(Rewritten and revised edition) Horizon scanning and monitoring for early warning are part of the family of activities used to foresee the future, anticipate uncertainty and manage risks. Their practice is crucial for successful strategic foresight and warning, risk management, futurism or any anticipatory activity.

While monitoring is a generic and common term used for many activities, horizon scanning is very specific and used mainly for anticipation. Horizon scanning is a term that appeared in the early years of the 21st century. It refers both to a specific tool within the strategic foresight process and to the whole anticipatory process (Habbeger, 2009).*

We shall here focus on horizon scanning as a specific tool within the entire strategic foresight process. We shall contrast it to monitoring for warning (hereafter monitoring). First, we shall present definitions for the two concepts. Then, using comparison of the practice of the two activities, we shall highlight the similarities and differences between the two. Meanwhile we shall identify best practice. Finally, we shall conclude that horizon scanning, as a tool, is, actually, the first step of any – good – monitoring for anticipation.

Definitions for horizon scanning and monitoring

Horizon Scanning

As a tool, horizon scanning allows for the identification of potential new themes or meta-issues and issues, answering our concerns as defined in our agenda or context. We shall then need to analyse in-depth the issues thus identified.

Horizon scanning looks thus for weak signals indicating the emergence of new meta-issues and issues. As a result, a scan must adopt the largest possible scope for the core question under watch.

The idea of horizon scanning is built upon older ideas and methods such as “environmental scanning,” “strategic foresight” and “indications and warning” (also labelled “strategic warning” and “warning intelligence” see Grabo, 2004). Actually, as Glenn and Gordon underline, in the 1960-1970s, most futurists used the term “’environmental scanning’. However, as the environmental movement grew, some thought the term might only refer to systems to monitor changes in the natural environment because of human actions. To avoid this confusion, futurists created various labels, such as “Futures Scanning Systems”, “Early Warning Systems” and “Futures Intelligence Systems”. The military, for its part, uses “strategic warning’ and related terms. The objective is to avoid strategic surprises (e.g. Pearl Harbour).

The English “horizon scanning” is not the same as the French “veille”, on the contrary from what some authors assert – e.g. Nicolas Charest (“Horizon Scanning” 2012 and pdf). We could best translate “veille” by “monitoring” – taken in a general way, and not more specifically for warning as here. We could also translate it as “intelligence gathering”.

Charest, actually, refers to a process: “an organised formal process of gathering, analysing and disseminating value-added information to support decision making”. Yet, this is a process from which the future and anticipation are absent. Strangely enough, the author himself underlines that the English meaning of horizon scanning implies foresight, anticipation.

Rather than conflating two practices and two words, “veille” and horizon scanning, it is necessary to distinguish both. Indeed, even though the two activities are closely related, one, horizon scanning, has to deal with the future, when the other does not have to face this challenge.

It is the anticipating quality, the necessity to “make a judgement on the future” to use Grabo’s word (Ibid.), that generates the essential difference between the two related activities.

The use of “horizon scanning” in the denomination of various governments’ offices contributed to popularise the name. For example, we had the UK Horizon Scanning Centre, created in 2004 after a call for developing such centres of excellence across government (Habbeger, 2009, p.14), or Singapore’s Risk Assessment and Horizon Scanning (RAHS) programme, launched in 2005 (Lavoix, 2010). The way the idea became fashionable also contributed to the confusion surrounding its meaning.

Monitoring for warning

Monitoring is a part of the strategic warning process. The literature on intelligence, warning and strategic surprise documents well the idea and the process. Indeed, actors have used strategic warning since at least WWII, while intelligence studies are now a constituted body of knowledge and a discipline. For further readings, there used to be an excellent bibliography of reference on intelligence related matters: J. Ransom Clark’s Bibliography on the Literature of Intelligence, notably the section on strategic warning. Unfortunately, it has been taken down. However, it can still be accessed through the internet archive, even the section on strategic warning, but with various dates which may not correspond to the latest version, now lost.

Monitoring issues will allow for the identification of warning problems. We shall then use adequate models and related indicators for the surveillance of those problems. As a reminder, an indicator is a concept and abstraction for something. An indication is the reality corresponding to the indicator at a specific instance. We thus use indicators to collect indications. For example growth of gross domestic product (GDP) is an indicator and 5% is an indication for a specific country and time. Speed can be an indicator and 60 km/h an indication on a specific place for a specific device at a specific time.

Both monitoring and surveillance lead the collection of necessary information, as defined by the model and related indicators.

As a reminder, throughout the whole SF&W process, we process to a narrowing down of our focus, which the vocabulary used reflects. We move from the most general and encompassing to the most detailed. Let us take as example energy as a “meta-issue”. Then, “issues” could be “oil security,” “peak oil,” “peak uranium,” “the volatility of oil prices,” “the politics of energy between Europe and Russia,” “energy for China,” etc. “Problems” could be the more specific “Gasprom policies,” “the Keystone pipeline,” “Energy in the Belt and Road Initiative”, or “Energy and the Belt and Road Initiative in Pakistan”, or even “tension around this or that plant”, etc.

Horizon scanning and monitoring for warning in practice

If definitions differ, is there truly a difference in the way we do horizon scanning on the one hand, monitoring for warning on the other? Is scanning included in monitoring for warning? Should we use the same processes and the same tools for scanning and for monitoring? Or do we have to use different approaches?

Similarly grounded in models, but different sophistication of models

A first difference between horizon scanning and monitoring is the location of each within the overall SF&W process. A scan is the first step of any analysis. What does that imply?

As it is the very first thing you do when tackling an issue, then scanning the horizon implicitly assumes that no understanding or little understanding of the question exists. Yet, actually this is only an appearance.

Try to make the exercise mentally: if you start looking for something, even in the loosest way, to do that you need to have an idea, even minimal, of what you are looking for. What happens is that, unconsciously, you rely on a cognitive model. This cognitive model is implicit. Thus, to scan the horizon you already use a model, even if it is a very imperfect one.

Further away in the process of foresight or risk analysis, you monitor an issue. This is meant to happen towards the end of the analytical process, thus once you know very well your topic. On the figure above, monitoring takes place after we have created the scenarios and identified the indicators for warning.

Monitoring is thus also grounded in a model. However, we have made that model explicit. We have improved and refined it through the process of analysis.

Thus, fundamentally, both horizon scanning and monitoring are similar. Their difference, here, resides actually in the sophistication of the model used, not in the actual process utilised to do scanning or the first steps of the monitoring. Hence, scanning and monitoring can utilise most often the same of tools or supports.

Broad outlook, enmeshed outputs

Second, the definition of a scan suggests that it should only identify weak signals. However, to select beforehand signals according to their strength – assuming this is possible – would be counter-productive and in some cases impossible. Indeed, a strong signal for an issue can also, sometimes, be a weak signal of emergence for something else.

Thus, when gathering signals through a scan that aims at identifying emerging meta-issues and issues, it is desirable to be as broad and encompassing as possible.

In practice, you can note new signals, and loosely start linking them to other meta-issues or issues.

Similarly, monitoring of an issue and surveillance of a problem may also pick upon signals of novel issues emerging. Again, you should make sure you record these findings.

Thus for both horizon scanning and monitoring, you need to have a cognitive make up that is as open and as broad as possible, while also, at the same time being able to link precisely this or that fact, trend or “thing” to this issue, that problem and this indicator.

Signals and their strength for horizon scanning, indications and timeline for monitoring

Last but not least, because of various biases, both analysts and clients, decision-makers and policy-makers are often unable to see, identify, and consider some signals “below the horizon.” They will be able to accept those signals only when they are “above the horizon,” which means when they are much stronger, as exemplified in the article on timeliness.

The position of the signal below or above the horizon, or the strength a signal needs to have to see actors perceiving and accepting it, will vary according to person.

It is thus not practically desirable to try sorting out signals according to their strength too early in the process.

In the case of monitoring and surveillance for warning, it is also crucial to sort the indications according to a timeline. That time sequence warns us about the evolution of the issue under watch. Finally, it will allow for the warning and its delivery. At least mentally, each indication or signal, or group of indications and signals must be positioned on their corresponding timelines. We use a plural here, because indications and signals can feed into different dynamics for various issues, as seen in the previous part.

We thus look at strength – for signals. On the other hand, we focus on timeline for indicators and their indications. Thus, does that mean that scanning and monitoring are different?

Actually, the strength of a signal for horizon scanning may be seen as nothing else than an indication of the movement of change on a timeline. Let me explain that further. If the signal is weak, then the situation is far from the actual occurrence of an event or phenomenon. On the contrary, if the signal is strong then one is close to it. A scan would thus be an instance of monitoring, where only indications leading to judgements according to which an event will not happen soon, but nevertheless deserve to be put under watch, are selected.

However, as we saw that it is neither desirable nor sometimes possible to sieve through signals according to their strength, then this vision of a scan is idealistic and impractical.

As a result, and practically, at the end of the process, a scan will gives us signals of varying strength. At that stage, we shall only have a relatively weak confidence of the very strength of the signals identified. In that case, using strength of signal would be a precursor to a much more refined judgement made in terms of timeline.

Horizon scanning thus corresponds to the first stage of monitoring (and surveillance) before judgements related to the signification of the signal, or indication in terms of timelines, are made. It thus exists not only at the very beginning of the whole SF&W process, but each time we monitor.

---

* The debate on national security is rich and features many authors. For a brief summary of and references to the many outstanding scholars who inform it, e.g. Helene Lavoix “Enabling Security for the 21st Century: Intelligence & Strategic Foresight and Warning,” RSIS Working Paper No. 207, August 2010.

---

This is the 2d edition of this article, substantially rewritten and revised from the 1st edition, June 2012.

Featured image: U.S. Navy by tpsdave. CC0 Public Domain

About the author: Dr Helene Lavoix, PhD Lond (International Relations), is the Director of The Red (Team) Analysis Society. She is specialised in strategic foresight and warning for national and international security issues. Her current focus is on Artificial Intelligence and Security.

Bibliography and References

Charest, N. (2012), “Horizon Scanning,” in L. Côté and J.-F. Savard (eds.), Encyclopedic Dictionary of Public Administration.

Gordon, Theodore J. and Jerome C. Glenn, “ENVIRONMENTAL SCANNING,” The Millennium Project: Futures Research Methodology, Version 3.0, Ed. Jerome C. Glenn and Theodore J. 2009, Chapter 2.

Grabo, Cynthia M., Anticipating Surprise: Analysis for Strategic Warning, edited by Jan Goldman, (Lanham MD: University Press of America, May 2004).

Habbegger, Beat,  Horizon Scanning in Government: Concept, Country Experiences, and Models for Switzerland,    Center    for    Security    Studies    (CSS),    ETH    Zurich,    2009.

J. Ransom Clark’s Bibliography on the Literature of Intelligence.

Lavoix, Helene, What makes foresight actionable: the cases of Singapore and Finland. (U.S. Department of State commissioned report, December 2010).

Lavoix, Helene, “Enabling Security for the 21st Century: Intelligence & Strategic Foresight and Warning,” RSIS Working Paper No. 207, August 2010 (also accessible here).