Content uploaded by Ilkka Tuomi
Author content
All content in this area was uploaded by Ilkka Tuomi on Jun 13, 2024
Content may be subject to copyright.
Futures Foresight Sci. 2019;e11. wileyonlinelibrary.com/journal/ffo2
|
1 of 15
https://doi.org/10.1002/o2.11
© 2019 John Wiley & Sons Ltd
1 | INTRODUCTION
As illustrated in this paper, all human action is based on anticipa‐
tory models, and imagined futures play a central role in human
sense‐making and in policy and strateg y development. Yet, time
and anticipation have remained poorly conceptualized also in
future‐oriented disciplines and practices. In the domain of fore‐
sight practice, this has lead to a growing interest in anticipatory
systems theory (Louie, 2010; Rosen, 1985) as a possible foun‐
dation for a discipline of anticipation and futures studies (Fuller,
2017; Poli, 2017), and related calls for building competences in
using the future (e.g., Miller, 2015, 2018b). We contribute to this
agenda by linking anticipatory systems theory with two influential
approaches in foresight and futures studies, and by contrasting
these with a third emerging “constructivist” and design‐oriented
approach. We focus on the different underpinning “chrono‐
topes” of these three alternative approaches, and show how their
characteristics make dif ferent types of futures visible and pos‐
sible. An important claim of this paper is that the constructivist
approach to foresight is able to address qualitative novelty and
Received:21November2018
|
Revised:15Januar y2019
|
Accepted:15Januar y2019
DOI: 10.10 02/f fo2.11
ORIGINAL ARTICLE
Chronotopes of foresight: Models of time‐space in
probabilistic, possibilistic and constructivist futures
Ilkka Tuomi1,2
1Meaning Processing Ltd, Helsinki, Finland
2Stellenbosch Institute for Advanced Study
(STIAS), Wallenb erg Rese arch Centre at
Stellenbosch University, Stellenbosch, South
Africa
Correspondence
Ilkka Tuomi, Meaning Processing Ltd.,
Helsinki, Finland
Email: ilkka.tuomi@meaningprocessing.com
Abstract
The concept of chronotope was introduced in the 1920s by the Russian neurophysi‐
ologist A.A. Ukhtomsky, and extensively used by Mikhail Bakhtin in his analysis of the
development of literary forms. Chronotope structures the possibilities for meaning‐
ful action and different chronotopes thus generate different forms of agency and
future. In this paper, three approaches to foresight are analyzed, showing how their
different chronotopes lead to different ways of understanding the future. We dif‐
ferentiate between probabilistic, possibilistic and constructivist frameworks for fore‐
sight. Probabilistic approaches are shown to rely on recursive chronotopes that
capture future as a repetition of the past. Possibilistic approaches, here exemplified
by the “gold standard” Schwar tz/GBN scenario method, are shown to rely on narra‐
tive chronotopes that can tell stories of emergent futures and the impact of innova‐
tion. Scenario methods, however, describe changes in the environment as forces and
trends in a recursive chronotope. As a result, they have limited capacity to address
qualitative novelty. In contrast to possibilistic and probabilistic approaches, a con‐
structivist dialogical approach described in this paper explicitly aims at integrating
qualitative novelty and radical innovation as important elements of foresight.
Constructivist foresight does not aim for “knowing” the future; instead, it aims at
creating the future. Knowing when to use these different forms of foresight is an
important element in strategy and policy development.
KEYWORDS
anticipatory systems, constructivist foresight, design‐oriented policy development, ethics of
foresight, forecasting, scenario methods, unpredictability
2 of 15
|
Tuomi
radical innovation in a theoretically justified and practically imple‐
mentable way. As considerable terminological and conceptual va‐
riety underpins the literature and practice of anticipatory methods
(Bishop, Hines, & Collins, 2007; Bradfield, Wright, Burt, Cairns, &
van der Heijden, 2005), a simplified characterization of these ap‐
proaches is necessary. We argue that chronotopic analysis high‐
lights key differences among common foresight methods and that
it shows how the constructivist approach can effectively address
important theoretical and practical challenges, such as unpredict‐
ability and ethics of foresight.
Mikhail Bakhtin made the concept of chronotope a central ele‐
ment in his studies of the historical development of narrative forms
(Bemong et al., 2010; Holquist, 1990; Keunen, 2011; Morson &
Emerson, 1990). Bakhtin is now best known as a literary theorist.
He was, however, motivated by the need to understand the ethics of
human action and agency, and how action can become responsible
(Clark & Holquist, 1984; Steinby, 2013). Bakhtin synthesized the his‐
tory of Western literature to show how different historical forms of
narrative embed different assumptions about time and space, cau‐
sality, and human actors that make change happen. Each historical
form of chronotope carries with it a unique model of causality and
agency, enabling and limiting the t ypes of meaningful stories that
can be told in its context.
A study on chronotopes that underpin dif ferent forms of fore‐
sight practice is useful as it makes explicit important assumptions
about dynamics of change. Chronotopes form the basic time‐space
organization that makes meaningful accounts of causalit y and
agency possible, and they also define what kinds of stories can make
sense.
In this paper, we describe three essentially different chronotopic
structures. The first of these can be called Newtonian. In this chro‐
notope, time and space are independent and absolute. In Kantian
terms, they are transcendential conditions that provide the possibil‐
ity for knowing the world. Building on Rosen’s (1978b, 1978c, 1985)
work on anticipatory systems, we show that in this chronotope fu‐
ture is recursively determined by the past. A unique characteristic of
systems that can be accurately modeled in the Newtonian chrono‐
tope is that they have a single set of “state variables” that completely
describe the dynamics of the system.
The second chronotopic structure can be called a narrative chro‐
notope. As Bakhtin scholars have mainly focused on literary theory,
this is the m ost common interpr etation of Bakhtin ian chronotopes. I n
particular, in his famous essay, Forms of Time and of the Chronotope
in the Novel, Bakhtin (1981) studied how different narrative forms
have tried to capture time, agency, and change from early Greek my‐
thologies to the 19th century polyphonic novel. In this view, chro‐
notopes provide the foundation for understanding experience and
the world, each chronotope creating different meaningful realities.
Chronotopic analysis, thus, can be a tool that reveals implicit models
of causality and agency embedded in narrative accounts. We analyze
the chronotopic structure of the “gold standard” Schwartz/GBN sce‐
narios (Schwartz, 1991), showing that also scenario methods rely on
recursive Newtonian chronotopes.
The third chronotopic structure that we describe in this paper
can be characterized as “constructive.” It returns back to Bakhtin’s
(1990, 1993) early writings where he focused on the nature of re‐
sponsible and “answerable” acts. In contrast to Bakhtin’s best‐known
texts that can easily be read as purely literary scholarship, these
texts reveal sophisticated philosophical thinking. In these partially
preser ved texts, theoretical and conceptual modes of knowing are
linked with aesthetic and ethical forms of knowing. In the “construc‐
tivist ” interpretation, a chronotope is a constantly evolving ground
for experience, and each unique act productively participates in the
non‐repeatable and unique process of Being. Each actor occupies
her own place in reality and from this time‐space position partici‐
pates in a dialogical process with the nature and other people that
creates yet‐to‐be futures. In the spirit of these early texts, we inter‐
pret chronotopes as the foundation for creating new realities.
Historically, the concept of chronotope was introduced in 1925 by
the Russian neurophysiologist Aleksei Ukhtomsk y, who argued that
we perceive the world as anticipations of its future (Chebanov, 2015;
Sokolova, 2015). We claim that the constructivist interpretation is
closely aligned with Ukhtomsk y’s idea of chronotope as a biologic al
structure that defines how living organisms can perceive and ob‐
serve their environment and act upon it . Although Bakhtin's concept
of chronotope has deep roots in the works of leading 20th century
phenomenologists and ethical theorists, including Scheler, Bloch, and
Hartman (Poole, 2001), we adopt a more biology‐grounded cognitive
interpretation of chronotope that can also be linked with Bergsonian
ontology and enactivism, socio‐cultural theories of cognition, and
Rosen's work on anticipatory and living systems (Tuomi, 2017). We
also follow Ukhtomsky in interpreting human chronotopes as social
and cultural phenomena. Indeed, as spatial relations have been an
important focus of sociology since Durkheim (1933), it would be pos‐
sible to interpret much of existing sociological theory from a chro‐
notopic point of view. As social spaces are produced in a historical
process that becomes represented and cr ystallized in material struc‐
tures, architectures, and also collective and organizational memo‐
ries, a contemporary interpretation of constructivist chronotopes
could also be linked with, for example, anthropological studies on
institutional memory (e.g., Douglas, 1987), distributed cognition (e.g.,
Hutchins, 1995), and studies on time‐space geography or the evolu‐
tion of social spaces and their social visibilities and accessibilities (e.g.,
Hillier & Hanson, 1984; Holloway & Kneale, 200 0).
In the last two decades, narrative analysis has become an im‐
portant starting point in organizational studies and strategy re‐
search (e.g., Barry & Elmes, 1997; Boje, 20 01; Czarniawska, 1997;
Engeström, 1995; Vaara, Sonenshein, & Boje, 2016), and also
Bakhtinian chronotopes have been used to understand how past,
present , and future are constructed in strategy narratives (e.g., Vaara
&Pedersen,2013).In thefield offuturesstudies, Jar va (2014)has
noted the potential of narrative approaches and chronotopes, and
Aaltonen (2007; Aaltonen & Holmström, 2010) has located different
foresight methods in a Bakhtin‐inspired “chronotope space,” where
system linearity, complexity and disruptiveness provide the under‐
pinning dimensions. Although the discussion in this paper is aligned
|
3 of 15
Tuomi
with these earlier studies, in line with Ukhtomsk y we maintain that
all action is oriented toward the future, and in line with Bakhtin,
that all action is “authored” in relation to its “emotional‐volitional”
content and expectations of it s impact. Action, thus, can be inter‐
preted as enacted prediction. The space and time of action forms an
integrated whole, and this whole is expressed in the spatio‐temporal
structure of anticipatory models that guide and enable experience,
observation, imagination, and action. We therefore use the con‐
cept of chronotope in a distinct way, in the context of anticipator y
systems theory. In this context, chronotope is not only a diagnostic
instrument or a unit of analysis that can be used to describe the nar‐
rative structure of existing organizational and policy stories; instead,
it charac terizes the time‐space organization of anticipator y models
that motivate present action and make it meaningful. Anticipatory
models can use different chronotopes, and different chronotopic
structures are able to generate and anticipate dif ferent imaginary
futures. Choosing a different chronotope, thus, creates different fu‐
tures and new possibilities for action.
1.1 | Structure of the paper
In the following sections, we describe and compare three different
chronotopic structures that underpin different approaches to fore‐
sight practice. We st art by clarify ing the concept of chro notope, illus‐
trating it with Bakhtin’s analysis of three early literary chronotopes.
Section 3 then describes the chronotope of probabilistic futures.
“Probabilistic” in this context does not imply statistical methodol‐
ogy: In the probabilistic approaches the future is understood as a set
of possible future st ates, some of which are considered more prob‐
able than others. A key aspect of probabilistic foresight and fore‐
casting is that—following dynamical system models from physics—it
conceptualizes systems as trajectories in a state space. Different
trajectories lead to different endpoints and some of these may be
more probable than others, contingent to possible interventions. As
discussed in Section 3, this conceptualization is rooted in the dif‐
ferential Newtonian formalism, and it can only encode very special
forms of entailment and causality.
Section 4 describes the chronotope of possibilistic foresight,
using the Intuitive Logics scenario approach popu larized by Schwart z
and the Global Business Network as an example (Schwartz, 1991;
Wilkinson & Kupers, 2014). In Intuitive Logics, narrative fragments
of possible futures play a central role. As discussed below, many
alternative chronotopes are available for scenario development.
We show that narrative chronotopes are mixed with a recursive
Newtonian chronotope in Intuitive Logics scenario development.
This analysis, therefore, is also aligned with Derbyshire and Wright
(2017), who argue that Intuitive Logics scenarios are grounded on
Newtonian models of causality.
Section 5 introduces a constructivist approach to foresight.
Whereas probabilistic and possibilistic approaches have been ex‐
tensively used in foresight and anticipator y practice, elements of
the constructivist approach such as policy experiments and design
thinking have only recently gained visibility. The constructivist
approach is illustrated using a “design‐oriented next‐generation”
foresight model developed for the European Commission’s European
Forum on For ward‐Looking Activities exper t advisory group
(Tuomi, 2013). The construc tivist approach is based on a distinc‐
tively ontological approach to foresight and it views the future as
something that is created. In this regard, it is aligned with Gaston
Berger’s emphasis on “prospective,” in contrast with “retrospective”
futures (Cournand & Lévy, 1973). It also adopt s the Bakhtinian idea
that new realities can emerge in a dialogical interac tion in the con‐
text of existing chronotopes. The constructivist model thus imple‐
ments a multi‐voiced process of futures design that links foresight
with models of responsible action and innovation.
Section 6 contrasts the distinguishing characteristics of these
three dif ferent approaches to foresight, in effect summarizing the pre‐
vious discussion. Finally, Section 7 makes some concluding remarks.
2 | THE CHRONOTOPE
Bakhtin used the concept of chronotope in many different ways in
his writings. The proper interpretation of the concept, therefore, has
been extensively debated by Bakhtin scholars. In this paper, we inter‐
pret the concept in a way that links human cognition, anticipation, ac‐
tion, and choice. A specific chronotope defines a structure for what is
achievable, accessible, and proximal from the point of an actor, defin‐
ing how agency and future can be expressed. In mathematical terms, a
chronotope could be associated with a topology or a metric space. In
this interpretation, a chronotope makes some futures possible with lit‐
tle effort, and different chronotopes make dif ferent futures possible.
In general, time in a chronotope is not an abstract dimension; instead,
it is a necessary medium that allows natural events and human activity
to leave concrete traces in the world. This chronotopic time is a gen‐
erator of history as a process of change and emergence.
In The Forms of Time and of the Chronotope in the Novel,
Bakhtin points out that in the different periods of human history
people have lived in different concrete chronotopes, and these have
been reflected in their social life and the stories they tell. Time and
space in the agrarian world was tightly bound with the natural pro‐
cesses of birth, growth, reproduction, and daily and yearly cycles.
Agrarian folkloristic time is a time of collective labor, productive
growth, and deeply grounded in the concrete location. Time is social,
infinite, but also cyclical. Because of this, it does not have a clear be‐
ginning, a unique present, or an end: “The passage of time does not
destroy or diminish but rather multiplies and increases the quantity
of valuable things; where there was but one seed sown, many stalks
of grain appear; the new issue always eclipses the passing away of
individual specimens...Perishing and death are perceived as a sowing,
after which flows increase and harvest, multiplying that which had
been sown.” (Bakhtin, 1981, p. 207).
Time is also profoundly spatial and concrete. It is not separated
from the nature, and human life and nature are perceived in the same
categories or growth, flourishing, multiplication, and reproduction.
The generative capacity of this agrarian time, however, has a major
4 of 15
|
Tuomi
limitation: it is cyclical. Because of this, growth remains multiplica‐
tion and expansion of the same, and cannot achieve true “becoming”
as something truly emergent and creative.1
This folkloristic chronotope forms Bakhtin’s starting point for
studying the development of different forms of chronotope that
became influential in the European literary history. In the following
sections, we include in this literary history also scientific narratives.
To make the abstract concept of chronotope more concrete, it is,
however, useful to briefly revisit Bakhtin’s descriptions of major
chronotopes that structure time, space, and agency in ancient Greek
romances.
2.1 | Bakhtinian chronotopes of ancient novel
Bakhtin found different chronotopes in different historical peri‐
ods of literary history. He started with Greek novels, written dur‐
ing the Roman Empire in the first four centuries CE. Chronotopes
are not explicitly described in written texts; instead, they form
the background that makes meaningful interpretation possible.
Chronotope defines the ways in which events, action, and agency
can be described as meaningful elements of a story, as well as
what counts as an event, action or agency. Literature offers many
different chronotopes, and as a result of long cultural‐historical
development, we can now deploy many alternative time‐space
structures in understanding the present and its dynamics of
change.
Bakhtin suggested that the ancient novel developed three
major chronotopes. In “the adventure novels of ordeal,” written
between the second and sixth centuries CE, the chronotope con‐
sists of “an alien world in adventure‐time.” (Bakhtin, 1981, p. 89)
In these Greek romances, the adventures leave no trace, and af‐
fect nothing. The heroes and heroines do not change, mature, or
even age as a result of their adventures. There is no inherent rea‐
son why the string of adventures and surprising events could not
have occurred in a different order. According to Bakhtin, time in
Greek romances, therefore, is “reversible.” Time is a crucial factor
in this literary genre, but not because it would underpin change,
becoming, “Bildung,” or emergence, as in the 19th century novel; it
is important because rescues and critical turning‐points occur at
the last possible moment. Thus, Bakhtin notes, this genre is char‐
acterized by the use of expressions such as “suddenly” and “at just
that moment.” As Morson and Emerson put it;
This is a world in which simultaneity, random con‐
tingency, miraculous coincidence, and sheer change
play a key role. Wars happen unexpectedly and with‐
out apparent cause at crucial moments; storms come
from nowhere to cause fatal shipwrecks. ...nothing
that happens to the hero and heroine in any way
shapes or reflect s the historical process... That is one
reason why adventuristic plot s are so easy to adapt by
authors of different countries and eras and so difficult
to date accurately. (Morson & Emerson, 1990, p. 378)
In the Greek romance, time is reversible or repeatable and also
place becomes abstract and undifferentiated. A sea is needed for a
shipwreck, but which sea makes no difference. Surprising events and
incredible encounters require an alien country, but any alien country
will do. The adventure chronotope is thus characterized by a technical
and abstract connection between space and time, by the reversibility
of moments in temporal sequence, and by their interchangeability in
space. What happens in Babylon could as well happen in Byzantium.
In the Greek romance, the hero thus travels in alien time‐spaces
where almost anything can happen. One thing that stays the same,
however, is the hero himself. Events happen to the hero but his ac‐
tions have no influence on them. As Bakhtin notes, in this chrono‐
tope individuals are completely passive and the actions of heroes
and heroines “are reduced to enforced movement through space (es‐
cape, persecution, quests); that is, to a change in spatial location”
(Bakhtin, 1981, p. 105).
Another form of chronotope can be found in what Bakhtin calls
“the adventure novel of everyday life.” An example of this is in The
Golden Ass of Apuleius. While tr ying to perform a spell that would
metamorphose the protagonist Lucius into a bird, he accidentally
turns into an ass. This leads to a long chain of adventures before
finally a goddess gives Lucius advice how to become a man again.
In the chronotope of The Golden Ass, the chronotope of adven‐
ture time is still strongly present. The logic of change, however, is
subordinated to another higher logic, where the hero’s adventures
become elements in an overarching stor y of the protagonist ’s crisis
and redemption. The hero at the beginning of the stor y is different
from the one at the end, and change is represented as a process of
metamorphosis that first makes Lucius an ass and, after all the ad‐
ventures, human again. At the end, Lucius is a different man. Time,
therefore, becomes irreversible.
A third type of chronotope can be found in ancient biographies
and autobiographies. Bakhtin argued that in the early antiquity there
was no difference between autobiographies and biographies. This
was because the characteristics of an individual were assumed to
equal his publicly visible characteristics. There was no concept of
“private” person as we may now understand it, and no “internal
perspective” or “inner self ” that could have differentiated autobiog‐
raphies from biographies. In this chronotope of the public square,
an individual was understood to be totally open to public gaze,
and nothing in him could avoid being subject to public control and
evaluation.
As Bakhtin argued, the development of literary chronotopes re‐
flect s the development of concrete chronotopes that cultures use to
make sense of their evolving realities. This development, however,
is not about a linear progress; all the historical chronotopes are in
constant use, whether in the search of a Lost Ark, in the fights with
aliens in space and on Earth, or in the public squares of social media.
In particular, they underpin three common but different ways of un‐
derstanding future in foresight.
In the next section, we focus on a chronotope that has become
extremely important during the last four centuries. This is the
Newtonian chronotope that underpins almost all mathematical and
|
5 of 15
Tuomi
scientific descriptions of the world. The Newtonian chronotope or‐
ganizes time and space in a very particular way, where only some
forms of agency and causality can be expressed. Because of this,
only a limited spectrum of possible futures can be modeled in this
chronotope.
3 | THE CHRONOTOPE OF PROBABILISTIC
FUTURES
Probabilistic approaches perceive futures as a set of trajectories that
lead to future states of the world, each with a probability of being
realized. Probabilistic futures are generated with tools such as struc‐
tural equations used in economics; dynamical simulation models
such as those used in the climate models and policy impact analysis;
time‐series analysis, used, for example, in econometrics, marketing
and algorithmic trading; and supervised and unsuper vised learning
models in machine learning. In all these, futures are forecast based
on historical data.
Most probabilistic approaches interpret the system under study
as a continuous or discrete dynamical system. Dynamical systems
have been used to describe a vast amount of phenomena in many
different domains of life, ranging from celestial mechanics to eco‐
nomics, chaotic systems, and biological neural networks. Due to
the unquestionable successes of this approach in physics, it is rarely
noted that this description implies very specific assumptions about
time, space, and causality. A formal analysis of these assumptions
requires that we move beyond the mathematics used to model dy‐
namical systems and use a more abstrac t description where the
structural characteristics of the mathematical formalism itself c an
be described.
A suitable formalism is known as category theory, originally
developed in the 1940s to study structure‐preserving mappings
between different domains of mathematics (Awodey, 2010; Mac
Lane, 1978). Category theory also provides the formal foundations
for anticipatory systems theory, as developed by Rosen (cf. Louie,
2009). In the present context, it suffices to note that category the‐
ory makes it possible to study the question what kinds of natural
systems c an be mapped to dynamical systems so that their observ‐
able characteristics and dynamics are preserved (Rosen, 1978a).
From a practical point of view the question is very simple: What
types of economic al, social, political or innovation processes can
be described using dynamical systems models without losing their
essential characteristics. Rosen has shown that such descriptions
are possible for a special class of natural systems where a single
set of “state variables” can completely describe trajectories of
system change. In general this is not possible, and many alterna‐
tive and incompatible descriptions are necessary for systems that
Rosen calls “complex.” Algorithmic systems and dynamic al systems
are the most important examples of systems that are not complex,
and this has important consequences on how their dynamics and
change can be captured. The next section briefly reviews the very
basic elements of classical Newtonian physics from this point of
view; these elements are often considered to be defining char‐
acteristics of scientific models and prediction, and thus also pro‐
vide the foundation for many areas of economics and other social
sciences.
3.1 | Recursive systems: Future as
repetition of the past
Dynamical systems have a very unique chronotope. There are many
mathematical formalisms that can be used to define dynamical sys‐
tems, but the most influential one is based on Newtonian mechanics.
In such a dynamical system, change in the system st ate is determined
by the current system state and external “forces” that encode the
impact of the environment. System state, itself, is defined by a set
of coordinate values that represent a point in a “st ate space.” In the
case of Newtonian particles, the state space is called “phase space,”
and the “location” in this space is fully determined when particle’s
position and its change rate, i.e. velocity, are given. Everything that
can be known about a Newtonian particle is known when its loca‐
tion in this state space and some constitutive parameters are known.
In the case of a Newtonian particle, only one constitutive param‐
eter is needed to describe how the “environment” couples with the
particle, and this parameter is called its “mass.” When mass is used
to scale velocity, as is commonly done in physics, a location in the
phase space fully determines everything that can be known about
the particle.
The Newtonian formalism relies on the fact that future can be
predicted if we know the position of a particle, the rate of change of
its position, the rate of change of this change, and so on until infin‐
ity. Newton’s great insight was that this infinite sequence can be cut
after two of its first terms if we assume that without environmental
impact particles stay in constant motion. The effec ts of the environ‐
ment, or the “forces” that influence change, can then be defined as
the generators of acceleration, or change in motion. In mathematical
terms, a Newtonian force is defined as acceleration multiplied by a
constant that represents the “reactivity ” of the par ticle in question.
This leads to Newton's famous second law of motion, F = ma, where
the “force” F is assumed to depend only on spatial position and veloc‐
ity. Thus, when the “force” is given or measured, the future and the
past follow deterministically from the current state of the system.
Stated in this form, it is clear that the possible futures of a
Newton ian system are so mehow already em bedded in the N ewtonian
concepts of force and system state. In fact, as Rosen (1985, 1991) has
shown in detail, dynamical models are based on the assumption that
the future state of the system is entailed by its previous state. From
a mathematical point of view, the dynamical systems of physics are
essentially recursive “state machines” (Rosen, 1978a).
In Newtonian physics, the idea that a future state can be com‐
puted from the earlier state is deeply related to the concept of “infin‐
itesimal” quantities, which Newton invented to boot up his celestial
mechanics. In basic calculus, infinitesimals are now known as differ‐
entials and this relation between the present and the future is known
as Taylor’s expansi on. It says that the val ue of a differentia ble function
6 of 15
|
Tuomi
at a specific point can be calculated using its value at a different point
when the value is corrected by rates of change. In particular, the
“points” can be point s in some abstrac t time. In exact mathematical
terms, Taylor expansion then says that, if derivatives exist, the value
of a function given its value at t0, is at a future point in time t1 = t0 + h:
In this equation, f´ indicates a time derivative of f, f´´ indicates
its second time derivative, the exclamation mark is the usual indi‐
cator of factorial function (n! = 1 × 2 × 3 × … × (n −1)×n), and the
sequence continues with further derivatives multiplied by the factor
hn/n!. In calculus text‐books, Taylor's expansion is introduced as a
consequence of the concept of derivative; an alternative interpreta‐
tion is that it actually defines what the concept of derivative tries to
do and why New ton invented a calculus of infinitely small quantities.
When a Newtonian description of system state and time is
valid, if we know the state and its derivatives at one point in time,
we can use Taylor's expansion to compute the state at a dif ferent
point, which we can call “the future” or “the next state.” If the
series used for this computation converges and the later terms in
the series are less and less important for the result, we get in‐
creasingly good approximations for the future state. In the case
of a Newtonian particle, only the position and its first derivative
need to be known to tell the future, assuming that we know the
forces that define the second derivative as a function of particle’s
position in its state space.
Rosen has shown that a very special form of entailment un‐
derpins this formalism of Newtonian particle physics. The use of
differential forms implies that system states need to be described
as “points” in a state space that is essentially a multi‐dimensional
Cartesian coordinate system. Time, in this representation, is a
label that defines what is earlier and what is later in this space. The
“trajectories” of particles in this space are solutions to differen‐
tial equations. New tonian systems, therefore, are “recursive.” The
next state of the system is determined by the earlier state and a
transition rule that represents “external forces.” Computer algo‐
rithms follow the same formalism, with the difference that time is
represented using integers instead of real numbers and the “exter‐
nal forces” are called the “program.”
Time in the Newtonian chronotope, therefore, has very little to do
with the physics of observed natural systems. In Newtonian physics
“true mathematical time” emerges as an absolute context that needs
to be posited to make the New tonian formalism possible. It is a for‐
mal thing, arising from the definition of derivatives and systems as se‐
quences of states located in the state space. In physic al terms it is just
a label that is needed to separate p oints in a state space. Thus New ton
is able to define in his Principia that true time is absolute and “from it s
own nature, flows uniformly without regard to anything external.”2
When Newton uses differential forms in modeling the trajecto‐
ries of celestial objects, a specific form of entailment and causality
becomes exported back to nature. The “topos” in the Newtonian
chronotope becomes a topological space of real‐valued Cartesian
coordinates, and “chronos” becomes an independent absolute pa‐
rameter that is needed to distinguish states and introduce dynamics
as transitions from one state to another. As a result, objects of na‐
ture become described as systems that have states, represented as
points in their “phase space,” and the internal causality of the system
becomes represented as a recursive sequence of states that gener‐
ates “trajectories” in this phase space. This representation proves
to be highly successful in generating predictions about Newtonian
systems, and nature thus becomes a mirror image of the Newtonian
formalism.
The key insight of Rosen is that this mirror c an reflect only very
extraordinar y systems. Only recursive systems of nature can be ac‐
curately modeled in this formalism. Rosen calls such systems “mech‐
anisms.” These systems are systems where the previous states of the
system determine its next state as soon as we are given a recursion
rule that embeds all the effects of the “environment .” This, however,
implies a very special understanding of the dif ference between a
system and its environment:
Newton’s Laws thus ser ve to transmute the initial
dualism between system and environment into a
new dualism, that bet ween phase (states) and forces,
or between states and dynamical laws. The states
or phases constitute a description of the system.
Environment, on the other hand, gets an entirely
different kind of description; it is described in terms
of the specific recursion rule it imposes on states or
phases. (Rosen, 1991, p. 94)
Rosen’s work on recursive systems was motivated by his search
for a mathematical formalism that could represent biological sys‐
tems. In particular, Rosen pioneered the use of category theory to
show that living systems have models of their environment that
allow them to act on expected future (Louie, 2009; Rosen, 1958b,
1978b, 1978c). Such anticipatory systems cannot be consistently
represented using the Newtonian formalism. If future states can
have an impac t on present change, the recursive Newtonian formal‐
ism breaks down. In this formalism, the impact of future states on
present, therefore, is not impossible because of the laws of Nature;
instead, it is impossible because the Newtonian formalism describes
the laws of Nature in a specific way that can only model processes
where history determines the future.
A simple heuristic to detect that recursive assumptions under‐
pin a model is that it uses differential forms. An alternative repre‐
sentation of recursive systems is a computational iterative system,
where the “forces of nature” are represented by a “program” that
operates in the context of a given “hardware” and the system state
is represented by “data variables.” Probabilistic models of the fu‐
ture, therefore, can be represented as algorithmic state machines
or as dynamical systems. Equilibrium economics and various forms
of trend analysis are replete with such models. They rely on a spe‐
cial mathematical formalism that uses dif ferential forms defined in
the space of real numbers. As a result, locations in the state space
f
(t0+h)=f(t0)+hf�(t0)+h
2
2!
f��(t0)+
...,
|
7 of 15
Tuomi
can be described using numbers. Current, past, and future system
states can be described in “quantitative” terms and many mathe‐
matical tools become available for their study.
The downside of this approach is that it excludes systems that
are not recursive. In it s standard form, the Newtonian formalism
usually assumes that the “forces” that encode the environment do
not change.3 Therefore, if we explicitly spell out the assumptions of
the probabilistic chronotope, we can say it captures the dynamics of
change if the system is recursive and the “rest of the world” does not
change. In other words, we may get good predictions as long as the
future is repetition of the past and there is nothing new under the
sun. In physics, where the fundamental objective is to find invariant
laws of nature, this approach is natural.
As the Newtonian formalism has during the centuries become
interpreted as the scientific way of describing nature, it is rarely no‐
ticed that there are many other types of systems in nature beyond
mechanisms. These include all living systems, as well as systems en‐
abled by living processes. Human societies and economies, ecosys‐
tems, and cognition are important examples of these. To more fully
describe the futures of such systems, we therefore need chrono‐
topes that differ from the probabilistic ones.
4 | THE CHRONOTOPE OF POSSIBILISTIC
FUTURES
Whereas probabilistic approaches rely on a recursive Newtonian
chronotope, possibilistic approaches use qualitative and narrative
forms of chronotope. Below the possibilistic approach is discussed
using Intuitive Logics scenarios as an example (Huss & Honton,
1987). This form of scenario development became popular after the
Royal Dutch/Shell adopted it as part of its strategy process around
the first oil crisis in 1972 (Wilkinson & Kupers, 2014). We use spe‐
cifically the Global Business Network scenario model popularized by
Schwartz (1991). Although the term “scenario” is used also in proba‐
bilistic contexts, Intuitive Logics scenarios produce alternative views
of the future which are assumed to be equally probable (Bradfield
et al., 2005).
4.1 | The chronotope of scenarios
There are many variations of scenario development processes.
Some of these are essentially probabilistic and use, for exam‐
ple, computer simulations and probabilistic models to automati‐
cally generate possible futures with varying probabilities. Below
we focus on the “gold standard” model popularized in the Art
of the Long View by Schwar tz (1991). It combines an essentially
Newtonian chronotope in a narrative framework that explicitly ad‐
dresses uncertainty. Although it embodies important probabilistic
assumptions in what scenario developers often interpret as the
“contextual environment” beyond control, it develops narratives
for several possible futures in an attempt to detect important ob‐
servables and to generate robust plans for action. The elements
that distinguish Intuitive Logics from probabilistic foresight can
therefore be framed in Bakhtinian terms as a shift towards a
chronotope where the focal ac tor becomes an active—although
highly constrained—participant and agent of change. Most im‐
portantly, however, a sharp differentiation between the “system”
and its contextual environment allows the environment to be de‐
scribed within a recursive Newtonian context.
Scenario development, as outlined by Schwartz, follows in many
respects closely the New tonian approach in modeling natural sys‐
tems. First, a topic is selected that defines the overall objectives of
the study and what the “system” is about. Second, the system is de‐
composed into components, also known as “local factors and key
element s.” In the Newtonian formalism, these correspond to the ele‐
mentary system particles. Third, the “environment” is represented as
“macro forces driving the change.” These are the Newtonian forces,
now operating on the focal system. In the general Newtonian case,
the forces c an explicitly depend on time, in which case the system is
called “non‐autonomous.” Time‐dependent forces can equivalently
be described in systems models and control theor y as time‐varying
external inputs (Rosen, 1991, pp. 81–89). In scenario work, the con‐
textual environment is typically by definition uncontrollable and ex‐
ogenous, and the environmental forces are structured in descriptive
thematic domains (e.g., political, economic, social, technological, “en‐
vironmental”, and legal) that obey laws of their own. The dynamics
in these contextual domains are typically assumed to follow simple
growth or decay pat ters, i.e., change in these environmental domains
is driven by constant forces that generate trends. These trends,
often based on historical data, then become interpreted as the “driv‐
ing forces” that constrain and control the focal system under study.
In policy‐oriented scenario work, key developments in this “environ‐
ment” are often represented as “megatrends” that are big enough to
be uncontrollable by any human agency.
The scenario approach, thus, avoids the determinism of
Newtonian particle physics by replacing the autonomous systems
typically studied in physics by non‐autonomous externally driven
systems w here time‐varying forces b reak the symmet ry between the
past and the future. The “trends” that represent these exogenous ex‐
ternal forces, however, look very much like Newtonian trajectories.4
In practice, trends are often described using time‐series data and
characterized by their first derivatives. This specific representation
of the environment follows naturally if we assume that the environ‐
ment itself obeys Newtonian dynamics. In contrast to the Bakhtinian
pre‐literary agrarian chronotope where growth and dec ay also have
a clear cyclical pattern, trends are recursive dynamical trajec tories
and extrapolations of the past. Although the focal system is poten‐
tially non‐Newtonian and may follow other narrative chronotopes,
the environment and the changes it induce are implicitly represented
using a Newtonian chronotope, and the focal system thus becomes
embedded in a recursive context.
An impor tant strength of scenarios arises from the fact that
the Newtonian chronotope is not used for probabilistic predic‐
tion. Trends are of ten approximated simply as patterns of grow th,
and, instead of prediction, the focus is on trend uncertainty. In the
8 of 15
|
Tuomi
Schwartz/GBN scenario approach, this uncertainty is used methodi‐
cally to generate alternative future states of the system. Qualitative
judgment is used to sor t out the most important forces that are also
the most uncertain ones and, based on these, alternative scenar‐
ios are developed. Scenarios themselves are described in narrative
forms where different chronotopes may be applied.5
Scenario methods, therefore, follow abductive reasoning. The
aim is to generate surprising, novel and relevant stories about pos‐
sible futures, and these are then explained as causal consequences
of events that could lead to the imagined outcome. Scenarios can
therefore generate images of the future that are considerably richer
than those provided by probabilistic approaches.
In practice, however, scenario development processes have
strong biases that weed out stories that do not fit with culturally
dominant narrative structures. Accepted narratives, in general,
rapidly evolve to established archetypes (Eliade, 1991), and sce‐
nario narratives tend to be “scientifically” coherent narratives that
follow the Newtonian chronotope. In Intuitive Logics, a central
requirement is the “plausibility” of resulting scenarios and key au‐
thors in this tradition emphasize the importance of creating “good”
scenarios (Spaniol & Rowland, 2018). For example, if a scenario
would follow a chronotope of a Greek romance, it’s “suddenlys”
and “at the last moments” would probably be considered unjusti‐
fied, ad‐hoc, and lacking sufficient evidence. It might be adventur‐
ous to propose such scenario narratives to well‐educated business
managers and policymakers. Although during scenario develop‐
ment esoteric, novel, and revolutionary ideas may be expressed,
these easily disappear in the process simply because meaningful
narratives can only be shared in already shared chronotopes. In
practice, scenarios therefore often become expressions of shared
dominant beliefs about the present, in a form that the process
participants expect to be easy to communicate to key stakehold‐
ers. Some narratives are inherently believable, diffuse virally, and
are recreated whenever imaginary visions of the future are pro‐
duced. Despite the valiant effort to generate qualitatively novel,
engaging, and relevant images of the future, scenario development
therefore often leads to scenarios where some rather isolated new
trends and phenomena are highlighted in the context of an overall
story that is not ver y different from what we have heard before.
The hero of the future may change from a steam engine to flying
car, computer, 3D printer, genetic scissors, or AI, but the dynam‐
ics of the story remain more or less the same. As to validate the
Bakhtinian claim that literar y forms both reflect changed realities
and provide chronotopes that make sense of them, the “heroes” of
scenarios often see the first light of the day in literar y imagination
and Hollywood movies.
Yet, at least in theory, scenario development may also lead to
new narrative chronotopes and new conceptualizations of action,
agency, and causalit y. A multi‐voiced dialogical chronotope, as high‐
lighted in Bakhtin’s studies of Dostoevsky, is of specific interest here.
It constructs actors as agents that can learn and become responsi‐
ble for their acts. In this chronotope, future is not simply something
that happens; instead, it results from human action, agency, and
interaction. Through multi‐voiced dialogue also qualitatively new
chronotopes can emerge that enable the participants to make sense
of their realities in new ways.
Constructivist foresight, described in the next section, builds on
this creative potential of dialogue and multiple interacting chrono‐
topes. In particular, in an attempt to capture unpredictability and
qualitative novelty, it tries to address the problem of moving beyond
already known chronotopes. The constructivist model implements a
process where multiple incompatible narrative structures and chro‐
notopes are explicitly used to generate expanding contexts for ac‐
tion. At the same time new chronotopes emerge where new forms of
agency and responsibility can make sense.
5 | THE CHRONOTOPE OF
CONSTRUCTIVIST FORESIGHT
The constructivist approach to foresight starts from the common
sense observation that future is not only repetition of the past.
Innovation creates things that did not exist before. Social, cultural,
and biological evolution and innovation therefore cannot be fully
described using historical facts, concepts, or data. Although we can
describe new in terms of the old, for example c alling electronic mes‐
saging “e‐mail” and personal mobile devices “phones,” we struggle
to describe those aspects of qualitative novelty that are essentially
new. It is easy to talk about the Internet as something old and we
have many words and concepts suitable for that; in contrast, it is
difficult to say how it is new. As the uses of the Internet evolve and
emerge, new competing interpretations and new suggestions for
novel concepts and conceptual systems are constantly introduced.
A remote‐access computing network becomes a global document
librar y, collaboration infrastruc ture, new public sphere and “social
media,” enabler for alternative finance and real‐time global produc‐
tion networks, and many other things (Tuomi, 2002). At the point of
their introduction these proposed concepts are ambiguous, ill‐de‐
fined, badly justified, odd, confusing, and lack good factual basis.
This is not because the emerging concepts would be “wrong.” It is
simply because the reality they try and describe is still in the process
of becoming.
Whereas probabilistic forecasting and narrative scenarios adopt
an epistemic focus, constructivist foresight has a distinctively on‐
tological focus. The future is not something to be known; indeed, it
cannot be known as it does not exist yet. In the constructivist ap‐
proach, the future is not known or understood; instead it is some‐
thing to be created.
The constructivist approach, therefore, comes with a novel cut
on epistemology and ontology. Whereas probabilistic and possibilis‐
tic foresight is almost always framed as a problem‐solving activity,
constructivist foresight starts from the obser vation that “problems”
are necessarily formulated with historically established concepts,
terms, and anticipatory frames. To see the future as a set of prob‐
lems and foresight as a process of problem solving implies that fu‐
ture is perceived as a continuation of the past. In such a perspective,
|
9 of 15
Tuomi
it is difficult to perceive or appropriate the latent possibilities of the
present. The constructivist approach, therefore, is closely related to
the idea that we need “futures literacy” as a capability to be open
to emergence (Miller, 2018a). From an ontological point of view, it
is aligned with the idea that ontology “expands” and new things and
phenomena become real as a result of human and natural creativit y
(Tuomi, 2017).
The constructivist approach starts from both theoretical and
practical considerations. On a formal level, it accepts Rosen's analy‐
sis of the capability of the Newtonian formalism to embed different
forms of entailment and causality. As Rosen (1958a) and his men‐
tor Rashevsky (1954) pointed out, a different formalism is needed
to model biological and biolog y‐based systems. The constructivist
assumption is that societies, economies, organizations, and ecosys‐
tems are biology‐based systems.
From an epistemological point of view, the term “constructiv‐
ist” gains some rather unconventional characteristic s. In line with
Rosen’s strict empiricism (1978a), constructivist foresight assumes
that observer’s modalities of interaction and the interaction modal‐
ities of the observed system jointly define the ontological reality. A
model, in general, is a system that maps a subsystem of the observed
system into another system with congruent dynamic organization.
Anticipatory systems are systems that have internal models that
enable them to make predictions about the future and adapt their
behavior accordingly (Rosen, 1985).
In this framework, constructivism therefore implies a special
form of realism that could be called “multiple‐worlds” realism. It says
that pieces of rock, ant s, and humans live in different realities, where
different causal factors influence change. For example, interest rates
may have causal influence on human action, whereas ants may live in
a chemical world of communication that remains outside the reality
of stones and humans. Most importantly, for living beings, realities
are continuously created as a result of evolutionary change, innova‐
tion, and undetermined choice. In constructivist foresight, models of
possible futures are created using human imagination so that they
can subsequently be realized with concrete causal influences. In ef‐
fect, non‐existent and yet‐to‐be futures are modeled with the intent
of making them real.
This form of “constructivism” therefore is ontologically different
from the more standard forms of epistemic constructivism or con‐
structionism associated with learning theorists such as Vygotsky,
Piaget, Papert, or von Glasersfeld, and research on social construc‐
tion in sociology and in science and technology studies (Ackermann,
2001; Berger & Luckmann, 1966; Bijker, Hughes, & Pinch, 1987; Cole
& Wertsch, 1996; von Glasersfeld, 1995). Instead of learning knowl‐
edge about an already existing reality, the constructivist approach
aims at creating novel futures and, in the process, learning some‐
thing that is not yet knowledge but will be knowledge in the future.
In the constructivist framework, imagined futures and expec ta‐
tions can therefore have causal effects, as Shackle suggested in his
economic theory (Shackle, 1958, 2012). As Rosen has shown, the
Newtonian framework necessarily excludes future influences and
“final causes” that are required for modeling functional systems.
Functional description requires that we represent the “what for,”
and this Aristotelian “final c ause”—necessary for modeling living and
anticipatory systems with implicit and explicit expectations—is the
missing part in the Newtonian formalism.
Rosen suggested in 1958 that the mathematical theory of cat‐
egories provides the required strength to model such systems (cf.
Louie, 20 09, 2013). The relational character of category theoretic
constructs is able to capture characteristics of systems where or‐
ganization is more important than the particular characteristics of
system component s. In biology, living organisms have lifetimes that
greatly exceed the lifetimes of their constituent elements, and the
same can be said about societies and economies. The Newtonian
formalism, in contrast, is based on a reductionist approach, where
system characteristics can be derived from the characteristics of
their constituent elements. When a system is more than a sum of it s
parts or when it has dynamic s that produce qualitative novelty, the
Newtonian formalism breaks down, and category theoretic models
become relevant.
The emphasis of constructivist foresight on the creation of nov‐
elty, combined with the recognition of the limited value of historical
observation, means that the traditional epistemic focus of foresight
is replaced by an ontological focus. By shifting the focus from epis‐
temic challenges to ontology, the cons tructivist approach also avoids
the common problem of bridging knowledge with action. Many
future‐oriented initiatives have difficulties in “knowledge trans‐
fer” and struggle with the generation of “actionable knowledge.”
Constructivist foresight focuses on joint creation of meaningful de‐
signs for possible futures and their experimental implementation,
and it is therefore inherently action‐oriented. In constructivist fore‐
sight, action emerges with a shared interpretation of the meaning of
this action, and there is no need to separately address interpretation,
knowing and action.
5.1 | “Next‐generation” foresight as an
example of the constructivist approach
To illustrate the constructivist approach, we use the “next‐genera‐
tion foresight” process model, developed for the European Forum on
Forward‐Looking Activities expert advisory group at the European
Commission (Tuomi, 2013). The model adopts some key ideas from
agile models used in new product development processes and com‐
bines these with a Bakhtinian idea of dialogue as a way of generating
new interpretations and meaning. The model views possible futures
as innovations that are jointly constructed by several communities of
practice, each with their own interpretative systems.
The “next‐generation” model links three basic insights. First , as
research on learning (e.g., Engeström, 1987; Vygotsk y, 1986), com‐
munities of practice (Brown & Duguid, 2001; Constant, 1984; Fleck,
1979; Lave & Wenger, 1991; Orr, 1986), and knowledge creation
(e.g., Tuomi, 1999) has shown, systems of meaning are grounded in
social practice and socially meaningful activity. Different communi‐
ties have dif ferent and often incompatible systems of sense‐making
and interpretation. In collaboration that involves specialization and
10 of 15
|
Tuomi
division of labor, these dif ferent systems of meaning need to be ne‐
gotiated. Negotiation, however, does not have to result in a shared
interpretation, and often collaboration can proceed without consen‐
sus (Star, 1992). The concepts and ar tifacts generated in the process
can then become “boundary objects” and “boundary infrastructures”
(Bowker & Star, 1999; Star, 2010) or equivocal and ambiguous “cryp‐
tic concepts” (Bougon, Weick, & Binkhorst, 1977) that link differ‐
ent socially meaningful interpretations, allowing work to proceed.
Different groups, or representatives of different sense‐making com‐
munities, generate in this model different futures using their idiosyn‐
cratic meaning horizons; to generate something that moves beyond
given systems of meaning, these, however, need to be collapsed into
a shared ar tifac t that operates as a boundary object across several
systems of meaning. This requires a “dialogic” process where sev‐
eral alternative narratives become reinterpreted in the context of
a shared story and a new chronotope where new kinds of future,
action, and agenc y can make sense.
The “next‐generation” model, in particular, aims at expanding the
perceived reality and its opportunities for action, and therefore sto‐
ries are evaluated in this model also based on how effectively they
generate new insights and “out‐of‐the‐box” thinking. A specially tai‐
lored incentive system can be established to support novel insights
and to reduce the tendency to repeat already established visions of
accepted futures. The process itself is conceived as a playful contest
of generating the most innovative and engaging narratives of future.
Second, an ontological assumption in the constructivist ap‐
proach is that the future does not exist yet. What counts as data,
facts, and knowledge in the emerging future is known only retro‐
spectively. Although elements of our cognitive and tacit models of
reality may remain useful also in future, there is no guarantee of this.
Socially and economically important discontinuities, by definition,
break our existing models, and make new types of knowledge pos‐
sible. Future, therefore, needs to be approached as a problem of de‐
sign, experimentation, and learning. The way to generate knowledge
is to experiment with futures and simultaneously construct activity
and concepts that effectively capture the emerging future. The con‐
structivist approach, therefore, is a creative and “poetic” activity in
the Aristotelian transformative and productive sense. It has to rely
on old concepts and words to describe something that is qualita‐
tively new, and an effec tive way of doing this is to create artifacts
that change the meaning of old terms so that they capture some key
aspects of novelty. This emphasis on concrete construction as an
objective of joint action is therefore also aligned with the practices
of design thinking (Brown, 2009).
Third, influenced by the available historical evidence, the “next‐
generation” model takes seriously Alan Kay’s famous claim that “the
best way to predict the future is to invent it.” Kay st ated this in one
of his presentations at Xerox PARC soon after it was founded, in
1971. 6 One could argue that, simply looking at the history, Kay and
his colleagues have excellent track records in predicting the future.
Without idolizing the Silicon Valley innovation culture, one may
argue that as a social system, Silicon Valley represents a collective
process of foresight that generates futures by testing alternative
interpretations of them. Although it is clear that in this specific case
imagination and experimentation are strongly filtered by economic
interests, the constructivist design‐oriented model of foresight is
compatible with the experimental dynamics of Silicon Valley. For ex‐
ample, it is well known that star t‐up firms in Silicon Valley rarely rely
on long‐term planning; instead, they reinvent business models many
times based on experiences gained during implementation before
a working business and product model is created (e.g., Bahrami &
Evans, 2000; Maidique & Zirger, 1985). Whereas foresight initiatives
are often conceived as knowledge production activities, in the so‐
cial system of Silicon Valley, experiment ation tells what is relevant
and what counts as a fac t. As the “next‐generation” model views
futures as something to be designed and constructed, the process
shares many characteristics with new product development in rap‐
idly changing environments. As it emphasizes “making the future,”
instead of more passive “new understanding,” it could also be as‐
sociated with Intuitive Logics approaches that emphasize “seeding”
instead of “seeing” (Wilkinson, Kupers, & Mangalagiu, 2013).
In the “next‐generation” process model, several teams are set
up and par ticipants are given a “theme,” around which they play
and improvise to generate a specific narrative model of the future.
Different teams work to generate alternative futures, and use in‐
struments they know how to play with. Different foresight tools,
including forecasting and scenario development, may be used by
different teams. In the “next‐generation” model the teams in the
first phase of the process explicitly aim at producing qualitatively
different stories around a common theme, without requiring that
these stories are conceptually compatible. The assumption is that
different interpretative systems are associated with different chro‐
notopes, agency concepts, and value systems.7 The process aims at
capturing dynamics of complex systems, and the theory of anticipa‐
tory systems shows that many incompatible models are necessary to
describe such systems.
The “next‐generation” model therefore starts by generating al‐
ternative narratives about the future, each based on the interpre‐
tative and value system available for the team that generated the
narrative. The resulting narratives are first kept separate and then
used as dialogical inputs to generate qualitatively new stories. In
metaphorical terms, the teams are provided with “theme notes”
around which they develop melodies that fit the genre and instru‐
ments available for the team. In the second phase of the process,
these “melodies” are shared and combined to create a multi‐voiced
piece that “sounds right ” for the participants. The assumption is that
the participant s cannot know what sounds right until they hear the
music.
A theoretical assumption that underpins this process is that
different communities of practice have different stocks of knowl‐
edge and they use different conceptual models to characterize their
worlds. Whereas diversity and heterogeneity is introduced in con‐
ventional Intuitive Logics processes by engaging individuals with
different backgrounds, in the “next‐generation” model knowledge is
assumed to be a social thing, created in heterogeneous systems of
meaning, and actualized in social interaction and socially meaning ful
|
11 of 15
Tuomi
action. A “composition” of the most interesting narrative elements
can be built as the par ticipants’ capability to recognize interesting
things greatly exceeds their capability to articulate things.
The chronotopic assumption that underpins this model, there‐
fore, is that emergent realities c an be articulated in a dialogical
process where different local chronotopes interact with concrete
realities. The process is creative and poetic. The ontological assump‐
tion of the construc tivist approach, in turn, implies that we don’t
have fact s or data about things that are in the process of becoming.
We can know what counts as knowledge only retrospectively. The
“next‐generation” foresight model, therefore, adopts an agile prod‐
uct development approach, and composes “future designs” that are
used to generate rapid “experimental sprints.” The underlying model
is essentially a model of knowledge creation and future‐oriented col‐
lective sense‐making. The model thus also builds on Nonaka’s early
work on agile product development (Nonaka & Takeuchi, 1986).
Borrowing from agile product development methods, each “sprint”
is based on expectations of what designs of future would work, and
the expectations are verified and corrected based on results from
practical implementation. As in agile methods such as the recent
variations of Nonaka’s SCRUM approach, the scale and duration of
experiments are kept short, so that ef fective learning becomes pos‐
sible (Schwaber & Beedle, 2002).
6 | COMPARING THE THREE TYPES OF
FORESIGHT
The “next‐generation” foresight model discussed above should look
very familiar to innovation and new product development practi‐
tioners. In contrast to marketable products, it, however, develops
futures that enable the participants to make sense of the possibilities
of the present. Most importantly, it is based on epistemic and onto‐
logical assumptions that are different from probabilistic and possi‐
bilistic approaches to anticipation. The probabilistic approach builds
on an empiricist epistemology, where increasingly accurate models
can be generated by incremental improvements that result from ob‐
served differences between prediction and observation. The under‐
pinning learning model is adaptive adjustment based on empirical
feedback. This model is known as adaptive or associative learning or,
in Bateson’s (1973) terms, Learning I.
The possibilistic approach rejects the idea of predictive mod‐
els. Instead it uses uncertainty as a heuristic to generate narrative
fragments about alternative futures. By making uncertainty explicit,
it expands the space for possible future trajectories. As discussed
above, the Intuitive Logics approach describes the environment
in recursive Newtonian terms. The environment is represented as
external forces and trends that drive change in the focal system.
Somewhat paradoxically, Intuitive Logics also assumes that scenario
developers can know how cert ain or uncertain the driving trends
are. The learning model is abductive in the sense that the generated
scenarios are used as hypotheses that can be suppor ted by fur ther
collection of data and evidence. Using Bateson’s terms, the learning
model in possibilistic foresight can therefore also be characterized
as deutero learning or Learning II. In this model, scenario processes
aim at expanding the set of alternative choices in given possible
environments.
The constructivist approach has a very different epistemological
foundation. It starts from the assumption that future does not exist
yet and therefore we do not yet know what will count as relevant
data and knowledge af ter this emerging future is realized. Instead
of epistemic uncertainty, it therefore emphasizes ontological unpre‐
dictability ( Tuomi, 2012). This le ads to a view where the environment
and the focal system are integrated elements in a larger picture. The
ontology that foresight tries to understand and make sense of is not
given but expands as a result of innovation and evolutionar y pro‐
cesses. Innovation implies that neither the system environment nor
the focal system it self are mechanisms. The system needs to be de‐
scribed in functional terms, where the system components can only
be understood in relation to the system itself. A relational approach
is therefore needed to model such systems. This relational approach
is essentially the one that underpins the Rashevsky‐Rosen‐Louie
school of “relational biology,” as well as Rosen’s work on anticipatory
systems.
In the constructivist approach, innovation and change do not
just “happen.” On the contrary, they are result s of active action and
negotiation in a heterogeneous social system with many different
social practices and communities that constantly reproduce these
practices. The learning model of the constructivist approach can
thus be characterized as innovative and experiential learning. It also
emphasizes the inseparable link between knowledge creation and
new forms of action, viewing knowledge as a capabilit y for intelli‐
gent action (Tuomi, 1999).
These dif ferent forms of foresight have also different objectives
and criteria for success. In the probabilistic approach the expected
impact is optimal choice, and the implicit measure of success is pre‐
diction accuracy. As discussed above, the recursive chronotope is
based on a formalism that is able to generate accurate predictions
about future if the system is a mechanism. In domains where social,
economic, or biological phenomena are important, or where innova‐
tion creates qualitative novelt y, dynamical models have limited pre‐
dictive capacit y. If model accuracy is tested, it is therefore usually
tested with historical data and retrospective prediction. In contrast,
in the possibilistic approach the quest for optimal choice is moved to
a higher level of abstraction. As the world is assumed to be complex
and uncer tain, the best we can do is to find directions for ac tion
while avoiding dead‐ends. Scenario methods therefore, in general,
aim at generating better maps of the future and in improving the
capacity of participant s to make sense of a complex world. The ef‐
fort is successful if dead‐ends are avoided, new promising paths for
action are found, and the resilience of the focal system is increased.
Whether the right trends and system descriptions were used can,
however, only be known retrospectively.
Also the constructivist approach aims at expanding the space
of possible action. In contrast, it emphasizes action as a means of
creating knowledge. The choices and decisions, in this view, are
12 of 15
|
Tuomi
not outcomes based on given data or knowledge; instead, they are
processes that require effort and take time. The set of “choices” are
essentially functional design hypotheses generated and articulated
simultaneously with actions that implement these choices. Action is
based on anticipator y models, and the outcomes can only be eval‐
uated retrospectively, but the process aims at making the chrono‐
topic assumptions explicit, thus allowing the participants to develop
sense‐making frames that can capture qualitative novelty and incor‐
porate new modes of agency.
The objec tive of construc tivist foresight is not to know the f uture
but to make it. Actors in the process are viewed as agents that do not
only generate knowledge but also more broadly develop capabilities
that enable new forms of value. The evaluation model, therefore, is
not simply predictive accuracy or organizational survival and suc‐
cess; instead, the evaluation model is at least partly generated si‐
multaneously with the designs for the future. In the constructivist
approach, also values internal to the focal system are part of the
design process. In particular, an important aim is to articulate new
forms of value that become possible as the future is realized.
This also means that the designers of the future can become
responsible for their action in the Bakhtinian sense. The construc‐
tion of future is a dialogical process, where the authors of future
act based on the expected ef fects of their ac ts. The “construction,”
“authoring” or “architecturing” of the future always occurs in the
context of values that evaluate the futures based on the expected
impact of action. Agency is expressed in a dialogical process where
the expected response of the world shapes the act, and where the
underpinning system of values make distinctions possible. All au‐
thentic and creative action, thus, necessarily reflects the values of
the actor, and makes the author in Bakhtin’s (1990) terms “answer‐
able” to his or her acts, at the same time making it possible for the
actor to be responsible for his or her acts. Constructivist foresight,
therefore, is not just a means to given ends; it also requires that the
ends themselves are negotiated in the process. If shared values are
found, commitment follows naturally but this also means that the
actors take responsibility for their action.
It is therefore clear that the chronotopic structure of these three
approaches to foresight lead to different practical implementations
of foresight processes and also different views on epistemolog y, on‐
tology, and agency. Table 1 summarizes some of these differences.
7 | SUMMARY: MAKING FUTURES REAL
In this paper, we have discussed three alternative approaches to an‐
ticipation and foresight. To study the different domains where these
approaches can coherently model the future, we used the concept
of chronotope, and illustrated this with Bakhtin’s narrative chrono‐
topes. Building on Rosen’s anticipatory systems theory we charac‐
terized systems that c an be modeled using recursive chronotopes,
and argued that non‐Newtonian chronotopes are needed to model
emergence and innovation in complex systems.
The probabilistic approach to foresight relies on a formal
description that has its roots in Newton's physics and its repre‐
sentation of natural systems using differential equations and envi‐
ronmental forces. As the underlying formalism requires recursive
processes, it can only capture future as repetition of the past. In
this formalism, it becomes impossible to ask “why” or “what for,”
and the concept of agency and the possibility for responsible
action effectively disappear. The underlying formalism was de‐
veloped to describe the invariant laws of nature and their conse‐
quences; the “how” instead of the “why.” Probabilistic foresight
underpins many different models of forec asting and policy anal‐
ysis, and the limitations of this approach need to be understood
when studying the impact of innovation, learning, and qualitative
change on emerging futures.
In contrast, possibilistic foresight, here illustrated with Intuitive
Logics scenarios, was shown to be able to capture alternative mod‐
els of the future using different narrative chronotopes. Whereas
probabilistic foresight is rooted in a single recursive chronotope,
scenario methods are based on alternative stories that may have
their own chronotopes. However, as the environment is implic‐
itly described as a Newtonian mechanism, the complex processes
that underpin innovation, evolution, and social change cannot be
modeled in this approach. Furthermore, the narratives generated
in scenario development processes often become filtered and mod‐
ified so that they appear scientific in the Newtonian sense. Many
scenario narratives can be told—but only if they do not “break the
laws of physics.” The unacknowledged consequences of adapting to
these “laws of physics” and associated recursive chronotopes are
many. These include the assumption that time is reversible, antic‐
ipated future cannot influence change, and, for example, that all
TABLE 1 Three types of foresight
Probabilistic (Forecasting) Possibilistic (Scenarios)
Constructivist (Design‐based
Foresight)
Epistemic model Empirical convergence Uncertainty and complexity Relational ontology, unpredictability
Innovation model Incremental Partly incremental and disruptive Distributed, multi‐focal
Base chronotope Recursive, algorithmic Narrative in mechanical environment Relational, dialogic
Learning model Adaptive Abductive Expansive, action‐oriented
Impact model Optimal choice Insightful sense‐making, cooperation,
preparedness
Expanded space for action, value
creation
Evaluation model Prediction accuracy Resilience, optimal action, routing
around dead‐ends
Heedful action, realization of latent
opportunities, emergent
|
13 of 15
Tuomi
interesting systems can be simulated by computer algorithms. The
narrative potential of scenario approaches, therefore, is rarely fully
realized in practice.
To address the challenges posed by possibilistic and probabilistic
approaches, constructivist foresight builds on a relational approach
originally developed in the context of mathematical biology. This re‐
lational starting point leads to epistemolog y and ontology that are
different from those commonly adopted in probabilistic and possi‐
bilistic foresight. In particular, knowledge, in this view, is a phenome‐
non that c an only be unders tood by considerin g its role in organi c life,
and knowing thus becomes closely related to capability for action. In
this view, technological and social development expands the domain
of knowing and generates qualitatively new types of facts and data.
This also means that the traditional strict separation between know‐
ing subjec ts and the known objects blurs. Living organisms are active
agents in an environment that becomes an environment for them
through active effort of organizing and anticipating it; in other words
by making it meaningful.
In the previous sections, we have emphasized the limitations of
probabilistic and possibilistic foresight, and highlighted the poten‐
tial of constructivist foresight to address some of these limitations.
Possibilistic and probabilistic foresight methods are influential in
strateg y and policy development, and we have emphasized their
limitations as it seems they have not received suf ficient theoretical
attention. Those “adventure‐time” chronotopes that Bakhtin found
in Greek romances can still be found in policy and strategy narra‐
tives, and important policy choices are frequently made based on
recursive chronotopes that are applied far from their domain of va‐
lidity. The “next‐generation” foresight model used to illustrate the
constructivist approach is obviously a very simplified model and has
its own limitations. The comparison of these different approaches,
however, shows that there are different ways to understand time
and future and that these lead to different practical consequences
in the present. Moreover, anticipatory systems theory provides
well‐developed formal methods that can be used to fur ther study
the conceptual foundations of dif ferent foresight approaches. This
is one of the main messages of the present article. Different ways
of structuring the relations between time, space, and agency make
different stories about future possible and thus constrain and enable
different ways of interpreting the present and its possibilities.
CONFLICT OF INTERESTS
None.
ACKNOWLEDGEMENTS
The research was par tly conducted at the Stellenbosch Institute
for Advanced Studies. The Finnish Technology Agency TEKES
provided partial funding through its Innovation Policy Research
Programme. The sponsors had no role in the design of the study
or its reporting.
NOTES
1This interpretation of pre‐modern cyclic chronotope is thus sub‐
stantially different from Eliade’s influential characterization of time
in myths , religions and culture. Eliade (1991) argued that manifes‐
tations of the sacred establish the struc ture of the world, and that
“profane” non‐religious experience has only geometric directions,
lacking value. Bakhtinian chronotopes, in contrast, have their roots
in the action of biological organisms, and all action expresses values.
The concept of chronotope trie s to capture exactly that structure,
whether it is profane or sacred.
2“Tempus abso lutum verum & Math ematicum, in se & na tura sua absq; rela ‐
tiona ad externum quodvis, aequabiliter fluit, alioq” p. 5 Scholium I.
3Technically, it is possible to define time‐var ying forces. In general sys tem
theor y and control theory, this leads to external inputs that vary in time
and in dynamical systems to non‐autonomous dynamical systems.
4Derbyshire and Wright (2017) have also noted that Intuitive Logics sce‐
narios rely on an implicit New tonian causality, and they argue that a
broader Aristotelian model of causality is needed in foresight. The dif‐
ferent forms of Aristotelian c ausality have been extensively analyzed
in anticipatory s ystems theory since the 1980 s. A recent formal study
can be found in Louie (2013).
5Scenarios therefore require implicit anticipatory models that can be used
to distinguish “probable” and “uncertain” forces.
6The inspiration for the maxim probably came from Dennis Gabor.
Somewhat paradoxically, Kay later stated that “the best way to pre‐
dict the future is to invent it, because we can then say, ‘the future’s
there for us to shape — we’re not helpless.’ As long as we don’t vi‐
olate too many of Newton’s laws, we can probably make new tech‐
nology work out.” (Kay, 1983) In fact, although physical models of
transistors and microprocessors obey the recursive chronotope of
quantum mechanics, they do not obey Newton’s laws, and as pointed
out above, we may need to break not only Newton’s laws but their
implicit chronotope to invent socially and economically interesting
futures.
7The teams may therefore also be viewed as pre‐modern value‐based
communities in the Durkheimian (1933) sense. As was pointed out by
Constant (1984), also different communities of scientific and engineer‐
ing expertise have different criteria for evaluating designs, including
their ef fectiveness and aesthetic “quality.”
ORCID
Ilkka Tuomi https://orcid.org/0000‐0002‐4179‐7103
REFERENCES
Aaltonen, M. (2007). Chronotope Space: managing the complex trade‐
offs between the properties of the strategic landscape and the
time frame being considered. Foresight, 9(4), 58–62. https://doi.
org /10.1108/14 636 680710773830 .
Aalton en, M., & Holm ström, J. (2 010). Multi‐onto logy topo logy of the
strategic landscape in three practical cases. Technological Forecasting
and Social Change, 77(9), 1519–1526. ht tps://doi.o rg/10.1016/j .
techfore.2010.07.004.
Ackermann, E. (2001). Piaget’s constructivism, Papert’s construction‐
ism: What ’s the difference? In Constructivism: Uses and Perspectives
in Education (Vol. 1 & 2, pp. 85–94). Geneva: Research Center in
Education.
Awodey, S. (2010). Category theory. Oxford; New York: Oxf ord University
Press.
14 of 15
|
Tuomi
Bahrami, H., & Evans , S. (2000). Flexible recycling and high‐technolog y
entrepreneurship. In M. Kenney (Ed.), Understanding Silicon Valley:
The anatomy of an entrepreneurial region (pp. 165–189). Stanford, CA:
Stanford University Press.
Bakhtin, M. (1981). Forms of time and of the chronotope in the novel:
Notes toward a historical poetics. In M.Holquist (Ed.), The dialogic
imagination: Four essays by M.M. Bakhtin (pp. 84–258). Austin, TX:
University of Texas Press.
Bakhtin, M. (1990). Art and answerability: Early philosophical essays by M.
Bakhtin (M.Holquist & V.Liapunov, Eds.). Austin, TX: University of
Texas Press.
Bakhtin, M. (1993). Toward a philosophy of the act. Austin, TX: Universit y
of Texas Press.
Barry, D., & Elmes, M. (1997). Str ategy retold: Toward a narrative view of
strategic discourse. Academy of Management Review, 22(2), 429–452.
https://doi.org/10.5 465/AMR .1997.9707154065 .
Bateson , G. (1973). Steps to an Ecolog y of Mind. London: Paladin.
Bemong , N., Borgha rt, P., De Dobbele er, M., Demoen, K ., De Temmerman ,
K., & Keunen, B. (2010). Bakhtin’s theory of the literary chronotope:
Reflections, applications, perspectives. Gent: Academia Press.
Berger, P. L., & Luckmann, T. (1966). The social construction of reality: A
treatise in the sociology of knowledge. New York: Penguin Books.
Bijker,W.E.,Hughes, T.P.,& Pinch, T. J.(1987).The social construction
of technological systems: New directions in the sociology and his tory of
technology. Cambridge, MA: The MIT Press.
Bishop, P., Hines, A., & Collins, T. (2007 ). The current state of scenario
development: and overview of techniques. Foresight, 9(1), 5–25.
htt ps://doi.org /10.1108/1463668 0710727516
Boje, D. M. (2001). Narrative methods for organizational & communication
research. London: SAGE Publications.
Bougon, M. G., Weick , K. E., & Binkhor st, D. (1977). Cognition in orga‐
nizations:An analysisoftheUtrecht Jaz zOrchestra. Administrative
Science Quarterly, 22, 606–639. https://doi.org/10.2307/2392403
Bowker, G., & Star, S. L. (1999). Sorting things out: Classification and its
consequences. Cambridge, MA: The MIT Press.
Bradf ield, R., Wright , G., Burt, G ., Cairns, G. , & van der Heijden, K . (2005).
The origins and evolution of scenario techniques in long range busi‐
ness planning. Futures, 37, 795–812. https://doi.org/10.1016/j.
futures.2005.01.003
Brown,J.S.,&Duguid,P.(2001).K nowledgeandorganization:A social‐
practice perspective. Organization Science, 12(2), 198–213. https://
doi.org /10.1287/orsc.12. 2.198.10116
Brown, T. (2009). Change by design: How de sign thinking trans forms organi‐
zations and inspires innovation. New York, NY: Harper Collins.
Chebanov, S. V. (2015). Ukhtomsk y’s idea of chronotope as frame of an‐
ticipation. In M. Nadin (Ed.), Anticipation: Learning from the past—The
Russian/Soviet contributions to the science of anticipation (pp . 13 7–
150). Cham: Springer.
Clark, K., & Holquist, M . (1984). Mikahil Bakhtin. Cambridge, MA: The
Belknap Press of Harvard University Press.
Cole,M .,&Wer tsch ,J.V.(1996).Beyond theindividual‐socialantinomy
in discussions of Piaget and Vygotsky. Human Development, 39(5),
250–256. ht tps://doi.o rg/10.1159/00 0278 475
Constant, E. W. (1984). Communities and hierarchies: Structure in the
practice of science and technology. In R. Laudan (Ed.), The nature of
technological knowledge: Are models of scientific change relevant? (pp.
27–46). Dordrecht: Reidel.
Cournand, A., & Lévy, M. (1973). Shaping the Future: Gaston Berger and
the Concept of Prospective. London: Gordon and Breach Science
Publishers.
Czarniawska, B. (1997). Narrating the organization: Dramas of institutional
identity. Chicago, IL: The University of Chicago Press.
Derbyshire, J., & Wright, G. (2017). Augmenting the intuitive log‐
ics scenario planning method for a more comprehensive analysis
of causation. International Journal of Forecasting, 33(1), 254–266.
https://doi.org/10.1016/j.ijforecast.2016.01.004.
Douglas, M. (1987). How institutions think. London: Routledge & Kegan
Paul.
Durkheim, E. (1933). Div ision of labor in soci ety. New York: The Free Press.
Eliade, M. (1991). The myth of the eternal return: Cosmos and history.
Princeton,NJ:PrincetonUniversityPress.
Engeström, R. (1995). Voice as communicative action.
Mind, Culture, and Activity, 2(3), 192–215. https://doi.
org /10.108 0/10749039509524 699.
Engeström, Y. (1987). Learning by expanding: An activity theoretical ap‐
proach to developmental work research. Helsinki: Orienta Konsultit.
Fleck, L . (1979). Genesis and development of a scientific fact. Chicago, IL:
The University of Chicago Press.
Fuller, T. (2017). Anxious relationships: The unmarked futures for post‐
normal scenarios in anticipatory systems. Technological Forecasting
and Social Change, 124(November), 41–50. https://doi.org/10.1016/j.
techfore.2016.07.045.
Hillier, B., & Hanson, J. (1984). The social logic of space. C ambridge:
Cambridge University Press.
Holloway, J., & K neale, J. (20 00). Mikhai l Bakhtin: Dia logics of space .
In M. Crang & N. Thrift (Eds.), Thinking space (pp. 71–88). London:
Routledge.
Holquis t, M. (1990). Dialogism: Bakhtin and his world. London: Routledge.
Huss, W. R., & Honton, E. J. (1987). Scenario planning–what style
should you use? Long Range Planning, 20(4), 21–2 9. h t tps ://do i.
org /10.1016/0 024‐63 01(87 )90152‐x
Hutchins, E. (1995). Cognition in the wild. Cambridge, MA: MIT Press.
Jarva,V.(2014). Introductionto narrativeforfutures studies.Journal of
Futures Studies, 18 (3), 5–26.
Kay, A. (1983). Learning vs. teaching with educational technologies.
EDUCOM Bulletin, Fall/Winter(3/4), 16–17.
Keunen, B. (2011). Time and imagination: Chronotopes in western narrative
culture. Chic ago, IL: Northwestern Universit y Press.
Lave,J.,& Wenger,E. (1991).Situated learning: Legitimate peripheral par‐
ticipation. Cambridge: Cambridge Universit y Press.
Louie, A . H. (20 09). More than life itself: A synthetic continuation in rela‐
tional biology. Frankfur t: Ontos Verlag.
Louie, A . H. (2010). Robert Rosen’s anticipatory systems. Foresight, 12(3),
18–29. https://doi. org /10.1108/14636 6810110 49848
Louie, A . H. (2013). The reflection of life: functional entailment and immi‐
nence in relational biology. New York: Springer.
Mac Lane, S. (1978). Category theory for the working mathematician. New
York: Springer Verlag.
Maidique, M. A., & Zirger, B. J. (1985). The new produc t learn‐
ing cycle. Research Policy, 14, 299–313. https://doi.
org/10.1016/0048‐7333(85)90001‐0
Miller, R. (2015). Learning, the future, and complexity. An essay on the
emergence of Futures Literacy. European Journal of Education, 50(4),
513–523. https ://doi.or g/10 .1111/ej ed.12157.
Miller, R. (2018a). Sensing and making‐sense of Futures Literacy:
Towards a Futures Literac y Framework (FLF). In R . Miller (Ed.),
Transforming the future: Anticipation in the 21st century (pp. 15–50).
London: Routledge.
Miller, R. (Ed.) (2018b). Transforming the future: Anticipation in the 21st
Century. London: Routledge.
Morson , G. S., & Emerson, C. (1990). Mikhail Bakhtin: Creation of Prosaics.
Stanford, CA: Stanford University Press.
Nonaka, I., & Takeuchi, H. (1986). The new new pro duct development
game. Har vard Business Review, 64(1), 137–146.
Orr,J.E.(1986).Na rra tivesatwork :Sto rytellinga scoo per ati vediagnos‐
tic activity. In H. Krasner & I. Greif (Eds.), Proce edings of the 1986 ACM
conference on Computer‐supported cooperative work (pp. 62–72). New
York, NY: ACM. https://doi.org/10.1145/637069.637077.
|
15 of 15
Tuomi
Poli, R. (2017). Introducing anticipation. In R. Poli (Ed.), Handbook of an‐
ticipation (pp. 1–14). Cham: Springer International Publishing. https://
doi.org/10.1007/978‐3‐319‐31737‐3_1‐1.
Poole, B. (2001). From phenomenology to dialogue: Max Scheler’s phe‐
nomenological tradition and Mikhail Bakhtin’s development from
“Toward a philosophy of the act” to his study of D ostoevsky. In K.
Hirschkop & D. Shepherd (Eds.), Bakhtin and cultural theory (2nd ed.,
pp. 109–135). Manchester: Manchester University Press.
Rashevsky, N. (1954). Topology and life: In search of general mathe‐
matical principles in biology and sociology. Bulletin of Mathematical
Biophysics, 16, 317–348. https ://doi.org /10.10 07/BF0 248 4495
Rosen, R . (1958a). A relational theor y of biological systems. Bulletin of
Mathematical Biophysics, 20, 245–260. https://doi.org/10.1007/
bf02478302
Rosen, R . (1958b). The representation of biological system from the
standpoint of the theor y of categories. Bulletin of Mathematical
Biophysics, 20, 317–341. htt ps: //doi.o rg/10.10 07/bf 02477890
Rosen, R . (1978a). Fundamentals of measurement and representation of
natural systems. New York: North‐Holland.
Rosen, R . (1978b). On anticip atory systems: I. When c an a system conta in
a predictive model of another? Journal of Social a nd Biological Sys tems,
1, 155–162. https://doi.org/10.1016/s0140‐1750(78)80 003‐7
Rosen, R . (1978c). On anticipatory systems: II. The nature of the model‐
ling relation bet ween systems. Jo urnal of Social an d Biological System s,
1, 163–180. htt ps://doi.org /10.1016/s 0140 ‐1750(78)8 00 04‐9
Rosen, R . (1985). Anticipatory systems: Philosophical, mathematical and
methodological foundations. Oxford: Pergamon Press.
Rosen, R . (1991). Life itself: A comprehensible inquiry into the nature, origin
and fabrication of life. New York: Columbia University Press.
Schwabe r, K., & Beedle, M. (20 02). Agile software development with scrum.
UpperSaddleRiver,NJ:PrenticeHall.
Schwar tz, P. (1991). The art of the long view: Planning for the future in an
uncertain world. New York, NY: Doubleday.
Shackle, G. L. S . (1958). Time in economics. Amsterdam: North‐Holland.
Shackle, G. L. S . (2012). Expectation in economics. Cambridge: Cambridge
University Press.
Sokolova, L. V. (2015). On the legacy and life of academician Alexei A.
Ukhtomsky. In M. Nadin, (Ed.), Anticipation: Learning from the past—
The Russian/Soviet contributions to the science of anticipation (pp.
113–136). Cham: Springer.
Spaniol, M. J., & Rowland, N. J. (2018). Defining scenario. Futures &
Foresight Science, e3. https: //doi.org/10.10 02/ffo2.3
Star, S. L. (1992). Cooperation without consensus in scientific problem
solving: dynamics of closure in open s ystems. In S. Easterbrook (Ed.),
CSCW: Cooperation or conflict? (pp. 93–106). Berlin: Springer‐Verlag.
Star, S. L. (2010). This is not a boundar y objec t: Reflections on the origin
of a concept. Science , Technology & Human Values, 35(5), 601–617.
https://doi.org/10.1177/0162243910377624.
Steinby, L. (2013). Bakhtin’s concept of the chronotope: The viewpoint of
an acting subjec t. In L. Steinby & T. Klapuri (Eds.), Bakhtin and his oth‐
ers: (Inter)subjectivity, chronotope, dialogism (pp. 105–125). London:
Anthem Press.
Tuomi, I. (1999). Corporate knowledge: Theory and practice of intelligent or‐
ganizations. Helsinki: Metaxis.
Tuomi, I. (2002). Networks of innovation: Change and meaning in the age of
the internet. Oxford: Oxford University Press.
Tuomi, I. (2012). Foresight in an unpredictable world. Technology Analysis
& Strategic Management, 24(8), 735–751. https://doi.org/10.1080/09
537325.2012.715476.
Tuomi, I. (2013). Next‐generation foresight in anticipatory organisations
(Report for the European Forum on Forward‐Looking Activities
(EFFLA)). Brussels: European Commission.
Tuomi, I. (2017). Ontological expansion. In R . Poli (Ed.), Handbook of an‐
ticipation (pp. 1–35). Cham: Sprin ger Internationa l Publishing. ht tps://
doi.org/10.1007/978‐3‐319‐31737‐3_4‐1.
Vaara, E., & Pedersen, A. (2013). Strateg y and chronotopes: A Bakhtinian
perspective on the construction of strategy narrative s. M@n@ge‐
ment, 16(5 ), 593–604. ht tps://doi.org/10 .3917/man a.165.0593 .
Vaara, E., Sonenshein, S., & Boje, D. (2016). Narratives as sources of sta‐
bility and change in organizations: A pproaches and directions for fu‐
ture research. The Academy of Management Annals, 10(1), 495–560.
htt ps://doi.org /10.1080/19416520 .2016.1120963.
von Glasersfeld, E. (1995). Radical constructivism: A way of knowing and
learning. London: The Falmer Pre ss.
Vygotsk y, L. (1986). Thought and language. Cambridge, MA: The MIT
Press.
Wilkinson, A., & Kupers, R. (2014). The essence of scenarios: Learning from
the shell experience. Amsterdam: Amsterdam University Press.
Wilkinson, A., Kupers, R., & Mangalagiu, D. (2013). How plausibilit y‐based
scenario practices are grappling with complexity to appreciate and ad‐
dress 21s t century ch allenges . Technological Forecasting and Social Change,
80(4), 699–710. https://doi.org/10.1016/j.techfore.2012.10.031.
How to cite this article: Tuomi I. Chronotopes of foresight:
Models of time‐space in probabilistic, possibilistic and
constructivist futures. Futures Foresight Sci. 2019;e11. ht t p s ://
doi.org/10.1002/ffo2.11
