Prediction and Futures Studies

views updated



Meaning. Prediction (previsione) is "seeing beforehand" how the future will be; that is, the situation that will come about in the short, medium, and long term. This, however, is a "seeing beforehand" that is not content with simply knowing what the final situation will be but it is also concerned with knowing with how it will be reached. It is concerned not only with the end-point but also with the process that the present situation will undergo to transform itself into (or remain as) the end-point. This two-part nature of prediction (end-point and development/process) enables us to take prediction into a scientific dimension, because the process by which the end-point is reached is comprehensible by means of the elaboration of a standardized method of identifying variables and their positions in a model of relationships that connect the before to the after, and carry the before to the after, and lead from the starting point to the end-point. Without this scientific attention to the passage from the former point to the latter, prediction could simply be a matter of taking the present and reading its evolution by means of an external "operator," such as a magic formula, casting stones, or the interpretation of animal innards.

This scientific attention to the process leading to the end-point—the prediction—is so crucial because by identifing elements of the process we can modify their developments to bring about the prediction we postulated. In other words, the variable now to be discovered and defined is not so much the prediction (with the extrapolative method) as the scientific process to be predicted so as to achieve a situation we desire (with the normative method).

In the light of the above, the term "prediction" as a name for this discipline is clearly less appropriate than others, such as "futurology," "future thinking," or "futures studies." A distinction may be drawn between two of these in that "futurology" is more of a disciplinary name and "futures studies" is more of an agenda of scientific activity aimed at achieving and discriminating between possible, probable, and desired futures.

In fact, if prediction is all of these things, it is of fundamental importance for taking action; that is, for deciding a priori how to achieve a given objective or to discover what a particular action will lead to.

Prediction is basically a scientific process in both of the senses just described, but it is also strongly related to other cognitive dimensions, such as ideology, ideas of utopia, action, and change. The theories underpinning prediction therefore combine or conflict with the theories and cognitive dimensions of reality. The links between these concepts should therefore be made clear.

Prediction, Utopia, and the Future of Traditional Society. A link of some kind between prediction, ideas of utopia, planning, and change certainly exists. We shall try to highlight it, starting from the problematic identified for prediction. Utopia is a "nonplace" in which is located a perfect society (perfect for he who first conceives it) dominated by a "cold synchrony," that is, by mechanical relations that serve only to maintain the utopian system without creating the emotional "warmth" or interests, including conflicting interests, whose relational outcome is unpredictable and therefore transforms the time-one system into something different from the time-zero system. Utopia is therefore the end-point of prediction, static and without relations, whose outcome cannot be predicted—an absolute end-point beyond which is nothingness. And between the present and the utopian state is "non-sense," that is, a black box whose contents do not interest us—the contents being the process that enables us to pass from the past to utopia.

What relation is there between prediction and utopia? We can interpret utopia as a residue of traditional society in which everything is tied to the past, but experienced in the pre-modern European society of the fifteenth and sixteenth centuries. In traditional society, real prediction, and therefore the future situation, is deduced by the past: Action is prescriptive, change is an aberration, organizations are similar because they all have the same structural contents and perform the same functions. Everything comes from the past: Rules are written in the past and actions are already perfected in the past. In these conditions, the process generating the future is a dejà-vu—not a perfect one, but a human condition and a destiny. The end-point that is different from such a prediction is not of this earthly condition but of the other life, in heaven or in hell. Utopia represents a sort of rebellion against the placing of heaven beyond earthly life—it is the secular dream of human omnipotence because it dares to place social perfection "not here" but nonetheless on earth. This utopia, an expression of the rebellion—no matter how fantastic—of traditional man, comes to represent a piece of "heaven" brought to earth, in which relations between people and social structures are so "sweet and delicate" as to strengthen the sweetness we have inside us rather than producing new situations and without automatically creating new equilibriums and new states. Because of all this, we can understand why constuctors of utopia do not need to know the process enabling us to pass from the present to the future (because it is a copy of the past) and why it is dominated by a static equilibrium that does not change once it has been achieved.

Thus prediction, utopia, and change in a society in which change is an aberration mean at the most bringing the perfection of the nonearthly world into the earthly world, but leaving it detached from reality, which also remains immutable. In other words, prediction is a game left to forces that are untameable and therefore ineluctable and at bottom mechanical, perpetuating positive and negative flows ad infinitum.

Prediction and the Objective as a Reference for the Plan: Ideology and the Future of Modern Society. Prediction becomes practically useful—able not only to reveal what will be but how this future may be controlled—when we lose sight of the perfections of the state we have called utopia and it takes on the role of an objective to be striven for, when an active value is ascribed to social ideals and single individuals' capacity for action. Here, society activates ideology as a resource and at the same time recognizes the ability of individual action, and above all the synthesis of individual actions, to create new situations.

Modern society thus shifts the focus from the perfection of utopia, which needs no modification since it is by definition perfect, and which is (an unreachable) vaguely defined objective, to the laboratory of process, which is concerned with relations and objectives to be achieved. If such an objective happens to be clearly defined, it is so accepted as provisional and therefore "adjustable," because certainty is only to be found in highly generic values such as justice, equality, and self-fulfillment in a fair, egalitarian, individual-enhancing society. This process of achieving the desired or probable prediction is guided by two resources activated by society. The first is ideology, the cognitive representation of the world used to guide practical action toward the objective that is the subject of prediction. The second resource is trust in the individual whose initiative may contribute to achieving the prediction, with the proviso that the action of this individual must be combined with that of other individuals to thus produce positive results for the predicted state of interpersonal and social relations.

In this view of modern society, it seems that the focus—aside from the objective defined in the probable or desired prediction that in its most complete form takes on the configuration of a plan—shifts for the most part to the process from which the prediction (and the predicted plan) emerges and thus to how this process is rationally manifested, how it may be scientifically explained, and what may be done to bend it to the achievement of the prediction.

All this comes about in modern society because change is conceived as normal, a factor built into the trajectory toward the future, tendentially and plausibly different from the present and above all from the past. All that remains of the past is the genetic origin of the present and a certain limited influence on it. If we have thus conceptually severed the deterministic link (at least in ideological terms) between past and future, modern society clearly has to focus very sharply on the processes and interdependent relations of the present in order to predict and dominate the future.

Determinism and Creativity in Prediction. To understand how prediction is to be orientated and manifested, rationality and the scientific method are essential because rationality and the scientific method provide the most effective and efficient ways of bringing about the realization of what we want to happen or what "must" happen. Hence the importance of method in obtaining a prediction and controlling it.

Methods have both deterministic components and creative components that are selected and embedded into techniques proper; these will be considered below. For the time being, it is sufficient simply to mention some features of these components. Deterministic components ground the formation of the end-point (prediction) in relations among the social, economic, environmental, and value structures defined in a model. Creative components include those that highlight the identification and pursuit of new and "invented" ways of controlling or accelerating the achievement of a prediction. Around such components, objective or subjective methodological techniques are developed that highlight the workings of a model and its simulation or formation of decisions.


The Scientific Problem of Prediction. Not only is the scientific nature of social sciences considered suspect by people outside the social sciences, but certain social scientists themselves consider the social disciplines nonscientific because the most they can do is provide a way of "reading" a social reality composed of individuals, groups, mutual relations, and formal organizations. For such people, the social sciences are not sciences but opinions. The reason is that—apart from some concepts providing interpretative keys for human, social and organizational action, mutual relations, and the products of those relations—interpretative theories stand the test of falsifiability only for a short time, often only until an event outside the phenomenon under investigation undermines the equilibrium and internal stability achieved by the phenomenon and explained by the theory. In the short term, the interpretative weakness of the theory even throws doubt on the ability of social science to explain the phenomenon. And this weakness of explanation obviously affects the strength of the prediction and consequently what has to be done to change it, that is, what is to be done to carry the present into the future.

Yet particular attention has been focused on four of the activities or purposes of which science is composed: (1) description as a pre-scientific stage and (2) explanation, (3) prediction, and (4) control as scientific activities proper. In point of fact, causal explanation is the central activity of the scientific process, since prediction is deduced from explanation and control is a "political" manipulation (and as such outside the phenomenon explained) of the variables of the explanatory model, undertaken deductively to achieve a modification of the prediction.

It may therefore be said that induction is at the root of description and explanation and deduction is at the root of prediction and control. But it is for precisely this reason that the first two activities are "more scientific," in that they are caught up in the bond between theory and theory-testing empirical research, whereas prediction and control are more rooted in utilization and change, in the final analysis in technical application. It is probably in this logic that we should see the contradiction between prediction as a science, whereby methods and techniques are elaborated as a deductive extension of methods for description and explanation or the perfection of methods, and prediction as techniques to help elites who have to make decisions to modify predictions and the explanatory picture deriving from them.

In other words, making predictions becomes scientific activity on the basis of data that are absent but that are plausible or possible or probable or desired, and whose relations may give rise to situations and scenarios that are equally possible, probable, and desired, but not certain.

The scientific nature of prediction is therefore based on rationality and the logic implicit in the links between events that have already come about and implicit in the possible reactions of, or to, a behavior that may come about. It is thus a matter of reasoning by analogy: Such-and-such has happened before in certain situations, so it may happen now in similar situations.

The scientific nature of prediction is also based on the fact that from the level of spatial analogy (if we have verified that an effect comes about here, we may infer that it will come about there in culturally analogous conditions) we may pass to the level of temporal analogy (if we have verified in the causal explanation that something comes about today, we may infer that it will come about tomorrow).

In more general terms, in prediction there is a "low-profile science," which becomes the rational study of what could happen in the future and above all how this might be more adequately dealt with so as best to guide or govern it.

Given these epistemological premises for prediction, the techniques that manifest its methodological paths are the result of the combination of certain features of the methods: qualitative and quantitative, those based on objective data or the opinions of elites (of power or knowledge), and extrapolative or normative. The predictive specificity of these three features increases from the first to the third. Quantitative and qualitative refer to the level of research and knowledge concerning a given phenomenon; the higher the level, the greater the chance of having indicators that are tried and tested and therefore more practically defined. Objective data and elite opinions are more closely tied to the usefulness ascribed to prediction. The objective datum reconstructs the model in a system, identifies the causal process and objective to which it leads, and starts from the assumption that it is "technically" possible to act on the structure of the process to modify its consequences. Leaders' opinions are privileged in that the basic assumption is that it will be their ideas and expectations, "true" or "false" as they may be, that condition, or even produce, the change in the objective/end-point. In the first of these approaches, there is extreme confidence in the scientific character of the epistemological canons of science; in the second, there is substantial lack of confidence that science can produce reality control—it becomes merely a formal exercise, though rationally useful. The extrapolative and normative features introduce action aimed at controlling the objective/end-point: Extrapolation is the projection of present processes and mechanisms into the future to postulate how it may possibly and probably be configured; the normative feature is fixing the desired future to identify the processes and mechanisms to achieve it. This is a dual approach to the future that is complementary rather than contradictory, in that its second (normative) part begins where the first (extrapolative) has served its purpose.

Table 1 shows the various techniques of prediction in relation to the three dimensions defined by the three features: qualitative–quantitative, objective–leaders' opinions, and normative–exploratory.

This plotting enables us to make some statements and develop some assessments of methods, especially of their function in establishing the various dimensions of prediction in terms of future studies.

Objective and Quantitative Techniques. Objective and quantitative techniques are common to all research. They are standardized and consolidated in practice. Scenarios, time series, causal models, simulations, and so on provide potent instruments for translating quantitative results into explained variance and high probability, although their considerable rigidity leaves them unable to cope with interference from new and sudden exogenous variables. These highly statistical methods are more effective in short-term prediction and for circumscribed rather than global events. They are also used in combined form to achieve both exploratory and normative prediction. Here the methodology may be used to explain the causal process by means of which a given trend is manifested, after which a "mission" is decided upon—a defined objective, such as a plan—and modifications are introduced into the contextual variables and their relational flows in order to achieve the objective. An example of these combined methodologies may be found in the research carried out

Table 1
methods of prediction in the combination of the three criteria
  normative methodextrapolative method
  gaming and simulationtime series
   regression analysis and canonical analysis
   econometrics and causal models
   nonlinear models
   trend impact analysis
   cross impact analysis
objective data  gaming and simulation
  relevance treesgaming and simulation
  science fiction 
  cross impact analysiscross impact analysis
   trend impact analysis
opinions of leaders   
 qualitative"expert group meetings," in-depth interviews, "genius forecasting""expert group meetings," in-depth interviews, "genius forecasting," intuitive logic
  intuitive logic 
  cross impact analysiscross impact analysis

on the Italian situation by Alberto Gasparini under the auspices of the Istituto di Sociologia Internazionale di Gorizia (ISIG) (Gasparini et al. 1983, 1988).

The first research project set out to identify how to meet the housing needs of a metropolitan area in accordance with a preestablished norm, which in this case was the habitation standard expressed in terms of the acceptable ratio between living space and family sizes. Housing needs were identified by applying several habitation standards, after which exploratory techniques were used, simulating natural demographic and social trends, simulating new housing markets and vacant housing markets, introducing factors such as filtering and vacancy chains. The result was the determination of how much of future housing needs (for example, ten years later) would remain with no intervention, that is, by leaving the area's housing situation to develop under its own impetus. Subsequently interventions were carried out on single variables: incentives for building new housing or for putting vacant housing on the market or for restoring existing housing or for providing financial help for needy families, and so on. The observation of the effects of interventions on the variables of the housing needs model indicated whether the objective had been achieved and, if it had not, provided indications as to the most appropriate modifications to be applied to single variables in order to achieve the objective.

The second project was on the quality of the environment in daily life in Italian towns. The questions were the following: What type of environment is it? How can quality be defined? What is the current state of environmental quality? How can high environmental quality be achieved in the daily life of the town? To answer these questions, prediction was developed over the following stages: (1) Quality of the environment was defined according to people's expectations in terms of services (number) and their spread or concentration in the town area, all placed in relation to individual and community values expressed by the local inhabitants. Surveys were carried out in each town by giving a questionnaire to 137 samples in as many communities, for a total of 33,000 interviewees. The result was the model of desired environmental quality for daily life as derived from subjective data converted into objective data. (2) This desired model was applied to each town (how many and what services existed and their location), which gave the model of environmental quality for daily life lived. (3) Observations were carried out on the context in which the above model was placed and the variables producing it, in order to identify the variables that influence the quality of the environment in it. These independent variables (clusters of multiple variables reduced in number by factor analysis and causally related to environmental quality through canonical analysis) represented the various features of community life: population, town territory, values, economic structure, social structure, local government, endogenous resources, exogenous resources, communications with the outside world, and so on. (4) The achievement of the desired environmental quality for daily life was explored by intervening on the variables that were causally most important for environmental quality (as they emerged from stage 3) and by simulating the effects that these interventions would have on environmental quality. This may entail further interventions on single variables until the achievement of the desired environmental quality, which is the subject of the prediction.

A third project was the definition of task environments and their dynamics in agricultural production organizations. This research was basically exploratory in nature, concluding with normative assessments. The exploration was not carried out by inquiring into how company task environments are modified over time (an inquiry into process), because the starting theory (to be subjected to verification) was that proposed by Emery and Trist (1965), whereby task environments are modified as companies expand and become increasingly causally important and disruptive, introducing irrationality into the decisions companies have to make. In this research, then, predictive exploration was not based on the projection of variables into the future, but on the investigation of two situations (small-company and large-company task environments) and their comparison in accordance with the Emery–Trist theory, reconstructing the dynamics by comparing the two potential stages of a single company that grows from a small one into a large one (Gasparini 1983). By means of defining the role of the agricultural entrepreneur and his relations with the task environment organizations, synthesising these into a few factors (by factor analysis) and linking them through canonical analysis, task environments were identified and articulated according to their influential and direct relations with the company, according to contacts not generating real influences in that relations were based on sporadic contacts, and so on. One substantial difference emerged between small and large companies. In small companies, there are few influential relations and a great many casual and sporadic contacts. In large companies, the relational task environment is very rich in relations, influences, and dependency on company decisions; the structure of the sporadic contact task environment is by contrast marked by relational links that are weak and few and far between. This exploratory projection produced by the hypothetical transformation of a small company into a large one therefore showed a radical change in the functional and power relationships of the task environment. The identification of concrete relations and contacts and their respective influence clearly leads to the intention to intervene according to the normative objective, which in this case is a rethinking of entrepreneurship, or an operational intervention in the agricultural economy to make sure that small and large entrepreneurs retain the power and responsibility assigned to them by the theory.

These three research examples show the great versatility of quantitative and objective techniques, that they need to be integrated with one another, and that they can be used in the exploratory dimension and some normative functions. These techniques are inextricably intertwined, as is exemplified by the fact that the exploratory dimension itself must be defined by reference to the criterion implicit in the normative dimension.

Leader's Opinions and Qualitative Techniques. The methods and techniques based on leaders' opinions, be they decision makers or experts in a particular field, are fundamentally qualitative in nature, that is, they are based on assessments that can be conventionally ordered in numerical values from which relations can then be highlighted. This can be done, as in the case of cross-impact analysis, but it should not be forgotten that the quantitative values manipulated are derived from percentages attributed intuitively to the occurrence of one event rather than another. Nevertheless, there are slightly differing degrees of formalization between these methods, and they are expressed in terms of their internal logic, reasoning experience that discriminates the more possible from the less possible, the ability to progressively refine judgments (the Delphi method), the compatibility between the reasoning and the context in which it is used, and the compatibility that has to give rise to the prediction for the phenomenon placed in the context.

This type of technique also contains scenarios, but they derive more from leaders' judgments than from the (highly implicit) model at the basis of the issues at stake and therefore of the variables defining the features of the model itself.

These are thus methods that can be used for the study and prediction of phenomena whose details are not known and/or which are relatively new, which means that recourse is made to qualified individuals equipped, for one reason or another, to see their own knowledge and predictions through the prism of research experience or familiarity with decision-making processes. If such is the case, the next step might be to transform the results of these subjective predictions into indicators and formal explanatory models, to be tested with exploratory methods and normative methods to obtain a (concrete) measurement of the projection or process required to achieve the predetermined objective–norm–criterion.

But it may also be the case that the simple results derived from these opinions are considered sufficient (expressed to various degrees of sophistication by means of in-depth interviews, the Delphi method, cross-impact analysis and the qualitative scenario), and this happens because, or probably because, the scientific component in the prediction is not held to be very important; it is considered as a set of rational instruments for reasoning about the plausibility of the prediction itself. Taking into account that these rationalized judgments come from policymakers (at the summit of the decision-making process) or opinion makers, this ascientific factor is even more worrisome.

In this case, the implicit conviction is that these are the players who will have a major role in the achievement of their own prediction—in which case, we are faced squarely with the principle of self-fulfilling prophecies.

Techniques in a Band of Abivalence. From Table 1 we still have to analyze the two intermediate bands in which predictive techniques are to be placed. Though these are conceptually different in some respects, they are also instrumentally contiguous, which in practice means that they often overlap, or are at least complementary, when being used. Objective qualitative methods indicate that phenomena are analyzed structurally with no measured data; methods that are quantitative but based on leaders' opinions provide judgments strongly based on facts, or at least measurable data, which involves a strong tendency to apply leaders' opinions to a concrete context.

Prediction techniques placed in these two bands of ambivalence are very similar to those devised for the predictive analysis of leaders' opinions, but they rely heavily on a detailed knowledge of context. Thus they also make use not only of the Delphi method, cross-impact analysis, and scenarios, but also simulation in objective quantitative methods. However, the most typical of these two bands are relevance trees, science fiction, and tendency impact. An example of predictive research in this context of ambivalence is condensed in what Igor Bestuzhev-Lada (1997) calls "technological prediction." It comprises seven procedures that use methods which are both quantitative and qualitative, objective and subjective. The procedures are: program elaboration, construction and analysis of the starting model, construction and analysis of the predictive background model, exploratory prediction, normative prediction, prediction verification, and formulation of recommendations for a proper management of technological prediction. Indicators are often measured quantitatively, but their treatment and assessment are mostly qualitative.

In summary, the combination of the three criteria detailed in Table 1 indicates the following:

  1. There are more specific techniques in the objective–quantitative methods and the leaders' opinions–qualitative methods.
  2. In the ambivalence band, methods that are typically objective and bound to opinion leaders tend to extend toward the quantitative and qualitative.
  3. The exploratory and normative methods are not alternatives but are fairly well integrated with one another. An exploratory projection is implicit in the normative method, and the exploratory method requires a criterion that is able to be transformed into the desired prediction–norm.
  4. Many methods are multivalent in prediction in that they are used to construct many types of prediction, but also in that they are technically versatile because they can be used with measured data and opinions alike. The most important example of this is the scenario.
  5. The effectiveness of the methods varies according to the type of prediction in question. For short-term predictions and those on a circumscribed subject, the quantitative–objective method is most effective. The longer the period involved and the broader the subject, the more effective are qualitative methods and methods based on leaders' opinion. This means that a broad to medium to long-term framework analyzed with qualitative methods contains specific short-term subjects studied with mathematical formalization.
  6. The scientific nature of prediction methods therefore varies with the variation of frames and times of reference, or at least there is a variation of the forms of expression of the scientific activities of description, explanation, and control.

From the above it thus emerges that the complexity of prediction of the future entails a multiplicity of studies because the future expresses itself in very different ways. This is why it is more than legitimate to speak in methodological terms of futures studies.


It is well known that studying the future becomes a strongly felt need in times of transition. The rules of the past no longer hold, and the rules for the future do not yet exist or are still untried. In addition, globalization accentuates the need to build niches within which some form of autonomy may be regained. But what does niche autonomy mean when globalization bombards it with the upheaval external to it, upheaval which is therefore experienced as irrational?

This question gives rise to the need to predict and to achieve prediction through studies on the future or futures. It is hardly surprising that many public and private institutions have set up study centers for prediction. An indicator of this growth is provided by the large number of Web sites for such centers, which are being put together to form a Futures Studies Internet Society.

An ISIG study (Apuzzo et al. 1999) has found that worldwide two hundred and eight institutions working in futures studies have Web sites. Of these, one hundred and forty publish papers, one hundred and twenty-three provide links, sixty-nine publish their own journals on-line and conduct training, sixty-four advertise books, fifty-seven provide on-line shopping, forty-seven pass on chat news, thirty-eight make available projects, thirty-four are concerned with software, and twenty-eight deal with methods and techniques. Most of the sites are American (one hundred and seventeen futures studies institutes) and only twenty-three are international; fourteen are in Britian; eleven in France; eight in Australia; six in Sweden; five in Canada; three each in Germany, Finland, Belgium, Norway, and Italy; two in Switzerland; and one each in Argentina, Denmark, Zaire, Austria, Israel, and Russia.

There can be no doubt that the future, especially in the United States, is strongly perceived as a subject requiring analysis. We have interpreted this as a way of finding autonomous futures for individual niches in the context of sweeping globalization. But it may also happen that despite this intention, the incipient Futures Studies Internet Society will accentuate the very standardization of feeling, thinking about, and planning the future that such great efforts are being made to curtail.


Aa.Vv 1987 Futuro e complessità. Milan: Angeli.

Apuzzo, Gian Matteo, Bruno Maltoni, and Moreno Zago 1999 "Futures Studies Bookmarks." Quadernidi Futuribili 3:98.

Bell, Wendell 1994 "Futuro." In Enciclopedia delle ScienzeSociali, vol. IV. Rome: Istituto Enciclopedia Italiana.

——1997 Foundations of Futures Studies. London: Transaction publishers. Vol. I and vol. II.

——, and J. A. Mau (eds.) 1971 The Sciology of theFuture. New York: Russell Sage Foundation.

Bestuzhev-Lada, Igor 1997 "I fondamenti metodologici della previsione tecnologica." Quaderni di Futuribili 1:110.

de Jouvenel, Bertrand 1967 The Art of Conjecture. New York: Basic Books.

Emery, F. E., and E. L. Trist 1965 "The Causal Texture of Organizational Environment." Human Relations 18:21–32.

Flechtheim, Ossip K. 1966 History and Futurology. Meisenheim am Glau, Germany: Verlag Anton Hain.

Gasparini, Alberto 1983 Ambiente operativo e aziendaagricola. Milan: Angeli.

——, Aldo de Marco, and Roberto Costa (eds.) 1988 Ilfuturo della città. Milan: Angeli.

Glenn, Jerome C. 1994 Introduction to the Futures Research Methodology Series. Washington, D.C.: Unu Millennium Project Feasibility Study.

Godet, Michel 1979 The Crisis in Forecasting and theEmergence of the Prospective Approach. New York: Unitar, Pergamon Policy Studies.

——1994 De la anticipation a la accion. Barcelona: Editorial Marcombo.

——1995 "Global Scenarios: Morphological and Probability Analysis." In 1995 profutures meeting, Ipts, Jrc, Ec: 17–30.

Gordon, Theodore J. 1992 "The Methods of Futures Research." Annals 522(July):25–35.

Irvine, J., and B. Martin 1984 Foresight in Science; Pickingthe Winners. London: Pinter Publishers.

Jantsch, Eric 1967 Technological Forecasting in Perspective. Paris: OECD.

Kahn, Herman, and Antony Wiener The Year 2000—AFramework for Speculation in the Next 33 Years. London: Macmillan.

Land, Kenneth C., and Stepeh H. Schneider (eds.) 1987 Forecasting in the Social and Natural Sciences. Dordrecht, The Netherlands: D. Reidel.

Malaska, Pentti 1995 "Survey of the Use of the Multiple Scenario Approach in Big European Companies Since 1973." In 1995 profutures meeting, Ipts, Jrc, Ec.: 42–46.

Mannermaa, Mika, Sohail Inayatullah, and Richard Slaughter (eds.) 1994 Coherence and Chaos in OurUncommon Futures. Visions, Means, Actions. Turku, Finland: World Futures Studies Federation.

Masini, Eleonora 1993 Why Futures Studies? London: Grey Seal.

——(ed.) 1994 "La previsione. Idee, protagonisti, nodi problematici." Futuribili 1:223.

——, and Giorgio Nebbia (eds.) 1997 "I limiti dello sviluppo. 1972–2022. Che cosa resta dopo 25 anni, che cosa resterà fra 25 anni." Futuribili 3:193.

McHale, John 1969 The Future of the Future. New York: George Braziller.

Meadows, Donella H., Dennis L. Meadows, Jurgen Randers, and William W. Behrens, III 1972 TheLimits to Growth. New York: Universe Books.

Norse, David 1979 "Scenario Analysis in Interfutures." Futures October.

Ortegon, Edgar, and Javier E. Medina Vasquez (eds.) 1997 Prospectiva: construccion social del futuro. Santiago de Calí, Colombia: Ilpes.

Pocecco, A. (ed.) 1994 Il futuro come progetto degli scienziatisociali. Gorizia, Italy: ISIG (Institute of International Sociology).

Polak, Fred 1971 Prognostic: A Science in the Making. Amsterdam: Elsevier.

——1973 The Image of the Future. Amsterdam: Elsevier.

Schwartz, Peter 1996 The Art of the Long View. Planningfor the Future in Uncertain World. New York: Currency Doubleday.

Slaughter, Richard A. (ed.) 1993 "The Knowledge Base of Futures Studies." Futures 25:3.

Wilson, Ian 1978 "Scenario." In Handbook of FuturesResearch. Westport, Conn.: Greenwood Press.

Alberto Gasparini