‘This business will get out of control’: What the Hunt for Red October teaches us about intelligence and uncertainty

Nothing of what you are about to see…ever happened 

We live in an age where the Cold War infrastructure of arms control is crumbling and when debate has returned to ‘great power competition’. Not only has Russia’s aggression against Ukraine, both its prelude and conduct, ensured a prominent place for intelligence in media headlines, but current challenges also include China’s ‘rise’ and ambitions towards Taiwan, Iran’s regional and nuclear policy, and North Korea’s nuclear programme. How might we present and explain the challenges inherent in using intelligence and tackling uncertainty for a sceptical or uncomprehending public? Perhaps by starting with something that never happened… 

The Hunt for Red October was released in March 1990. The Berlin Wall had fallen four months previously, the Soviet Union was crumbling and former Eastern Bloc leaders were being swept from power in competitive elections (or in the case of Nicolae Ceausescu, at the far end of a firing squad). As a film from the director of Predator and Die Hard, Red October is an undeniably great action thriller and probably the Best Submarine Film Ever.¹ But it is also a fantastic learning film touching on intelligence, deterrence, international relations, uncertainty, risk, and the classic security dilemma. While much of the film is a story about espionage, covert action, and ultimately a spectacular theft, it also contains lessons on analysis, uncertainty, communication and transparency.     

‘Engage the caterpillar drive’

For those unfamiliar with the film, some spoilers lie ahead. Set in the 1980s, the basic premise is that a Soviet submarine commander (Marko Ramius) and his senior officers steal a new prototype ballistic missile submarine equipped with a propulsion system that renders it almost silent. The US must first work out whether he is attempting to defect or launch an attack on the US and then determine what to do without triggering a war with the Soviet Union. It is a classic thriller that becomes an action movie, complete with a couple of twists and moments of high tension in confined spaces. The spoiler (you were warned!) is that Ramius is indeed attempting to defect, and we learn this fairly early on. But the Americans do not, and so much of the tension comes from the audience being in possession of more facts than most of the main characters, who then have to navigate, metaphorically and literally, the uncertainty this creates.     

‘Give him direct answers; tell him what you think’

Our protagonist, in the form of CIA analyst Jack Ryan, is presented almost as an academic rather than as some sort of black-clad covert ninja, albeit he has a background in the Marines to give him a physical ‘edge’ needed later for some action sequences. This is an all-source analyst of the kind rarely depicted well in fiction; someone who has to piece together an assessment (a judgement) on the basis of multiple types of information, usually fragmentary. All-source analysts, be they in the US system at the CIA (the Directorate of Intelligence, rather its clandestine service), the Defense Intelligence Agency, or in the UK at Defence Intelligence and the Joint Intelligence Organisation, are largely unsung heroes of intelligence work. We are accustomed to anything ‘intelligence’ being confused with espionage and thus associated with the Secret Intelligence Service (SIS). Red October is a reminder that in the real world someone else has to put together these (sometimes) contradictory pieces of information. 

Too often, the phrase ‘intelligence community’ in the UK is a shorthand for the three agencies that most are (loosely) familiar with: SIS (‘MI6’ to those weaned on Bond films), the Government Communications Headquarters (GCHQ), and the Security Service (‘MI5’, preferred to the rather more dubious alternative abbreviation). Yet these are not the totality of the UK’s intelligence apparatus, as important as covert operations, counter-intelligence, signals intelligence and human intelligence are. The war in Ukraine has seen Defence Intelligence come to prominence, given its involvement in imagery intelligence and responsibility for the bulk of the government’s all-source intelligence assessment. Both lend themselves readily to public diplomacy and explanations because they can either obscure the source of reporting or have a lesser risk of leading to harm if declassified than very sensitive single sources. 

Red October gives an example of this sort of challenge, albeit compressed over time. Ryan has an expert background, shown to us in an early sequence which gives flashes of books and computer graphics on naval capabilities and history. His interest is piqued when he is given a photo of the submarine itself, as an example of imagery obtained by a covert human source (IMINT provided by HUMINT). By the time he arrives at a briefing for the US National Security Adviser, he has further satellite imagery showing activity by the Soviet fleet, backed up with infra-red detections of their reactors (demonstrating the ability of different types of imagery to supply information beyond that detected by electro-optical sensors).

Ryan also has news the Red October has disappeared from the sonar screens of a US submarine following it, thus providing a form of measures and signature intelligence – MASINT. Additional information emerges during the briefing when the National Security Agency representative present reveals that they have picked up (presumably from some kind of interception through signals intelligence – SIGINT) the fact that Ramius has written a letter to his superiors; the contents are unknown, but shortly afterwards orders were given to scramble the ships now seen moving out of port at high speed. As an aside, this latter example might elicit knowing looks from analysts the world over. The ‘juiciest’ intelligence is sometimes withheld from desk officers until it can be revealed with a flourish during a discussion or sent on to more senior policy-makers as a means of ensuring ‘credit’ for the organisation collecting it. Bureaucratic jockeying for position is an unacknowledged influence on the analytical process, as the Butler Report into the UK’s work on WMD intelligence before the invasion of Iraq showed when it highlighted how some intelligence was briefed directly to the Prime Minister and other senior officials but not shown to the specialists who could have placed it in a proper context.²

This briefing scene neatly encapsulates the advantages and drawbacks of various types of intelligence. Ryan has photographs of the Red October, but does not know what they show. He has images of activity, but they do not explain why it is taking place. And he has a fragment of a communication, but only the context and a sequence of events rather than the detailed content. From this, and his background knowledge, he has to make inferences and draw conclusions. It is rare that our adversaries reveal their nefarious plans in a Blofeld-esque monologue, waiting to be packaged into a single human intelligence report or left lying around for electronic eavesdropping to collect. As the debate before 24 February 2022 showed, interpretation is required to make sense of the signals within the noise, and so as it is in Red October it is in the real world. We need analysts to test hypotheses, make connections and draw conclusions to navigate confusion and deal with deception.            

‘You’re just an analyst’

The sequence during the briefing where Ryan demonstrates his analytical understanding to conclude that Ramius is defecting – and puts a patronising general in his place – is a guilty pleasure for many. Many analysts have endured a lecture from a policy-maker or operator telling them that they ‘don’t understand how things work’ or have been told that assessment isn’t needed because that policy official (often senior) ‘is their own intelligence analyst’. But it also serves to highlight uncertainty. The film soon drops the pretence of ambiguity surrounding what Ramius intends. But at this point, we don’t know he is trying to defect, even though we suspect Ryan is right. What’s more, Ryan’s case is not strong, and he’s making some big assumptions, including on Ramius’s motivation based on the fact he is Lithuanian, not Russian. His comment on having met Ramius is intended to indicate his background working on the Soviet navy, but does a single meeting really give him an unassailable position?

The other question posed in the briefing is the purpose of Red October. When queried about the potential use of Red October as a ‘first strike weapon’ Ryan caveats his response: ‘that is a possibility’. The military officers in the room are more certain that ‘the goddamn thing is meant to start a war’. We later learn that Ramius defected because upon seeing the plans for the submarine, he concluded that it had ‘but one purpose’. But the film actually leaves the ‘truth’ of the matter ambiguous. It is never truly confirmed whether it was meant as a first strike weapon, or created as a ‘better’ deterrent and because the technology was available (often the way that the military makes advances). The film is right to highlight the role of Red October as more of a hypothesis than a fact, and treats it accordingly. 

In both cases, Ryan is being asked to make judgements with partial information and has compared the options in his head to try and give an idea of the likelihood of the various outcomes, an analytical technique known as ‘analysis of competing hypotheses’. Ryan is asked to prove his theory on defection, but because National Security Adviser Pelt has, as he tells Ryan, ‘to keep [his] options open’ the US must also make preparations to sink Red October. This is often the truth at the heart of assessments and how risk must be handled. Judgements can’t really predict the future, they can only give policy-makers the information on what to prioritise and how to think about preparing. Pelt can send Ryan, because the risk associated with him being right is small (if viewed as an opportunity) and because he is ‘expendable’. The gains are potentially great, but at this point only Ryan is lost if it proves to be a forlorn hope; that can be debated in light of the ensuing confrontation, of course. Meanwhile, the risk of not considering the ‘madman’ option is rather more pressing, so the US must send out the fleet accordingly. In a more simplistic film we’d simply go with the ‘knowledge’ that Ryan is right, and never mind about the other stuff, but the film uses this uncertainty as a driving force in how the characters act.

‘I give us…one chance in three’

That uncertainty and scepticism extends through to the US officers in a carrier group operating in the Atlantic and on the submarine USS Dallas, both of which are trying to find Red October. Indeed, Ryan has to ultimately use subterfuge to get the opportunity to communicate with Red October, bluffing and hoping that his guess on which direction Ramius will turn during a ‘Crazy Ivan’ manoeuvre will prove to be correct and so convince Dallas’s captain Mancuso that Ryan’s expertise is worth being heard. If the earlier briefing scene is a good example of approaching an analytical problem and conveying uncertainty to senior officials, then perhaps this scene is a worth highlighting as what not to do. Ryan is fundamentally guessing, and gambling on a 50-50 decision to try and demonstrate his credibility. We may inwardly cheer when his guess works, and it gives him the opportunity to make contact with Ramius, but in the real world this would be seen as a remarkably unprofessional risk, and any young analyst acting this way would probably be invited to re-attend training at the very least! Though to be fair, not many young analysts would find themselves on board a submerged nuclear submarine in the middle of the Atlantic. Perhaps we should accept the dramatic contrivance as entertainment in this case, especially as it gives us the memorable ‘one ping’ only scene, where the two submarines communicate via Morse code and active sonar

Conversely, Mancuso is not wrong to view the situation differently and his position is a good reminder of the pressures that commanders face; he has orders to follow, and the risk involved in not doing so is considerable. When might he again have the opportunity to almost guarantee sinking Red October? Intelligence collectors and analysts would do well to recognise that intelligence is no academic exercise, and beyond policy-makers stuck in offices, operational commanders and those on the literal front line have to make judgements in which intelligence is but one factor in their decision-making. Commanders also consider their experience, the operational advice they receive from their staff, and the situation they see unfolding in front of them. The intelligence community sometimes assumes that it represents the ‘truth’ of the world, and expects policy or activity to simply ‘follow the intelligence’. In reality, it needs to be acknowledged that understanding and judgement will also rely on priorities and an appreciation of risk. 

By this point in the film, we know what Ramius and his officers want to do, but the uncertainty seen on the Dallas is mirrored on the Soviet boat; they don’t know what the US intends, and are concerned that they might get the ‘wrong’ kind of officer on the other side: a ‘buckaroo’ as Ramius puts it. And this brings home another point about uncertainty and chance, familiar to any who have read the work of Mr and Mrs Clausewitz. Ramius has planned this heist for years. He has selected the crew and the situation, prepared accordingly and executed his part almost flawlessly. He knows how he will get his crew off Red October (something Ryan later realises that the US doesn’t need to worry about). But he can’t control everything, and can only prepare and hope for the ‘right’ person on the other side of the periscope. In both war and international relations, uncertainty and basic luck almost always play a part in any plan. It’s possible that if Ryan doesn’t make his near-suicidal trip to get on board the Dallas that Mancuso simply sinks Red October. It would make for a very unsatisfying film, but it wouldn’t be implausible. It works not just as a way of maintaining dramatic tension, but because it feels realistic. In fact, neither Ramius nor Ryan have their theories confirmed until they meet on the bridge of Red October (Mancuso retains some scepticism, and prepares accordingly³); notably, this doesn’t take place until 110 minutes into the film. Again, this is partly a desire to maintain some tension, but also an accurate reflection of the inherent difficulty in determining intent even in a world of ever-more pervasive intelligence and openly available information.     

‘Wars have begun over less, Mr Ambassador’

While those characters on the submarines are moving towards a resolution of some kind, we are reminded that the rest of the world has no good understanding of what is happening on board, and are having to prepare for other outcomes. The development of the film drives this home, even as the US begins to get a better insight into what Ramius is up to. The Soviets have ‘sortied their whole bloody fleet’ (at least in the Atlantic), and this brings up the more conventional aspects of the security dilemma. As Pelt later tells the Russian Ambassador, the US can’t tell if the massive Russian mobilisation is an exercise or a ‘prelude to war’ and therefore ‘prudence demands that we deploy our ships to observe yours’, even though having such a large concentration of forces in close proximity is ‘inherently dangerous’. In fact, the Soviet leadership knows that Ramius wants to defect, because he’s left a letter telling them this as a means of preventing his officers getting cold feet, but we don’t know what their naval forces think, and they certainly don’t know what the US thinks, leading to some lighter moments as the Russian Ambassador attempts to mislead Pelt. 

Pelt has already approved Ryan’s mission at this point, and is partly hyping up the tension to put pressure on the ambassador to extract information. But the truth is the US has deployed additional forces, and doesn’t know that this activity isn’t cover for a first strike. We are back to the uncertainty of the earlier briefing; even as Pelt explains the reality of the security dilemma to the ambassador in stark terms, the logic of the situation as seen from the US side means they feel they must chance feeding the tension, to head off the prospect of being caught unawares. 

This is true even in a supposedly more ‘transparent’ world, where we might assume that the tidal-wave of open-source information available reduces uncertainty. One potential retort to the use of Red October as a teaching tool is that as a film set in the 1980s and released over 30 years ago it doesn’t reflect modern social media use or journalistic excellence and thus is less relevant in a world of Instagram, YouTube and Bellingcat. The first defence would be that open-source information, though less prolific, was still relevant in the Cold War. Not only did Jane’s produce high-quality analysis (a copy of Jane’s Fighting Ships 1977-78 appears at the start of the film in Ryan’s book collection) but author Tom Clancy was able to write Red October based largely on publicly available information, which was accurate to the extent he was accused of having misused intelligence. The second defence would be to point out that open-source information can only take us so far, and that secret intelligence is still required to truly get inside decision-making, intent and hidden capabilities, as intelligence academic Dan Lomas recently wrote for RUSI.⁴ The scenario in Red October is one which, if reflected today, would no doubt see the use of commercial imagery from satellites or sailors posting on Facebook (much to the chagrin of their commanders!), but still focusses on the thoughts and views of a handful of people playing a dangerous game of nuclear poker. Maybe a 2023 version would include a scene in which Ramius tells his shocked officers he has just posted to Twitter/X a photo of them all marked #Defecting #EngagetheSilentDrive. However, NATO countries would still be scouring sensitive sonar systems, signals intelligence and human sources for some sort of additional insight. As the previous UK Chief of Defence Intelligence pointed out when extolling the use of open-sources around Ukraine, it complements – not replaces – secret intelligence. Though it appears likely to grow in importance and perhaps will soon be the foundation for many assessments⁵

‘Perhaps there’s another possibility we’re not considering’

The next 12 months could see a new US administration, while Europe continues to adjust to the major war raging on its doorstep and the UK attempts to implement its ‘Global Britain’ approach and the ‘refresh’ of the Integrated Review of 2021. All will probably be tested against existing points of international friction as well as unpredictable events. Investment in intelligence collection and assessment, and the development of officers and analysts who can piece together useful assessment for policy officials will therefore remain a vital part of modern national security. 

What Red October reminds us is that uncertainty and chance are a fundamental part of intelligence work, and even the very best expertise and insight can struggle in the face of gaps in knowledge: very slight differences in access to information can skew a policy response. It is striking that even in the modern case of Ukraine, where some countries – most notably the UK and US – appeared convinced of a Russian invasion, others (including possibly Ukrainian officials themselves) were more sceptical and the open-source and academic community similarly demonstrated a wide spectrum of opinions.⁶ While much of the drive since Butler has been to ‘improve’ intelligence, it’s worth recalling that a significant portion of the challenge around Iraq was around the communication of uncertainty. The setting may be that of the 1980s Cold War, but it’s as true now in trying to understand the policy of the Communist Party of China, the actions of Kim Jong-Un or the operations of the Iranian Revolutionary Guard, as it was for Soviet Russia. 

The main lessons from the film could perhaps be summarised as:

• Human factors, particularly intent, are one of the main drivers of uncertainty in intelligence. Open-source information is close to matching some collection techniques previously regarded as unique (eg satellite imagery) and the volume of information available on social media is huge. But the need to peer behind the curtain of deception, or look into closed systems means there is still a role for covertly-obtained information, even as governments can and should take advantage of the explosion of publicly-available information.
• The ability to communicate uncertainty in a meaningful sense to those taking decisions is a vital skill for analysts. Much is made of the ‘Probability Yardstick’⁷ used by organisations like Defence Intelligence in the UK, but the story doesn’t end when a mathematical value has been assigned to likelihood. Comparative descriptions and consequences are also an important part of placing uncertainty in context, and the results have to be intelligible: simply crowbarring probabilistic language into a judgement won’t always help.
• Analysts need to understand how policy officials handle risk and uncertainty when reacting to judgements. They will often have to hedge if their tolerance for risk is low, and so take action even if something is ‘unlikely’.  
• Don’t remove the safety features from your torpedoes…

Conversely, it is worth ending on a note of caution when drawing lessons from fiction, and one with a particular historical setting. The Hunt for Red October makes a virtue of expertise, both as a reflection of the professionalism of many of its characters, and as a means of conveying authenticity and ‘realism’ even in an exaggerated fictional setting. But biases affect intelligence officials too, and the human factors seen in Red October can be just as prominent in the intelligence community. Commentators with far greater knowledge than I have suggested that developments in artificial intelligence might mean we place less reliance in the future on human analysts⁸, and that technology should be player a larger role in making decisions in the future.⁹ As a result, we might see fewer plucky analysts making the difference and more amoral algorithms being the key determinant of intelligence superiority, assuming we can understand how they produce their answers, and reduce human error in the inputting of information. Recruiting and training coders, scenario-designers and mathematicians will become just as important as teaching the next generation of intelligence officers and analysts. Nonetheless, with this caveat in mind, we can learn much from entertainment such as The Hunt for Red October – and it’s still an all-time great Cold War film!¹⁰     

1. Calm down, fans of Das Boot; we all know it’s the tv series which is the best format of that masterpiece.
2. House of Commons, Review of Intelligence on Weapons of Mass Destruction, HC 898, (London: TSO, 2004), 138-139.
3. And of course, his preparation does not go to waste when a shootout occurs: the principle of Chekhov’s gun, or in this case, Mancuso’s sidearm, applies even underwater.
4. Dan Lomas, The Death of Secret Intelligence? Think Again, RUSI https://rusi.org/explore-our-research/publications/commentary/death-secret-intelligence-think-again (accessed 26 August 2023).
5. Lt-Gen Jim Hockenhull, How Open-Source Intelligence Has Shaped the Russia-Ukraine War https://www.gov.uk/government/speeches/how-open-source-intelligence-has-shaped-the-russia-ukraine-war, (accessed 26 August 2023).
6. Neveen Shaaban Abdalla, Philip H. J. Davies, Kristian Gustafson, Dan Lomas, and Steven Wagner, Intelligence and the War in Ukraine: Part 1, War on the Rocks  https://warontherocks.com/2022/05/intelligence-and-the-war-in-ukraine-part-1/, (accessed 27 August 2023).
7. Defence Intelligence – communicating probability, https://www.gov.uk/government/news/defence-intelligence-communicating-probability, (accessed 8 September 2023).
8. Keith Dear, A Very British AI Revolution in Intelligence is Needed, War on the Rocks https://warontherocks.com/2018/10/a-very-british-ai-revolution-in-intelligence-is-needed/ (accessed 10 September 2023).
9. Keith Dear, Artificial Intelligence and Decision-making, RUSI Journal 165, No. 4 (2019).
10. While doing some background reading for this article, I came across a Twitter thread on the broader international relations and national security lessons of the film, which I can highly recommend. If The Hunt for Red October isn’t part of your national security syllabus, it should be. https://twitter.com/mcopelov/status/1323096233892814848, (accessed 28 August 2023).