‘Future sign’ is an event indicating what may happen in future. Hiltunen (2008) introduced this term as a synonym to more popular, yet maybe less exact terms such as ‘weak signals’, ‘early signals’ etc. Futures studies rely on interpretation of future signs significantly, yet there is still no general theory about how to interpret them effectively. Mathematical approaches to describing functions may help create one.
‘Effective’ (interpretation) means that an intended goal is accomplished. Following the framework by Hiltunen (2008), where a future sign has three aspects: signal (a smaller change), a possible emerging issue (a bigger change), and interpretation of the signal, I propose that the general purpose of interpretation of future sign is to identify an issue (issues) behind the signal. In other words, an observer sees a signal X that can be about an issue Y. The observer needs to find out what Y is. How can we do it?
First, let’s delineate what Y is and is not. Strictly speaking, a change or a set of changes Y does not include their history or their futures, any possible predecessors or successors of Y, causes or implications. In this respect, interpretation of a future sign can be compared to diagnostics: it shall say what the current state of things is but, in its simplest version, it is not intended to say why things are the way they are or what can happen next.
Coming back to our analogy with X and Y, we can think of X as a part of Y that has already been observed, i.e. we can see a signal as a part of the issue(s) behind it. Although it is an arbitrary connection, it is quite intuitive. For example, if people are getting sick in a strange way (a signal) and it indicates a new virus spreading (an issue), it is logical and customary to consider ‘getting sick’ as a part of ‘new virus spreading’.
One of the reasons why interpretation of future signs is difficult is that one signal can point to many possible issues. Continuing an example above, getting sick can indicate a new virus, contamination of food with a harmful chemical, polluted air, magnetic storms etc etc etc. Figure 1 illustrates how a better visible curve X can hypothetically be a part of many different Ys. Knowing just X is not enough to conclude which Y is happening.
Figure 1. Knowing just X is not enough to conclude which Y is happening.
To find out what issue X signals about, i.e. to interpret X effectively, we obviously need more information. What information is necessary and sufficient for effective interpretation?
Ahlqvist & Uotila (2019) point to relational difficulties here: detecting whether and what the signal is pointing to is dependent on several relational phenomena, such as the current state and properties of the observed phenomenon, observer, observer’s perspective, context, knowledge etc.
This problem reminded me of describing mathematical functions like we did it in the secondary school. Functions can be very different, yet mathematicians have developed several approaches to creating unequivocal standard descriptions that have a necessary degree of accuracy.
Briefly, these approaches were about locating the function in the system of coordinates and describing with some partial data what happens in every part of the function. It is not necessary to know all the values that a function could take on in order to create a description that is precise enough for many practical purposes. Red elements in the Figure 2 show what we’d need to know about function Y2 so that we could begin describing it in my secondary school.
Figure 2. Red elements in the diagram show what data would help describe a function Y2.
The red elements in Figure 2 represent quantitative information about function Y2, such as limits of the function and coordinates of several points. This data would not suffice alone without qualitative information that we perceive visually from the diagram as a whole, such as knowledge about the type of function and its special properties, for example if it is symmetrical, if it grows or falls in certain intervals etc.
When we interpret an issue Y behind a signal X, we can use a similar approach. What we know in the beginning is often insufficient. If we do not rely on intuition, we need to collect more data before we can interpret a future sign effectively. Continuing an analogy with description of mathematical functions, I suggest asking the following questions.
1. Quantitative: locate the issue(s) in space and time.
a. Space: Where do the observed phenomena exist? In what locations, communities, industries etc? Where do they not? The aim of this step is to find possible borders, limits of the issue.
Other signals of the same issue may not look exactly the same as the one we have observed so far. It may be helpful to abstract from some properties of the already known signal. For example, if people begin to practice yoga more actively than they did in the past, we can make one or several hypotheses about the most essential element of this signal and check if such essential elements appear elsewhere, for example: (a) people take better care about health, (b) India and Indian things becomes fashionable, (c) yoga philosophy is getting popular, etc.
It can happen that the ‘issue’ in question is a combination of several ones and that all the essential elements contributed to the signal we observed first. Needless to say that a signal may turn out to be a ‘false alarm’, for example, a single occurrence of something that is not linked to any bigger change, or a part of normal fluctuation that was not known to the observer, etc.
To determine what is ‘essential’ in certain phenomenon, we can use methods suitable to the phenomenon and its context, e.g. logical analysis, measurements, participatory methods etc.
b. Time: When did the observed phenomena begin? Did they end? Are they continuous or occurring with some intervals? Have similar changes ever occurred before, including far past, or are they absolutely new? The aim is to describe the time borders of the issue(s).
What we do by amassing more information in this phase, in futurist terms, is collecting more ‘future signs’ (or weak signals). Knowing when and where particular instances of the change (issue) happened allows ‘connecting the dots’ and suggesting certain patterns of the change in question. Clarifying the spatial and temporal contours of the observed phenomena also helps in the next step – answering qualitative questions.
2. Qualitative: general and special properties.
a. General: What type of change is it? If no old frameworks help classify the change, what new way of classifying could be helpful?
A way to classify obviously depends on the aims of the observer. For some, it is enough to classify issues into ‘good’, ‘bad’, ‘neutral’. Others may prefer more sophisticated or specialized classifications.
b. Special: What is special or maybe unique about this change? How is it different from other similar changes?
In the qualitative research phase, we generalize the data collected during the quantitative research phase. We may need to refine the data or obtain some more by returning to the previous step one or more times.
In sum, if we choose rational methods over intuitive ones, we must come to peace with the reality: one ‘(weak) signal’ is seldom enough to make reasonably reliable conclusions about futures. To identify possible emerging change(s) behind a signal, we may need to collect more quantitative and qualitative data. By exploring spatial, temporal, general and special characteristics of these changes, we are nearing their sufficiently precise identification.
Hiltunen, E. (2008a) The future sign and its three dimensions. Futures. Vol. 40(3), 247-260.
Ahlqvist, T. – Uotila, T. (2020) Contextualising weak signals: Towards a relational theory of futures knowledge. Futures. Vol. 119, 1-12.