellfull.html ISSN
1466-4917
Schmidt, Eric, interview in Wall Street Journal on August 14 2010 accessed January 27th 2011
from http://online.wsj.com/article/SB10001424052748704901104575423294099527212.html
The Secret Life of Chaos, 2010. [TV program], BBC, BBC 4, 14. January 2010, 21:00
M. J. de Vries, N. Cross and D. P. Grant (Eds.) 1993, Design Methodology and Relationships with
Science. Dordrecht: Kluwer Academic Publishers.
WEINBERGER, D. (2011). Too big to know: rethinking knowledge now that the facts aren't the
facts, experts are everywhere, and the smartest person in the room is the room. New York, Basic
Books.
1. Persian recipe by Madame Benshaw given to Richard Sennett. Here Sennett’s own
notes: ‘‘Your dead child. [the chicken] Prepare him for new life. [bone] Fill him with the
earth. [stuff] Be careful! He should not over-eat. [stuff lightly] Put on his golden coat.
[brown before baking] You bathe him. [Prepare the poaching liquor] Warm him but be
careful! A child dies from too much sun. [cooking temperature: 130 Celsius] Put on his
jewels. [once cooked, pour the sweet-pepper sauce] This is my recipe.’’ Many Persian
recipes, I’ve since learned, are couched in such poetic language.” p. 190