HOW SPIES THINK: TEN LESSONS IN
INTELLIGENCE
Date: November 23, 2021
Disclaimer: This briefing note contains the
encapsulation of views presented by the speaker
and does not exclusively represent the views of the
Canadian Association for Security and
Intelligence Studies.
KEY EVENTS
On November 23, 2021, Sir David Omand, visiting Professor in War Studies at
King’s College London and Former Director General of the Government
Communications Headquarters (GCHQ), presented on How Spies Think: Ten
Lessons in Intelligence at the 2021 CASIS West Coast Security Conference. The
presentation was followed by a question and answer period session with
questions from the audience and CASIS Vancouver executives. The key points
discussed were the role of intelligence in decision making, and the SEES
model—Situational Awareness, Explanation, Estimation and modelling, and
Strategic notice—as a valuable tool for analysts.
NATURE OF DISCUSSION
Presentation
Sir David focused his presentation on the key takeaways of his publication, How
Spies Think: Ten Lessons in Intelligence. He started by defining the purpose of
intelligence and then used it as the backdrop for the SEES model. Sir David also
suggested that every critical decision should connect the emotional needs from
the decision and the rational constraints in making a sensible decision.
Question Period
During the question and answer period, Sir David discussed the role of the
security services, effective communication of intelligence assessments to the
public, the warning nature of intelligence, and the politicisation of intelligence.
Sir David Omand 245
BACKGROUND
Presentation
Sir David began his presentation by noting that the primary purpose of
intelligence is to enable better decisions by reducing the ignorance of the decision
maker regarding what is being faced. This applies to decisions made by prime
ministers, military commanders, senior police officers, policymakers, business
leaders, etc. As proposed in his book How Spies Think: Ten Lessons in
Intelligence, whether analysts utilise secret sources or open sources, the
following questions need to be answered: what does it really mean to say that
decision makers use information to improve the quality of their decisions? How
does one do that? What is needed to know to make good decisions? What did the
intelligence community need to know to generate sensible warnings? How
confident can one be that the information on which the warning is based is
reliable?
The history of secret intelligence is littered with examples of deliberate
deception, mistaken analysis, biases, prejudices, and, arguably, cognitive
dissonance on the part of the decision makers, which lead them to ignore
intelligence assessments. Therefore, when making critical decisions, satisfying
and understanding emotional needs from the decision itself and its results must
be tied to understanding the rational constraints of making a decision.
In his book, Sir David notes that keeping passion within rational bounds has
always been hard, more so in the era of social media and today's politics, where
respect for truth is not what it used to be. For instance, falsehoods spread on
social media during the Brexit referendum and during the U.S. and France
presidential elections, as well as the deliberate attempts to widen divisions in
democratic societies. Such reality can easily generate an induced feeling of ‘I
would like that to be true’, or, by the constant social media repetition, the feeling
of ‘it might be true,’ which can easily slide into the feeling of ‘well, for me, it is
as good as true, and I will act as if it is true.’ Sir David noted that a good example
is the emotions—based on a conspiracy theory still believed by millions of
heavily armed U.S. voters—that fueled the invasion of the U.S. Capitol building
on January 6, 2021.
Arguably, the world is now in the ‘post truth social media era,’ where respecting
the value of truth no longer seems as important as saying something, even if not
strictly true, to create the desired emotional impact. Thus, Sir David proposes
four outputs that are helpful for rational decision-making: 1) situational
The Journal of Intelligence, Conflict, and Warfare
Volume 4, Issue 3
Sir David Omand 246
awareness, 2) explanation, 3) estimation and modelling, and 4) strategic notice—
(abbreviated as SEES).
Situational awareness comes from accessing data about ‘what has happened,
where, and when,’ but it is strictly related to a past event since there is always a
time delay. Intelligence reporting comes with both latency and perishability.
Time carries several meanings for intelligence officers, one of them being, ‘the
objective measure of the duration of the interval between events. Time, thus, is
associated with location in space and it can be analysed as a sequence of events
to which we can apply the Bayesian inference to relate effects-backed causes.
Therefore, to analyse an event, it is helpful to use factual questions that start with
‘what, when, and where.’ Sometimes, establishing the facts in a confused
situation makes a decision easier, and it is necessary to establish some reliable
situational awareness before leaders try to decide what to do. However, one must
be humble to recognize the first lesson in intelligence: ‘our knowledge of the
world is necessarily fragmentary, incomplete, and sometimes wrong.’ One must
not jump to inductive conclusions since data by itself is dumb and even
established facts can have multiple interpretations. This leads to the second
lesson in intelligence: ‘facts need explaining.’
Confirmation bias means that individuals are liable to choose the explanation that
best fits their preconceptions and prejudices. Therefore, it is necessary to test
alternative explanations or hypotheses against the data and look for the
explanation that has the least evidence against it instead of the one with most in
its favour. Explanation in international affairs, for example, is difficult because
analysts must possess background knowledge, foreign language skills, and a
sense of the relevant history, geography, anthropology, psychology, current
affairs, and the region. Notwithstanding, when western analysts try to see into
the mind of someone from a very different culture, including the actions of
dictators by heedless international opinion, the risks of mirror imaging and
transfer judgement are always present. However, if analysts have a sound,
evidence-based explanation of the data, it is possible to be more confident about
proceeding to estimate how events are likely to unfold and model how others
might respond to possible actions. Estimation and modelling allow a decision
maker to answer vital questions about how likely is the adversary to react if one
acts or does not act in a particular way or adopt a particular policy.
According to Sir David, the inductive fallacy of moving straight from facts to
prediction, without having a sound explanation of what is really going on under
the surface is a trap for the unwary. However, with a good explanatory model
and sufficient data, it is possible to model different outcomes on the basis of
The Journal of Intelligence, Conflict, and Warfare
Volume 4, Issue 3
Sir David Omand 247
different assumptions, including assumptions about what the adversary might
think the reaction would be to his moods. This is the application of Bayesian
inference—working forward from data that was gathered to understand the
situation that led to it and use that understanding to estimate how events may
happen. Nevertheless, it is important to remember that data requires explanation,
and if there is not a stable, satisfactory explanation or path dependency, even the
most advanced methods or advanced AI cannot forecast anything.
Lastly, with strategic notice it is possible to work backwards from the future to
the present by identifying possible longer-term developments to help
policymakers consider whether it is worthwhile to take precautionary steps or
even to take steps that might avoid that future altogether. Using imagination or
abductive reasoning provides strategic notice of many of the future developments
of interest. For example, there is strategic notice that a working quantum
computer will likely be developed within the next 5 to 10 years that works at a
large scale; the strategic notice might be that China could develop it first.
However, that is not a prediction that will definitely happen but only that it is
plausible and that such a future will pose significant risks. Therefore, armed with
that strategic notice, it is possible to mitigate the risk by investing more now into
research and by keeping a very careful intelligence eye on Chinese quantum
technology. By devoting effort to acquiring strategic notice and using it to
prepare for a possible outcome, decision makers will not be so ‘surprised by
surprise.’
Sir David finalised his presentation by pointing out that analysts cannot fully
escape their unconscious emotional framing of issues, precisely because they are
unconscious. Information is obtained through our senses and goes straight into
our minds, and without knowing it, the process of emotional framing begins. For
that reason, analysis should be a team sport so that others can spot what a single
individual cannot. Policymakers can also benefit from such an approach.
Sometimes there can be specific warning failures that fall into the cracks between
adequate foreknowledge and appropriate precautionary action, often because
policymakers fail to probe each other's position, and adopting the SEES model
makes such gaps less likely.
Question Period
During the question and answer period, Sir David touched on a few different
subjects. First, he stated that it is a mistake trying to pose human rights and
national security as a trade-off and that the trade-off is within human rights
themselves. Depending on the circumstances, the rights of a few might have to
The Journal of Intelligence, Conflict, and Warfare
Volume 4, Issue 3
Sir David Omand 248
be infringed to provide protection to the majority. This interference, of course,
should be based on reasonable grounds. A democracy operating under the straight
Rule of Law can still deliver very impressive successes. Further, Sir David added
that the security and intelligence defence community should protect these rights.
In the UK, the security services have the obligation to look at subversion, which
nowadays includes digital subversion. Therefore, their job is to investigate if
there really is something important going on, in which case the government
should take action. Although such investigations are very covert and sensitive,
they are still subject to independent oversight mechanisms.
Sir David, noted that the UK’s parliamentary oversight committee publishes
reports, which he thinks are effective, communicating to the public after a
terrorist attack what the security services knew and did not know, as well as what
is and what is not reasonable to expect from the system. Those reports foster
public support and can deal with potential conspiracy thinking that tends to
circulate after a terrorist attack takes place. A critical discussion, then, is the
consequences of a government’s decision to make an intelligence assessment
public to justify, for example, additional security measures on a rise in the
terrorist threat assessment. Arguably, there is not a safe way in which intelligence
assessments can be made public without triggering the media to interpret such
action as a way to support the government's actions, which will likely create
politicisation.
In any important decision, one must bring together two different kinds of
thinking: one is emotionally values driven, and the other is rational analysis.
Politicians presenting policies and requesting evidence to support it should be
avoided. It is a dangerous line of argument because it may lead to the presentation
of evidence that might actually not support a policy or cause it to be modified.
Rational policy making is having one’s convictions (as a person in authority
democratically elected), while also understanding the constraints or boundaries
of any analysis.
On the warning nature of intelligence, however, while an assessment does
provide a key judgement, it is not a warning. A warning is a proposed act intended
to achieve something in which the person being warned realises that something
important is happening. Thus, it has to have a claim about the world or a claim
about knowing more about the world. It has to answer the question, ‘why is that
important?’ to the person it is trying to warn, which then leads into the policy
territory. For example, if there is an international incident, warning intelligence
directed to a country that has a stake in it will be very different from that written
for a country that has nothing at stake. Therefore, there is a crossing of the
The Journal of Intelligence, Conflict, and Warfare
Volume 4, Issue 3
Sir David Omand 249
boundary between intelligence and policy implications, which is why in the UK
system, for example, having policymakers on the Joint Intelligence Committee
arguably improves the credibility of warning intelligence. Paradoxically,
however, crossing the intelligence line into the policy territory to become an
advocate of a particular point of view, might not be a recommended practice by
intelligence analysts.
KEY POINTS OF DISCUSSION
Presentation
• The primary purpose of intelligence is to enable better decisions by reducing
the ignorance of the decision maker regarding situations faced.
• For critical decisions, it is necessary to bring together two things inside a
single mind: a) an individual’s emotional needs from the decision; b) an
individual's rational constraints in making a sensible decision.
• Lesson 1 in intelligence: an individual’s knowledge of the world is
necessarily fragmentary, incomplete, and sometimes wrong.
• Lesson 2 in intelligence: facts need explaining—even established facts can
have different explanations.
• Lesson 3 in intelligence: estimates need sound explanation, adequate data,
and explicit assumptions.
• Lesson 4 in intelligence: strategic notice will arguably allow decision makers
not to be so surprised by surprises.
Question Period
• It is a mistake trying to pose human rights and national security as a trade off
when the trade-off is within human rights themselves.
• Government reports that communicate to the public what the security services
knew and did not know after a terrorist attack, as well as what is and what is
not reasonable to expect from the system can foster public support and deal
with potential conspiracy thinking that tends to circulate after such events
take place.
• Rational policy making is having one’s convictions, while also understanding
the constraints or boundaries of any analysis.
• A warning is a proposed act intended to achieve something in which the
person being warned realises that something important is happening. Thus, it
has to have a claim about the world or a claim about knowing more about the
world.
The Journal of Intelligence, Conflict, and Warfare
Volume 4, Issue 3
Sir David Omand 250
This work is licensed under a Creative Commons Attribution-Non-
Commercial-NoDerivatives 4.0 International License.
© (SIR DAVID OMAND, 2022)
Published by the Journal of Intelligence, Conflict, and Warfare and Simon Fraser
University
Available from: [Link]
The Journal of Intelligence, Conflict, and Warfare
Volume 4, Issue 3