In the blind spot: Influence operations and sub-threshold situational awareness in Norway

ABSTRACT

Situational awareness in the face of hybrid campaigns is undermined by ‘blind spots’, which this article conceptualises. Blind spots are where legal or political frameworks preclude intelligence gathering below the threshold of war. Drawing on a historical case study of alterations in legislation and security police directives in Norway from 1950–2021, the analysis shows how an influence operation’s blind spot varied along with shifts in the balance between national security concerns and democratic individual rights. Recent initiatives to criminalise domestic collaboration with foreign influence operators are discussed, and the advent of AI and microtargeting presented as a challenge out of reach of such legislation.

KEYWORDS:

• Influence operations
• situational awareness
• hybrid
• liminal warfare
• intelligence
• Norway

‘It’s not the fact that we can’t connect the dots. We can’t see all of the dots’, said General Paul Nakasone, head of the U.S. Cyber Command and the National Security Agency, addressing a U.S. Senate hearing in March 2021.¹ He was pointing out a domestic ‘blind spot’ exploited by foreign actors to launch attacks from within the United States. Domestic privacy protections restrict the ability of the authorities to gather intelligence on American soil, or, in Nakasone’s case, from American Internet service providers. His testimony referred to major cyber-attacks, but the concept of blind spots holds wider relevance both thematically and geographically.² By blocking detection, blind spots impede not only the ability to uncover separate attacks but also the capacity for situational awareness in general. This article suggests a much-needed conceptualisation of such politically and legally generated blind spots.

The strategic significance of situational awareness has evolved with the past decade’s challenges to basic delineations of war and peace.³ Recently, the United States Marine Corps (USMC) and the British Ministry of Defence (MoD), two major institutions of Western military power, declared their intention to approach the future by organising for competition.⁴ At heart is their expectation that ongoing geopolitical and technological changes herald a world where different levels of conflict will be harder to discern. The emerging assumption is that any war will, at least in part, play out below what has traditionally been considered the threshold of overt, armed conflict. The ensuing complexity and ambiguity require that the capacity for situational awareness adapt accordingly. Only then may strategic decision makers ‘ascertain the adversaries’ ends’ in the manner necessary ‘to fulfil a rational calculus of one’s own ways and means’.⁵ Time is of the essence here, particularly if the attacker seeks to shape the battlespace in anticipation of escalation or to produce a fait accompli.⁶ Gaining any such ‘information advantage’, the British MoD has concluded, necessitates articulating the ‘credibility or completeness of information that is available, as well as clarifying where information is not available’.⁷ Put differently, strategic crisis management requires a continued and systematic identification of blind spots.⁸

Based on a historical case study of the conditions for detecting foreign influence operations in Norway from 1950 to 2021, this article shows how blind spots can be defined analytically and how they relate to conflict escalation.⁹ Their extent and significance is affected by three dynamic and interrelated main factors: (i) domestic political and legal frameworks, (ii) an adversary’s willingness to engage, and (iii) opportunities afforded by technology. The latter two form the backdrop for this article, and their treatment is limited to the conceptual discussion below. The article’s main topic is the first factor. This differentiates the present conceptualisation of blind spots from related variations in the existing literature on intelligence warning and horizon scanning. Whether the term chosen is ‘blindside’ or a ‘strategic shock’, the emphasis tends towards cognitive factors that negatively affect the quality of analysis.¹⁰ Thus applied, blind spots will represent inattention, poor imagination or insufficient knowledge about the strategic intent or capacity of adversaries. The present concept, in contrast, is constructed on externally observable factors that help identify sectors in which unknowns may be concealed.

The historical case study shows how changing contexts have caused alterations of domestic policies and thereby the contractions and expansions of a blind spot for locally assisted influence operations in Norway. The end of the Second World War as well as the advent of the Cold War caused a sharp contraction in the blind spot from the 1950s, first legislatively and then by way of flexible directives to the secret services. In the 1970s public criticism of police surveillance forced a limitation of mandates and thus a widening of the blind spot. Its most significant expansion followed the end of the Cold War when a relative political shift in emphasis from national security concerns to civil liberties contributed to the 2008 legalisation of domestic assistance to the influence operations of a foreign state.¹¹ Arguing that these dynamics of domestic frameworks reflect dilemmas of principle inherent to liberal democracies, rather than Norwegian particularities, the article concludes that blind spots pose a continuing risk to Western democracies more generally.

Several traits make Norway relevant for this study. Sharing a border with Russia that is proximate to the strategic capacities of its Northern Fleet, Norway risks involvement in regional conflict early on. Russian deployment of covert measures would make sense as Norway’s NATO membership increases the potential cost of crossing the threshold of overt confrontation.¹² Finally, findings about Norway should be suitable for future comparison with Western democracies marked by similar social and constitutional characteristics.

The article first discusses its approach and conceptual basis in the field concerned with ‘hybrid’ or ‘liminal’ campaigns. It goes on to present the concept of blind spots and the ways in which they relate to Norwegian emergency readiness and intelligence legislation. The conditions for and importance of detecting domestically assisted influence operations are given particular attention, including a striking Catch-22: By law the Norwegian Intelligence Service (NIS) can only gather information about its own nationals on Norwegian territory following a cabinet-level declaration of a threat to the ‘independence or security of the realm’. By implication, gaining access to the blind spot may hinge on information hidden therein. Following this discussion, the case study shows how shifting legal and political conditions affected the influence-operations blind spot in Norway from 1950 to the 2020s. Towards the end, the article presents current challenges which suggest that the blind spot may yet expand, then considers recent legislative attempts at rebalancing domestic intelligence mandates and basic civil liberties.

The case and its concepts

The historical approach of this article brings forth the enduring as well as dynamic impact of an inherent problem in any liberal-democratic state: national security policies, such as the regulation of domestic intelligence or of citizen involvement with foreign states, must be balanced with civil liberties. The literature on conflicts below the threshold seldom explores the relationship between these concerns beyond an acknowledgment that it exists. However, related long-running debates in terrorism studies have helped articulate positions that may enrich this field, as well. Two contrasting approaches of particular relevance can be distilled: One holds that constitutional rights of the individual rule out any form of state intervention but as prescribed by law. In effect this would limit domestic intelligence gathering about specific citizens to law enforcement investigations.¹³ The opposing view would argue that threats against the state require a different set of means than those available in regular crime-fighting. This could allow for domestic intelligence gathering independently of any specific suspected crime.¹⁴

Applied to real situations, the two positions are less easily separable than in the above simplification, as the limits of individual rights may themselves not be absolute. On the contrary, the limits to free speech or privacy are constantly renegotiated through the very political processes that democratic societies have set up to safeguard those rights.¹⁵ The present case study, with its broad temporal perspective, demonstrates the importance of acknowledging the dynamism of this balance. Whereas the balancing act itself may be permanent, the balancing point is not. Exploring these processes historically allows current events to be contextualised within decades-long trajectories. It also demonstrates that rebalancing and adaptation has repeatedly taken place only after the exposure of some shortcoming, as policymakers strive to catch up to the present rather than look to the future.

For its analytical framework, the article also relies on terms that belong to the ongoing effort to articulate conflict below the threshold of war. Concepts of ‘hybrid’ and ‘liminal’ campaigns are particularly relevant to the problem of situational awareness as it is presented here. As a general term, ‘hybrid’ has been linked to the operational integration of irregular means, both kinetic and otherwise.¹⁶ The deployment of such means in the ambiguous space below the threshold of overt confrontation has then been termed ‘liminal warfare’.¹⁷ A related term not employed here is that of a ‘grey zone’ or ‘grey area’.¹⁸ While this term appears to describe the same space as ‘liminal’, it lacks the latter’s utility in analysing escalation. Among other frequently used terms not applied here, ‘political warfare’ is perhaps a better fit for an analysis of offensive campaigns than it is for the present article’s interest in defensive situational-awareness processes.¹⁹

The analytical relevance of ‘hybrid’ and ‘liminal’ as denominators in this article lies in the perspectives they open upon the significance of detecting covert operations. Seen together, the concepts connect the detection of individual operations to strategic situational awareness and thereby, ultimately, to escalation control. Hybrid campaigns entail the ‘synchroni[sation of] multiple instruments of power to create effects greater than through (their separate) escalation’.²⁰ These instruments of power may be military, political, economic, civil or informational.²¹ Importantly, they are flexible parts in a whole which can be tailored to fit each campaign’s purpose and context. As a consequence, blindness to one instrument may bring about faulty analyses of these synergistic effects and thereby undermine strategic-level situational awareness: ‘[A] hybrid warfare campaign may not be seen until it is already well underway’.²² Compared to the Cold War emphasis on early warning signals emanating from military preparation, indications of hybrid threats will possibly appear as we ‘look at ourselves’ and ‘across all of society’.²³ These are arguments in support of exploring both military and civilian frameworks for detection, domestically as well as internationally. In terms of escalation control, early awareness is particularly important in the face of campaigns seeking to shape the battlespace in preparation for a more comprehensive military attack or to establish a fait accompli.²⁴

David Kilcullen’s application of the term ‘liminal’ captures the relationship between escalation control and hybrid-campaign situational awareness. He has ‘liminal’ denoting a space defined by successively higher thresholds: (i) detection, (ii) attribution and (iii) response.²⁵ Below ‘detection’ are clandestine operations known to exist only by their sponsor. Above it, but still below that of attribution, Kilcullen locates covert operations. Their existence is known to the attacked party, but not the identity of their sponsor. Climbing to the attribution threshold, the sponsor is suspected but unproven.²⁶ The ambiguity of this phase would favour the attacker, who could delay or preclude a state-to-state response by exploiting the ‘plausible’, or even ‘implausible’, ‘deniability’ of involvement.²⁷ The defending state therefore faces a persistent disincentive to act, as it risks getting blamed for any subsequent escalation. Not acting, however, means remaining at a strategic disadvantage in the ‘liminal zone’. In time, a small state with no viable way out could be left to consider escalation by summoning allies and risking a greater conflict. To gain the initiative and escape this ‘threshold manipulation’, the defending state must overcome its information deficit.²⁸

The developing scholarship on the forms of conflict alluded to above tends to include some measure of self-doubt surrounding the applicability and historical implications of key terms and concepts. The argument here will be that whether similar tactics have in fact been concocted before is somewhat beside the point. Both for the historical research problem and as a current matter of national security, the question is how and why change continues to happen. Another repeated warning is that focussing on specific ways and means, as is indeed done here, brings with it the risk of neglecting the strategic level.²⁹ This is taken into account in what follows. While the present case study emphasises the detection of one particular type of operation, the strategic level is integral to the analysis of national emergency preparedness and the blind spots which it faces.

Delineating blind spots

Offering a conceptualisation of blind spots in a nation’s emergency preparedness panorama, as it were, the discussion below amounts to a suggestion concerning their delineation. Legal and political conditions limit transparency as well as intelligence gathering, thereby defining areas within which an aggressor can operate largely unchecked. Frank Hoffman has argued that Russia has a ‘propensity to target seams and institutional gaps with its active measures’.³⁰ Thinking in terms of blind spots can help us to identify the extent of such gaps – places where the responsibility to detect or report not only lacks unequivocal assignment to one agency over another, but where nobody may be watching at all. This article applies the concept of blind spots to influence operations in particular, but it should be equally applicable to any set of threats.

As mentioned above, three factors help decide the extent of the blind spot: (i) domestic political and legal frameworks, (ii) an adversary’s willingness to engage there, and (iii) opportunities afforded by technology. The latter two appear to be increasing, as seen from the perspective of the West. The first factor, in contrast, is a function of domestic politics and the key object of analysis in this article. The three factors may vary independently, but their significance is relative to one another. Greater willingness to engage increases the significance of technological opportunity, and, importantly, of domestic frameworks that reduce the likelihood of detection. In times of increased tension, therefore, this dynamism places the domestic legal framework under pressure, challenging the balance between the population’s civil liberties and the state’s obligation to protect itself.

A blind spot concept offers at least three advantages to the study of national security in the liminal zone. Firstly, it helps operationalise the oft-mentioned limitation that Western democracies have imposed upon themselves in the shape of individual constitutional rights. Secondly, it demonstrates how tactical-level concerns – for instance, about detection – relate to the capacity for strategic decision-making. Thirdly, it incorporates historical dynamism into the analysis, thereby elucidating the causes of changing conditions for situational awareness.

Norwegian emergency preparedness below the threshold

The Constitution is the premise of all Norwegian crisis management, and therefore of the delineation of blind spots. It specifies the civil liberties and the rule of law as well as the government’s duty to ‘to engage in war in defence of the realm’.³¹ The latter implicitly requires war to represent an identifiable negation of peace.³² Accordingly, the structures of Norwegian national security crisis management appear to accommodate emergency powers only when necessary to ‘defend the realm’. There is no clear-cut recipe for the practical application of this clause, but the Emergency Preparedness Act (1950) requires that a crisis pose a threat to the ‘security of the realm’.³³ The dominant interpretations of that law have emphasised that this would entail an international dimension, or at the very least that the government’s ability to protect the country is at risk. The Norwegian Cabinet itself, at least temporarily, would make this judgment.³⁴ Doing so would require situational awareness demonstrating that someone was targeting the country to weaken its security or independence.³⁵ Notwithstanding its implicit interpretative flexibility, this lower bar could coincide with the threshold labelled ‘response’ by Kilcullen. Put differently, the lower end of emergency preparedness mandates overlaps with the high end of a crisis in the liminal zone. The dynamics of such crises, therefore, appear unlikely to correspond to the Norwegian crisis management structure tasked with their management.

The discrepancy between the Norwegian system and the shifting modes of great-power competition appears to be growing. Mark Galeotti has suggested that the ‘state-to-state contestation’ of ‘political war’ is ‘how war is fought in a new world’.³⁶ Øystein Tunsjø, considering a bipolar world dominated by the United States and China, predicts an era of ‘conflictual coexistence’.³⁷ The United States National Defense Strategy argues that America should ‘expand the competitive space’.³⁸ This ‘grey’ manner of conflict is relatively easier to meet flexibly outside of one’s own territory, whereas no comparable gradual scale of conflict applies to domestic crisis management. The only way to alter the Norwegian government’s mandates is through the Emergency Preparedness Act.³⁹

The closest the current crisis management system gets to operationalising a series of lower threshold options are the military and civilian emergency response systems.⁴⁰ They specify predefined tasks either preparing or implementing particular measures. Each task is tied to a decision maker and a legal statute.⁴¹ The latter means that these systems can only offer flexibility as previously granted by law – for instance, by the Emergency Preparedness Act.⁴²

From the outset, this would place Norway at a strategic disadvantage in the liminal zone. Moreover, a declaration that the threshold has been crossed depends on situational awareness which in turn could depend on emergency powers – for instance, to supersede restrictions on domestic intelligence gathering. A state under attack in the liminal zone could effectively be caught in a Catch-22. This point is potentially applicable to any state adhering to human rights and the rule of law, with its exact consequence dependent on each nation’s intelligence and emergency powers framework. Any decision to cross the threshold would be further complicated by the need to coordinate internationally. Maintaining escalation control demands in particular that any step would have to align with similar reactions among allied countries and their own situational understandings.⁴³

Intelligence law and the influence-operations blind spot

For the detection and attribution of influence operations, the thresholds of the Emergency Preparedness Act are of specific relevance to the intelligence sector. They are referenced in the 2021 law governing Norwegian intelligence, as their overstepping grants the King in council the right to expand the general purview of the Norwegian Intelligence Service (NIS) to gather any information that is ‘significant to the ability of the armed forces to handle enemy activity’.⁴⁴ This could include even ‘Norwegian persons’, meaning citizens and businesses on Norwegian territory. Save for these special cases, the NIS is precluded from gathering such domestic information.⁴⁵ It is strictly tasked with foreign intelligence, with exceptions for open-source information and foreign nationals on Norwegian territory.⁴⁶

Domestic intelligence-gathering concerning Norwegian persons falls within the domain of the Norwegian Police Security Service (PST).⁴⁷ Crucially, the PST is not a domestic mirror of the NIS. The NIS is tasked by law with assisting decision makers in the protection of Norwegian interests, in effect defining its scope via its purpose.⁴⁸ The PST, on the other hand, is a domestic authority and therefore bound by the constitutional principle of legality.⁴⁹ This means that the state needs a legal justification to intervene in the lives of its citizens. Consequently, the PST can only gather intelligence on the grounds that it may help to avert or uncover a specific crime. Legal activity, whether it constitutes a security threat or otherwise, falls outside the PST’s mandate. Conversely, the legality of a perceived threat is, per se, not a key concern of the NIS. The influence-operations blind spot is a result of these asymmetrical mandates.

As of 2021, Norway had no law against assisting a foreign state in carrying out influence operations.⁵⁰ This absence of a prohibition has excluded from PST’s purview Norwegian persons and businesses joining a foreign adversary in operations which are otherwise legal. As the historical analysis will show, a relevant clause was in place until 2008, when it was repealed in a major overhaul of the penal code. Since then, emergency powers have offered the only inroad, effectively delineating this blind spot via the absence of a government emergency declaration.

Information laundering for a Norwegian audience

For influence operations to have a significant impact in Norway, they would likely need domestic bridgeheads. Key tenets of Norwegian public life as well as established practices of hybrid campaigns support this assumption. From a defensive point of view, domestic detection is therefore crucial. The specific elements of a hybrid campaign against Norway will depend on ends, means and ways available to the adversary in a specific context and cannot be predicted. Yet, certain factors are difficult to disregard. For influence operations which propagate stories to Norwegians by legal means, one such factor would be the Norwegian public sphere’s enduring reliance on the Norwegian language, the domestic news media and locally organised social events.

Throughout the period studied here, local and national newspapers have remained widely read in Norway, and television and radio news have been dominated by national public broadcasting.⁵¹ The online scene is more fragmented, but even there major national and regional news outlets are hegemonic.⁵² Although the population’s English proficiency is high, news bulletins about Norway are unlikely to be widely trusted unless they are presented by the domestic media.⁵³ For an influence operation to succeed, this would suggest that communications ought to be in Norwegian and preferably presented through channels with which the target audience is already familiar. Exceptions to the rule would be where the intended audience is to be found among the limited but strongly engaged readership of the ‘alternative’ or ‘partisan’ media, whose trust depends less on the credibility often associated with established brands. However, where such media have gained a following, they have also published in Norwegian.⁵⁴

The implication of having to communicate in Norwegian and through existing media is that a foreign influence operation would require recruiting in Norway. As we have seen, such recruits would currently not constitute legitimate targets for surveillance. Yet, their role would hold great potential. Research fields that have occupied themselves with border-crossing influences offer theoretical illuminations of such processes. In his analysis of U.S. influences on post-war Europe, Rob Kroes pointed out the role of mediators who reshaped American cultural expressions, wittingly or otherwise, to integrate them into the domestic conversation and culture. The impact and ease of adoption is reinforced as ‘we make room … in a context of meaning and significance that is ours’.⁵⁵

Recently, influence operations have been studied with similar processes in mind, emphasising the importance of domestic collaboration and local communication. The concept of ‘information laundering’ (IL) has been suggested to capture the use of sequential intermediaries to legitimise deceitful information and obscure its origin.⁵⁶ They progressively distort and disseminate information, either digitally or using well-established twentieth-century methods which retain their relevance.⁵⁷ These may involve media but also local civic associations which may be exploited or set up specifically by the state performing the operations. Such ‘GONGOs’ (government-organised non-governmental organisations) comprise an established feature of Russian influence operations abroad today, as they were during the cold war.⁵⁸ The enlistment of civil society and volunteers helps to blur the distinctions among different actors and their relations with the state running the operations.⁵⁹

Leaving a blind spot for such operations has obvious repercussions for situational awareness and crisis management. Detection and attribution will have to be based on information which ends where any domestic participation begins. Were emergency powers to be invoked and the NIS to enter the blind spot, the operational consequences would still be significant. Intelligence gathering would begin from scratch in terms of both sensors and analytical preparation. Relevant units might have little prior training and lack target lists.⁶⁰ The following historical case study of Norway will show how this blind spot came to be.

Detection in Norway, 1950–2021

After the Second World War, the Norwegian government set out to reform legislation dealing with national security and emergency preparedness. Among the results was the Emergency Preparedness Act which today still governs the threshold under discussion in this article. As a ceiling for the liminal zone, it remains important, but its static nature over the last 70 years also indicates that explanations should be sought elsewhere for the significant development of the influence-operations blind spot over the same period. In the domestic legal and political domain, such insights are to be found in the changing status of Norwegian subjects suspected of assisting foreign operators. From this vantage point, the following represents a periodization of the influence-operations blind spot in Norway. Drawing on the terrorism studies distinction discussed above, it will show how the police task of preventing crime was quickly supplemented in the 1950s by directives strongly tending towards intelligence gathering. Any remaining blind spot was fast diminishing. The 1970s saw government backtrack, limiting the police to the specific enforcement of national security laws and in effect expanding the blind spot. Around the millennial turn, a penal code revision in effect instituted a far wider blind spot by repealing altogether two post-war clauses against influence collaboration.

The war itself, as well as the prosecution of Nazi collaborators in its aftermath, had formed the backdrop for the planning of several new chapters of the penal code, including clauses defining various acts of treason. The bill as introduced by the government made direct reference to the wartime realisation that previous laws had proven inadequate in the face of ‘modern warfare’.⁶¹ The distinction between military and civilian persons had been found wanting, and, along with the introduction of a more fluid delimitation of the theatre of war, this was among the reasons given for a wholesale reconsideration of the relevant laws. One could reasonably point out parallels to the twenty-first century preoccupation with ‘hybridity’. The Judicial Committee of the Norwegian Parliament mostly accepted the bill’s analysis and emphasised the increasing importance of covert operations, declaring, ‘the strategic concept of a “theatre of war” has been shattered’. The committee also anticipated that a covert phase might precede a future war: ‘Experience teaches us that an armed attack will often be prepared by saboteurs and fifth-columnists through a pre-phase that may be as decisive and fateful as the subsequent fighting itself’.⁶²

By the end of 1950, then, the penal code had been expanded, including two clauses specifically targeting what would be termed influence operations today. The first, § 97a, prohibited the reception of economic support from a foreign state ‘to influence public opinion on the form of government, on foreign policy or for the purpose of running a political party’.⁶³ In principle, a blind spot was left for any other influence purpose. Organisational collaboration as such was also targeted. The latter part of the clause intentionally impeded the Norwegian Communist Party [NKP] from receiving funds from the Soviet Union.⁶⁴

The next clause, § 97b, was directed towards those who placed national security at risk by ‘disseminating in public or to a foreign power false rumours or incorrect information’.⁶⁵ The bill explained that it was intended to protect against internal as well as external threats. Foreign adversaries as well as domestic coup-plotters might want the public to believe that the government had entered into ‘secret military agreements’ with a foreign state – for example, allowing them to establish bases on Norwegian territory. The purpose could be to trick an opponent into acting ‘rashly’ or developing a ‘false sense of security’. The tactical dimension remains pertinent: ‘In part the intention could simply be to provide an excuse for action’.⁶⁶ The continued relevance of these measures, of course, was not necessarily foreseen at the time, while the historical referent is evident. When Nazi Germany invaded Norway on 9 April 1940 it was under the false pretence that it would protect the country from an invasion by the United Kingdom. The idea that someone would attempt a coup by falsely claiming that the lawful government had lost legitimacy immediately recalled Vidkun Quisling’s speech on national radio in the evening following the German invasion. Lawmakers were looking ahead here, but their frame of reference was dominated by recent dramatic events.

This interplay between general and historical contingency was evident in the bill’s discussion of freedom of speech as well. The foreign influence clauses were considered ‘an instance where it is difficult to reconcile the demand for freedom of speech even at a time of war with the demand for national discipline and solidarity’. The importance gained by ‘psychological warfare in our time’ meant that ‘this form of treason should still receive particular emphasis’. Spreading false information about ‘enemy peace offers or disloyalty among our allies’ was considered to be ‘clearly’ within the scope of the law. The limit for culpability was drawn where the information was accurate if also ‘disheartening’ and thereby detrimental to the national effort – to punish such speech as treason was ‘going too far’.⁶⁷

The blind spot left by legislative restraint quickly shrinked as government formalised operational responsibility for domestic intelligence-gathering. The 1950 legislation had given the police specific crimes to investigate, but from 1953 their licence expanded immensely as The Ministry of Justice issued a classified directive on the organisation and mandate of the Police Surveillance Agency (POT).⁶⁸ It stated that the POT would prevent and counter all threats to the ‘internal security of the realm’, and ‘fifth columns’ were marked for surveillance in a separate clause. Specifically included were ‘persons, groups or organisations’ carrying out tasks related to ‘intelligence, propaganda, infiltration, sabotage, assassinations, etc.’.⁶⁹ Revisions from 1955 and 1959 left the clause intact.⁷⁰ According to the Lund Commission, which was set up by Parliament in 1994 to investigate the post-war practices of the intelligence and security services, this part of the directive had a clear legal consequence: it empowered the POT to gather and register even legal acts of infiltration and propaganda.⁷¹ Provided that the operational capabilities were suitably developed, the influence-operations blind spot in this period would have been modest. The 1950s POT directives remained in force until 1977.

Tightened mandates

In 1977 the mandate to pursue loosely defined ‘fifth columns’ and ‘propaganda’ was tightened. The government issued a revised directive in which those two terms were removed, in effect widening the blind spot.⁷² The new directive was unclassified and adopted by the King in council rather than issued as a secret missive to the POT from the Ministry of Justice. In retrospect, the symbolism is momentous. The greater openness as well as the explicit anchoring in the democratic process clearly signalled both reduced autonomy and increased accountability on the part of the security services. Surveillance of legal activity would be far more difficult for the POT, reflecting a change in emphasis towards law enforcement and away from intelligence.

These changes derived from political criticism which first arose in the mid-1960s. News reports and speeches by members of parliament had alleged that the POT entered into its records not only those suspected of representing a particular threat but also those merely exercising their civil right to political participation. Several of these allegations were later proven to be true. The police kept systematic records of the Norwegian Communist Party’s (NKP) candidates, party members and, in some instances, people who entered classified ads with personal messages in communist newspapers. The leadership of the Socialist People’s Party (SF) was also under surveillance.⁷³ At this time the POT supplied information for security clearances in the armed forces, but the intelligence on presumed communist sympathisers was also intended to identify potential fifth columns.⁷⁴ The surveillance of certain political parties, including eavesdropping on their national leadership, was carried out in part to investigate suspected collaboration with Soviet and East German influence operations.⁷⁵

In 1966 the prime minister of Norway, Per Borten of the agrarian Center Party, tasked an independent committee with considering the future organisation of domestic security services, and, importantly, the question of whether their ‘methods made necessary concessions to the legal rights of the individual’.⁷⁶ The committee’s report was eventually submitted to Parliament, where the Judiciary Committee found a risk that the secret services would bring upon themselves a ‘crisis of confidence’ among the general public. One antidote to this was the establishment of an intelligence oversight board. Another was the instruction that Norwegian citizens were not to be registered ‘simply on the grounds of legal political activity’.⁷⁷ The essence of this finding had been integrated into the POT’s internal regulations two years prior, in 1970, at the personal behest of the prime minister.⁷⁸

This political reaction against the secret services in the 1970s did not entail legislation to alter their mandates. Nor, argue Bergh and Eriksen in their seminal work on the domestic secret services, did it represent an unequivocal rebuke. The services’ purpose was instead reconfirmed, and their funds were increased.⁷⁹ Yet, the process did shorten the leash held by elected authorities and raise the bar for legitimate intelligence gathering domestically; many names were purged from the communist case files.⁸⁰ At the root of the changes was the growing discrepancy between the political culture of Norway in the 1960s and 1970s and the lack of transparency and democratic control which had marked the services in the early post-war era.

In institutional terms, the political pressure that mounted through the 1960s and 1970s reduced the flexibility of the executive branch to define its mission and the scope of domestic intelligence independent of the people’s representatives. When the 1977 directive removed ‘fifth columns’ and ‘propaganda’, what remained for the POT to consider were the actions still forbidden by § 97a and b. Specifically, the law’s prohibition of remunerated assistance to foreign influence operations concerned only those challenging ‘foreign policy or the system of government’. The POT directives prior to 1977 had contained no similar limitation. After 1977, anyone receiving Soviet funding to support influence operations outside the two topics stated in § 97a would now have a legitimate claim to do so freely.⁸¹ Moreover, as the clause specifically concerned influence on public opinion, it would also be legal to influence civil servants or other decision makers.⁸² The POT’s previous discretionary powers to pursue those it considered subversive had been granted by the political leadership, which was now responsible for tightening them.

The repercussions of the 1970s did not end with this tightening. The related debates and policy changes had cemented a public framing of the secret services as a risk to civil liberties in general and to the right to protest foreign policy more specifically. Just over a decade later, a lingering uneasiness would surface again.

After the Cold War

A few years into the new millennium, § 97a and b were abandoned as relics of the twentieth century. Counter-influence intelligence gathering had been curtailed in the 1970s, now the law enforcement approach was decimated as well. Three aspects of the 1990s as a historical period help to explain what turned out to be a far-reaching expansion of the influence-operations blind spot. All three are related to the end of the Cold War. The first was a national reckoning with the secret services; the second, a strengthening of liberal civil rights such as free speech; and the third, a fundamental reappraisal of threats to the national security.

The pressure against the Norwegian secret services had continued to build since the 1970s. Through the late 1980s and early 1990s, several books and news reports shocked the public by exposing the extent of political surveillance and its connections to the labour movement. With the authority of its parliamentary appointment, the Lund Commission of 1994–96 documented decades of abused trust. Some of the commission’s interpretations were themselves subjected to criticism from historians as well as observers who lived through the 1950s and 1960s, but the commission made an indelible mark on the public’s view of the POT.⁸³ Its report was also the direct reason given by the Norwegian prime minister as he set in motion the process to reform the POT, eventually turning it into the present-day PST.⁸⁴ The legal framework for military intelligence and protective security was also buttressed in the 1990s.⁸⁵ Those who had alleged wrongdoing two or three decades earlier had ‘emerged victorious’ in the eyes of the public, conclude Bergh and Eriksen.⁸⁶ The past actions of the secret services ‘did not seem to withstand the confrontation with the political and legal norms and values of the 1990s’. What the services had done, how they kept it secret and how they tried to undermine the reckoning of the 1990s eroded the public’s trust in them as the media were handed new sensations almost by the week.⁸⁷

Bergh and Eriksen point out that similar processes in several other countries were also related to a greater emphasis on personal freedoms not only in ongoing political considerations but in the law itself.⁸⁸ The anti-authoritarian currents implied in the shorthand ‘1968’ had long since spilled over into consumer-oriented, centre-right deregulation.⁸⁹ Public expectations concerning freedom of information in state entities affected even the degree to which the secret services could operate covertly.⁹⁰ In the broadest of terms, even the ongoing processes of European integration could be said to belong within this dominating tendency to confer rights on individuals and businesses. Rights-based governance tended to supplant hands-on political regulation.

The bureaucratic and legislative processes which facilitated the millennial expansion of the influence-operations blind spot simply must be seen in this wider context. Little captures the liberalising tendency of the 1990s more acutely in Norway than Ytringsfrihetskommisjonen, the independent government commission tasked with redrafting the constitutional article granting freedom of expression. As had been acknowledged by the legislators already in 1950, this right stood out among the counterweights to any law intended to hinder influence operations. The free speech commission’s highly influential white paper of 1999 recommended that constitutional protections be reinforced. It also suggested specific laws that needed changing, including penal code clauses 97a and b. For the first one, the commission recommended that the crime should reside in the secrecy of any financial aid from a foreign state, rather than in its reception as such. Open support should not be punishable, whereas ‘covert support would clearly be a problem’. For 97b the commission suggested complete repeal; it had never been used, and the concept of disinformation was ‘problematic … and open to many definitions’.⁹¹

The end of the cold war not only provided the breathing space to allow such reconsiderations of civil liberties and secret service mandates but also altered the national security calculus. Russian pressure did remain worthy of attention, but the idea of a massive military threat was increasingly considered a thing of the past. Instead, attention was being reoriented towards challenges such as terrorism, particularly following the attacks of 11 September 2001 as well as cyber threats to Norway’s increasingly ICT-dependent infrastructure.⁹² This shift affected clauses 97a and b in direct ways. Fifth columns were difficult to envisage in the absence of an enemy. One government document on the POT’s future complimented its downscaling of counter-subversion efforts. Subversive activities, including those ‘influencing certain actions or action patterns’ or ‘changing the attitudes of individuals or groups’ were no longer seen as peacetime threats: ‘Few countries can match the social and political cohesion of Norway, and our international surroundings today will not put it to any test worth mentioning’.⁹³

The fate of § 97a and b was decided in a major penal code revision which had spanned several decades only to arrive at its conclusion during this already transitional time. The chapters on national security received separate attention as the process neared its end, including that of a special committee whose white paper from 2003 explicitly engaged with and supported the main findings of the free speech commission. In charge of this committee was Supreme Court Justice Ketil Lund, who had become a household name less than a decade earlier when he led the eponymous commission investigating the secret services. In a sense, the 2003 white paper brought together the three tendencies which would eventually erase § 97a and b: the reinforcement of free speech, surveillance scepticism and the perception that Norway no longer faced a comprehensive military threat. By the time the white paper was presented for public consultation, the influence clauses had lost any remaining support. Repeal was recommended for both. With the breakthrough of social media still a few years away, the comparatively transparent public sphere appears to have been considered a sufficient line of defence: ‘Participants in the political debate will generally demonstrate their affiliations by the positions they assume. Covert financial support to [someone] is unlikely to be of any notable significance to Norwegian democracy’, the white paper reasoned.⁹⁴ Among the consultation letters submitted by government bodies, NGOs and other interested parties, only a single one mentioned this clause, supporting its repeal.⁹⁵ On 97b, a similarly lonely opinion came from the police department of Trondheim, Norway’s third largest city, arguing that ‘thinkable’ situations did exist where the clause might be justified.⁹⁶

In 2007, several years and a change of government later, the Ministry of Justice had finalised its proposal to Parliament for penal code revision. Underpinning the national security chapters was a threat assessment resembling those from earlier in the decade: ‘The cold war has ended and there is only one real military great power’.⁹⁷ On the relevance of laws to counter future influence operations, the bill quite literally planned for the last war instead: the fact that nobody had been indicted for breaking § 97a during its 50-year lifespan, ‘of which 40 were marked by the “cold war”, […] shows that there is no reason to keep it’.⁹⁸ Parliament unanimously approved the penal code revision with no particular mention of § 97 a and b.⁹⁹ As of March 2008 they were gone.

Blind spot expansion

Unbeknown to the legislators at the time, a twenty-first-century iteration of the blind spot would soon be expanding rapidly. They did not envisage the breakthrough of social media and smartphone technology a few years later, nor any resurgence of great-power competition. The Norwegian Parliament thus scrapped the influence-operations laws of the Cold War just as they were on the cusp of renewed relevance. Frank Hoffmann published his influential text on hybrid war in 2007, and Russian troops entered Georgia in 2008, which was also a signal year for data-driven political communication as Barack Obama won the U.S. presidency.¹⁰⁰ All proved to be harbingers of a complex and fast-evolving threat environment.

Over the past decade, the blind spot’s potential for harm in Norway has been exacerbated by the development of several business sectors, with consulting agencies and microtargeting particularly noteworthy. Consulting agencies help states and other clients influence both decision makers and the public, increasingly including the offer of ‘disinformation-for-hire’ services.¹⁰¹ In Norway, the revenue of PR and communication agencies increased by about 350% between 1995 and 2019,¹⁰² and they have become heavily internationalised.¹⁰³ There is no legal requirement in Norway for agencies to disclose assignments on behalf of a foreign principal, or of anyone else, for that matter. In theory, an adversary of Norway could hire a Norwegian agency to plan and conduct an influence operation without breaking any laws and without anyone having to know. Such partnerships could reach different and wider audiences than do the right- or left-wing partisan media which are more commonly associated with information laundering.¹⁰⁴

A similar disadvantage accompanies the challenge of microtargeting. Microtargeting is the use of data-driven profiling to reach an audience of individuals who are likely to respond in similar ways to similar stimuli. Messages address ‘issues which are important to an individual, adapting the format and language to meet the individual needs and interests for maximum effect’.¹⁰⁵ Artificial intelligence (AI) and the accelerating accumulation of data is assumed to significantly enhance the potential of such individualised contact, which has been pioneered in American election campaigns.¹⁰⁶ In Norway and the EU, such registration and use of personal data without the consent of the owner is lawful only for specific purposes and according to particular procedures. The mapping of individual traits such as ethnicity, language, sexual orientation, and religious and political views is particularly protected.¹⁰⁷ A foreign state without such obligations may identify and target audiences in ways that are unavailable to its opponent. Rich datasets, including those covering Norway, are already available from international vendors.¹⁰⁸

Microtargeting holds the potential of individualised contact through social media or messaging platforms. In contrast to public communication channels such as the news media or blogs, these platforms enable a foreign state to covertly approach any number of individuals according to operational relevance. Different people would receive different messages placing a conflict or growing crisis in a light which ‘nudges’ the recipient towards acting in a desired way.¹⁰⁹ The selection of individuals to be targeted would depend on the ends of the operation. Investigative journalism has shown how military intelligence and special operations personnel can be identified with relative ease.¹¹⁰ Awareness of such audience selection could help the defending state understand what sort of operation it is being subjected to. Whether the audience is comprised of national security professionals, ethnic groups, extremists, or people united by some otherwise invisible common denominator, the very knowledge that they have been contacted as a group would be valuable. One legal – if likely extremely difficult – inroad to that information for Norwegian authorities would be through intelligence gathering in the foreign state conducting the operation. An alternative route would be through the application of AI to look for patterns in online communication directed to domestic recipients.¹¹¹ Conceivably, the authorities could prepare in advance by ‘red-teaming’, in this case by simulating an adversary’s audience analysis and targeting process.¹¹² This would entail the mining of data to identify persons and groups which are susceptible to certain pressures or incitements. AI could then monitor relevant communication patterns for sudden changes. Obviously, these suggestions would be likely to breach individual rights to privacy and free speech, even at a red-teaming stage before any actual implementation of communication surveillance.¹¹³

By conducting operations in the private sphere of the targets, an attacker would gain an advantage in relation to the defending state. In regular military strategy as well as in the protection of civilian infrastructure, the identification of vulnerabilities is self-evident. Where are the weak spots in the supply of electricity or water? No comparable preparation appears possible in the face of microtargeted influence operations. Given the potential of AI and the increasing amount of available data, influence operations relying on microtargeting are likely to happen and to go unnoticed. Appropriate situational awareness is exceedingly unrealistic, as a willing adversary is able to hide in this rapidly expanding blind spot.

Legislative potential

The three factors affecting the size of the blind spot – domestic regulation, adversarial will, and technology – also mean there are three sides from which to reduce it. Two of them would depend on long-term international developments. Firstly, Norway’s potential adversaries would have to turn away from the means associated with influence operations in the liminal zone. Secondly, the technology would have to become less applicable – for example, by way of the stronger regulation of social media and data harvesting. Each of these developments requires major international efforts and reduced tensions between the great powers. Strengthened resilience would also reduce the utility of the technology but would not shrink the blind spot as such.¹¹⁴

The domestic factor differs from these, as the blind spot can be readily tightened nationally through legislation and governance. The historical analysis presented here has demonstrated that any such effort requires that individual rights will have to be balanced against national security concerns. Recent and current processes in Norway, as well as in two allied countries, affirm the continued dynamism of this dilemma. In the United Kingdom, oversight and regulation of threats to the ‘British information environment’ have been alleged to ‘lag far behind the online capability of most threat actors’.¹¹⁵ The British government has stated that legislation will be introduced in order to empower intelligence services and criminalise ‘harmful activity conducted by, and on behalf of, states’ while ‘appropriately protecting the legal rights of individuals’.¹¹⁶

Similar discussions took place in Denmark in 2018 and 2019 on the introduction of a bill to criminalise ‘cooperation’ with a foreign intelligence service for the purpose of ‘influencing decision-making or the formation of public opinion’.¹¹⁷ The bill passed the Danish Parliament but not unanimously. Those in opposition worried that freedom of speech would suffer and people would hesitate to participate in public debates, because they considered ‘influencing’ and ‘cooperation’ to be poorly defined.¹¹⁸ In addition to the recurring concern over individual rights, the Danish process again indicated a general tendency to let recent national security events affect legislation. Arguing its case, the Danish government specifically pointed to the integrity of the upcoming elections and recent experiences in the United States and fellow EU countries.¹¹⁹

In May 2021, the Norwegian government published a proposal for a law similar to that of Denmark, inviting public comments.¹²⁰ Among the motivations stated was the reenabling of the PST to handle information on foreign influence operations.¹²¹ The Danish concern that free speech should not be impeded was explicitly referenced. The act to be criminalised was cooperating with a foreign intelligence service to influence ‘decisions’ or ‘public opinion’. That is, the contents being mediated would not be of legal relevance but the perpetrator’s intent to cooperate would be.

The reintroduction of a prohibition would shrink the blind spot but not constitute a return to the status quo ante. Compared with the abandoned clauses 97a and b, the new proposal appears to be less restrictive in terms of free speech. Whereas the defunct law identified specific topics and the potentially detrimental effect of certain messages, such conditions were now explicitly left out. The definitions of the cooperative relation also differ. The defunct law did not limit the range of incriminating partners to intelligence services, as did the 2021 proposal. It did, however, limit illegal collaboration to that involving economic support, whereas the recent one placed no such condition. Whether these differences mean that the new proposal would restore lesser or greater insight than before 2008 would depend on the situation. It would almost certainly not, however, change the requirement that authorities specify suspicions of an illegal act for every instance of domestic surveillance of individuals. There is no political – perhaps neither constitutional – room for a 1950s POT-variety licence to gather general intelligence.

Most significant among the factors differentiating the effect of the proposed law from that of the old one is the contextual change. The digital revolution and the complexities of international ownership have produced a more fragmented, dynamic, and less transparent public sphere. The sender of a message, as well as its recipient, may be harder to discern from the outside. Private corporations or other mediators hiring a consultancy may fall outside the law’s purview. Unless it were demonstrably clear to the consultancy that their client acted on behalf of an intelligence service, any intent would be hard to prove. Microtargeting campaigns will be likely to escape open source intelligence. Seen together, these still developing factors imply that the pre-2008 mandates for intelligence gathering would cover relatively less ground in the 2020s blind spot than they did back then.

Conclusions

Both current and historical events serve as reminders that when human societies make laws and attempt to strike the balance between different ideals such as those of security and civil liberties, their contemporary concerns will be reflected in the outcome. In the first couple of decades after the Second World War in Norway, the balance tilted towards security. Recent initiatives to legislate against influence operations suggest that, as things are now viewed in the early 2020s, some consider the balance to have tilted too far in the other direction since the turn of the millennium.

Maintaining this balance in influence-operations legislation has been complicated further by the challenges presented by big data and individualised communication. Influence operations which make use of the public sphere exploit the freedom of expression in an open society. The use of personal data in microtargeting, on the other hand, hides itself by avoiding legal action in the open in favour of entering closed spaces where an individual is protected by law from outside intervention. The intruder weaponises the victim’s right to privacy, which keeps the authorities from detecting the crime being committed. Therefore, while the legislators in 1950 needed to find out where the exercising of free speech turned into a criminal act, the 2020s offer a more complex dilemma. Political microtargeting is already illegal, but so are important methods of its detection. This resembles the dilemma suggested by General Nakasone in the introduction to this article. Legislating against future threats combining AI, big data, microtargeting and influence operations would require balancing several scales at once, if it were possible at all.

The difficulties of legislating against influence operations have strategic ramifications. For situational awareness in the liminal zone, they may mean that the blind spot continues to grow. If technology keeps offering increasing opportunities, and adversarial ambitions do not abate, domestic legislation may not be enough to reverse the expansion of that which the authorities cannot see. In Norway, some have suggested a differentiation of the emergency preparedness threshold, although specific proposals have yet to materialise.¹²² Any such alteration would fit the emerging tendency to view geopolitics as requiring what the British Integrated Operating Concept calls ‘a more competitive approach’.¹²³ Yet, it would be likely to face the same domestic balancing challenges which have confronted the last 70 years of influence-operations legislation in Norway.

This historical dynamism shows that the asymmetry of influence operations in the liminal zone is not set in stone. Yet, the analysis also confirms that liberal democracies remain caught in their self-imposed balancing act between efforts to detect enemy operations and obligations to leave law-abiding citizens alone. This article has suggested a way of conceptualising the operational and strategic consequences of this balancing act by thinking of politics and laws as partial delineations of blind spots. Second best to situational awareness must be a realistic assessment of what cannot be observed, and at the very least where this blind spot may appear.

Acknowledgements

Funding for this research was provided by the Norwegian Ministry of Defence through the Norwegian Institute for Defence Studies’ research program Norwegian Security Policy in a Strategic Perspective. For their draft reading and valuable suggestions, the author wishes to thank Adelina T. Andersen, Kjell Inge Bjerga, Ingeborg Bjur, Sigbjørn Halsne, Nils A. Nadeau, Ida Maria Oma, Johannes Rø, Dagfinn Chr. Selvaag, Rolf Tamnes and Vivi Ringnes Wilhelmsen. Thanks also to the two anonymous reviewers for constructive and highly relevant advice.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Norwegian Ministry of Defence.

Notes on contributors

Hallvard Notaker

Hallvard Notaker is a Senior Research Fellow at the Norwegian Institute for Defence Studies and a Ph.D. in History from the University of Oslo (2008). From 2004 to 2019 he held various positions and fellowships at the University of Oslo. He is the author of several monographs, most recently a history of the Norwegian Labour Party and the 22 July 2011 terror attacks on Oslo and Utøya (Arbeiderpartiet og 22. juli (Oslo: Aschehoug, 2021)).

Notes

¹ ‘NSA Director Says Intelligence Has a Big Blind Spot: Domestic Internet Activity’, washingtonpost.com, 25 March 2021 https://www.washingtonpost.com/politics/2021/03/26/cybersecurity-202-nsa-director-says-intelligence-has-big-blind-spot-domestic-internet-activity/.

² Such attacks are commonly named after exploited companies or products, e.g., ‘SolarWinds’ (2020), ‘Microsoft Exchange Server’ (2021).

³ Patrick J. Cullen and Erik Reichborn-Kjennerud, Understanding Hybrid Warfare (London: MCDC, 2017), 21; Patrick J. Cullen, ‘Hybrid Threats as a New “Wicked” Problem for Early Warning’, Strategic Analysis 8 (May 2018); and Ministry of Defence [MoD], Information Advantage: Joint Concept Note 2/18 (Bristol: MoD, 2018).

⁴ MoD, Introducing the Integrated Operating Concept (Bristol: MoD, 2020); [USMC], MCDP 1–4 Competing (Washington, D.C.: USMC, 2020); cf. David Kilcullen, The Dragons and the Snakes: How the Rest Learned to Fight the West (London: Hurst, 2020), 151.

⁵ Robert Johnson, ‘Hybrid War and Its Countermeasures: A Critique of the Literature’, Small Wars & Insurgencies 29/1 (2018), 159.

⁶ MoD, Introducing the Integrated Operating Concept, 5.

⁷ Information Advantage: Joint Concept Note 2/18, 9.

⁸ This could be related to processes of ‘self-assessment’ (Cullen et al., Understanding Hybrid Warfare, 4) but may differ from ‘known unknowns’ and ‘unknown unknowns’ (e.g., Cullen, ‘Hybrid Threats as a New “Wicked” Problem for Early Warning’), as the object being identified is the opening for the unknown rather than its substance.

⁹ On hybrid threats and situational awareness in Norway in general, see Stein Malerud, Alf Christian Hennum and Narve Toverød, Situasjonsforståelse ved sammensatte trusler (Kjeller: FFI, 2021).

¹⁰ Peter Schwartz, Inevitable Surprises: Thinking Ahead in a Time of Turbulence (New York: Gotham Books, 2003); Francis Fukuyama, ed., Blindside: How to Anticipate Forcing Events and Wild Cards in Global Politics (Washington DC: Brookings Institution Press, 2007); Nathan Freier, Known unknowns: unconventional ‘strategic shocks’ in defense strategy development (Carlisle, Pa.: U.S. Army War College, 2008): and Cathy Downes, ‘Strategic Blind–Spots on Cyber Threats, Vectors and Campaigns’, The Cyber Defense Review 3/1 (2018), 79–104.

¹¹ The Norwegian Parliament repealed the relevant clauses in 2008, but they formally remained on the books until 2015 when the revised Penal Code of 2005 took effect.

¹² For examples from the Norwegian–Russian border area, see Kilcullen, The Dragons and the Snakes, 125.

¹³ Tending towards this view are e.g., Paul Wilkinson, Terrorism versus democracy: the liberal state response (London: Routledge, 2006), 81–83, 210; Kate Martin, ‘Domestic Intelligence and Civil Liberties’, The SAIS Review of International Affairs 24/1 (2004), 7–21.

¹⁴ Tending in this direction is e.g., K.G. Robertson, ‘Intelligence, Terrorism and Civil Liberties’, Conflict Quarterly 7/2 (1987).

¹⁵ E.g., Amy Searight, ‘Countering China’s Influence Operations: Lessons from Australia’, csis.org, 8 May 2020 https://www.csis.org/analysis/countering-chinas-influence-operations-lessons-australia.

¹⁶ Frank G. Hoffman, Conflict in the 21st Century: The Rise of Hybrid Wars (Arlington: Potomac Institute for Policy Studies, 2007); Cullen et al., Understanding Hybrid Warfare; for a critical perspective, see Donald Stoker and Craig Whiteside, ‘Blurred Lines: Gray-Zone Conflict and Hybrid War – Two Failures of American Strategic Thinking’, Naval War College Review 73/1 (2020); and Johnson, ‘Hybrid War and Its Countermeasures’.

¹⁷ David Kilcullen, ‘The Evolution of Unconventional Warfare’, Scandinavian Journal of Military Studies 2/1 (2019), 61–71; and Kilcullen, The Dragons and the Snakes, 119–120.

¹⁸ For recent definitions, see, e.g., Elisabeth Braw, The Defender’s Dilemma: Defining, Identifying, and Deterring Gray-Zone Aggression (Washington, D.C.: American Enterprise Institute, 2021); and Frank Hoffman, ‘Examining Complex Forms of Conflict Gray Zone and Hybrid Challenges’, Prism 7/4 (2018).

¹⁹ Foreign Relations of the United States, 1945–50, ‘Emergence of the Intelligence Establishment’, Document 269, George F. Kennan, ‘The Inauguration of Organized Political Warfare’, 4 May 1948 https://history.state.gov/historicaldocuments/frus1945-50Intel/d269; Mark Galeotti, Russian Political War: Moving beyond the Hybrid (London: Routledge, 2019).

²⁰ Cullen et al., Understanding Hybrid Warfare, 9–10.

²¹ Ibid. At issue here is not the question of whether this ‘instrument of power’ enumeration – as opposed to, say, ‘DIME’ or ‘DIMEFIL’ – is better suited but instead the observation that strategic effects depend upon the synchronisation of instruments.

²² see note 20 above.

²³ Cullen, ‘Hybrid Threats as a New “Wicked” Problem for Early Warning’, 4.

²⁴ E.g., Kilcullen, The Dragons and the Snakes, 69; for a discussion of such campaigns as ‘a replacement for force, not its harbinger’, see Galeotti, Russian Political War, ‘Introduction’; and Rob Johnson, ‘Hybrid Warfare and Counter-Coercion’, in The Conduct of War in the 21st Century: Kinetic, Connected and Synthetic, ed. by Rob Johnson, Martijn Kitzen and Tim Sweijs (London: Routledge, 2021).

²⁵ Kilcullen, ‘The Evolution of Unconventional Warfare’, 61–71.

²⁶ Here, attribution refers to the actual identification of a perpetrator, not to the public announcement of that identity, which would figure among the options to respond.

²⁷ Rory Cormac and Richard J. Aldrich, ‘Grey Is the New Black: Covert Action and Implausible Deniability’, International Affairs 94/3 (2018), 477–494; and Kilcullen, The Dragons and the Snakes, 154.

²⁸ Quoted from Kilcullen, The Dragons and the Snakes, 150.

²⁹ Nicole Jackson, ‘Deterrence, Resilience and Hybrid Wars: The Case of Canada and NATO’, Journal of Military and Strategic Studies 19/4 (2019); and Johnson, ‘Hybrid War and Its Countermeasures’.

³⁰ Hoffman, ‘Examining Complex Forms of Conflict Gray Zone and Hybrid Challenges’, 42.

³¹ The Constitution of the Kingdom of Norway, § 26, https://lovdata.no/dokument/NLE/lov/1814-05-17.

³² For a defence of ‘war’ as traditionally defined, see Stoker et al., ‘Blurred Lines’.

³³ Lov om særlige rådgjerder under krig, krigsfare og liknende forhold (1950) [aka ‘Beredskapsloven’], § 3; Sigrid Redse Johansen, ‘Krig på norsk jord: Hvilken rett gjelder?’ and Sigmund Simonsen, ‘Regjeringen og Forsvarets rettslige handlerett og handleplikt i nasjonale kriser’, in Strategisk ledelse i krise og krig: Det norske systemet, ed. Ann Karin Larssen and Gjert Lage Dyndal (Oslo: Universitetsforlaget, 2020); for a comparison with neighbouring countries, see N[orske] O[ffentlige] U[tredninger] [Eng.: Norwegian Official Report] 2019:13 Når krisen inntreffer, ch. 10.

³⁴ NOU 1995:31 Beredskapslovgivningen i lys av endrede forsvars- og sikkerhetspolitiske rammebetingelser, ch. 4.5.5.2; cf. Ragnar Auglænd, ‘Fordelingen av ansvar og myndighet mellom politi og forsvar ved indre beredskap’, Tidsskrift for strafferett, 15/3 (2015), 335; St.meld. no. 39 (2003–2004) [Eng.: Government Report to Parliament], 22; and Gro Nystuen, ‘Terrorbekjempelse og folkerettslige normkonflikter’, Nordic Journal of Human Rights 20/3 (2002), 21–23.

³⁵ NOU 1995:31, ch. 4.5.5.2.

³⁶ Galeotti, Russian Political War, ‘Introduction’.

³⁷ Øystein Tunsjø, ‘USA og Kina står foran en superkonflikt’, Dagens Næringsliv, 16 August 2020 https://www.dn.no/innlegg/superkonflikt/kald-krig/geopolitikk/innlegg-usa-og-kina-star-foran-en-superkonflikt/2-1-857459

³⁸ Department of Defense, Summary of the 2018 National Defense Strategy of the United States of America (Washington, D.C.: Department of Defense [n.d.]), 4–5. https://dod.defense.gov/Portals/1/Documents/pubs/2018-National-Defense-Strategy-Summary.pdf; see also MoD, Introducing the Integrated Operating Concept, 1; USMC, CDP 1–4 Competing.

³⁹ NOU 2019:13 Når krisen inntreffer, ch. 6.5, 8.4.

⁴⁰ Sivilt beredskapssystem (SBS) and Beredskapssystem for Forsvarssektoren (BFF).

⁴¹ Paal S. Hilde, ‘Bistand fra NATO og allierte: Norge utløser artikkel 4 og 5’, in Larssen et al., Strategisk ledelse i krise og krig.

⁴² NOU 1995:31, ch. 4.4.

⁴³ On the role of bilateral and NATO consultations and support in Norwegian crisis management, see Hilde, ‘Bistand fra NATO og allierte’.

⁴⁴ Lov om Etterretningstjenesten, § 4–2.

⁴⁵ Open-source information is allowed. Lov om Etterretningstjenesten, § 4–1.

⁴⁶ Operations targeting the latter require coordination with the Norwegian Police Security Service. Lov om Etterretningstjenesten, §§ 2–1, 4–3.

⁴⁷ No. Politiets sikkerhetstjeneste.

⁴⁸ For a more precise delimitation, see Lov om Etterretningstjenesten, § 3–1 to 3–5.

⁴⁹ The Constitution of the Kingdom of Norway, § 113.

⁵⁰ Provided the influence operation does not consist of acts which are themselves illegal – for example, assisting foreign intelligence operations or committing sabotage, terrorism or manipulation of election results. See particularly chapters 17 and 18 of the penal code. Lov om straff, https://lovdata.no/dokument/NL/lov/2005-05-20-28.

⁵¹ Henrik G. Bastiansen and Hans Fredrik Dahl, Norsk mediehistorie (Oslo: Universitetsforlaget, 2008), chapters 5–7.

⁵² Hallvard Moe and Hilde Sakariassen, ‘Norway’, in Digital News Report 2019 (Oxford: Reuters Institute, 2019), 98–99.

⁵³ A similar point is made about Germany in Jeffrey Mankoff, ‘Russian Influence Operations in Germany and Their Effect’, csis.org, 3 February 2020 https://www.csis.org/analysis/russian-influence-operations-germany-and-their-effect.

⁵⁴ Bente Kalsnes, Falske nyheter: Løgn, desinformasjon og propaganda i den digitale offentligheten (CappelenDamm Undervisning, 2019), 69.

⁵⁵ Rob Kroes, If You’ve Seen One You’ve Seen the Mall (Urbana: University of Illinois Press, 1996), 176.

⁵⁶ Used in this context by, for example, Belén Rodríguez, Information Laundering in Germany (Riga: NATO Strategic Communications Centre of Excellence, 2020); see also James Pamment, Howard Nothhaft, Henrik Agardh-Twetman and Alicia Fjällhed, Countering Information Influence Activities: The State of the Art (Lund: Lund University, 2018), 73; the concept was previously developed for another context in Adam Klein, ‘Slipping Racism into the Mainstream: A Theory of Information Laundering’, Communication Theory 22 (2012).

⁵⁷ Thomas Rid, Active Measures: The Secret History of Disinformation and Political Warfare (New York: Farrar, Straus and Giroux, 2020).

⁵⁸ Rid, Active Measures; Trond Bergh and Knut Einar Eriksen, Den hemmelige krigen: overvåking i Norge, vol. 2, Storhetstid og stormkast: 1955–1997 (Oslo: Cappelen Akademisk, 1998); Daniel Kliman, Andrea Kendall-Taylor, Kristine Lee, Joshua Fitt and Carisa Nietsche, Dangerous Synergies: Countering Chinese and Russian Digital Influence Operations (Washington, D.C.: Center for a New American Security, 2020), 12; and Ieva Bērziņa, Māris Cepurītis, Diana Kaljula and Ivo Juurvee, Russia’s Footprint in the Nordic–Baltic Information Environment (Riga: NATO Strategic Communications Centre of Excellence, 2018).

⁵⁹ Samantha Bradshaw, Hannah Bailey and Philip N. Howard, Industrialized Disinformation: 2020 Global Inventory of Organized Social Media Manipulation (Oxford: University of Oxford, 2021), https://demtech.oii.ox.ac.uk/wp-content/uploads/sites/127/2021/01/CyberTroop-Report20-FINALv.3.pdf.

⁶⁰ Foreign partners may have such information, which could be lawful to handle nationally upon the declaration of a national security emergency.

⁶¹ Ot.prp. 79 (1950–51), Norwegian Parliament (Proposition [bill] to the Odelsting).

⁶² Innst. O. XVI (1950–51), Judicial Committee, Norwegian Parliament (Committee recommendation [bill] to the Odelsting).

⁶³ Lov om straff, https://lovdata.no/lov/1902-05-22-10/§97a.

⁶⁴ Ot.prp. 79 (1950–51), 13.

⁶⁵ Lov om straff, https://lovdata.no/lov/1902-05-22-10/§97b.

⁶⁶ Ot.prp. 79 (1950–51).

⁶⁷ Ibid.

⁶⁸ No. ‘Politiets Overvåkingstjeneste’. The current acronym PST went into use in 2002.

⁶⁹ Dok. 15 (1995–96) (aka ‘Lund-rapporten’), Norwegian Parliament, 203–204.

⁷⁰ Dok. 15 (1995–96), 205.

⁷¹ Ibid, 204.

⁷² Bergh et al., Den hemmelige krigen, 2:100.

⁷³ Dok. 15 (1995–96), 205; and Bergh et al., Den hemmelige krigen, 2:92.

⁷⁴ Bergh et al., Den hemmelige krigen, 2:103.

⁷⁵ Bergh et al., Den hemmelige krigen, 2:108–09, 137–39, 392.

⁷⁶ St.meld. 89 (1969–70), Norwegian Parliament.

⁷⁷ Innst. S. 164 (1971–72) (Parliamentary committee recommendation), Norwegian Parliament.

⁷⁸ Bergh et al., Den hemmelige krigen, 2:97.

⁷⁹ Bergh et al., Den hemmelige krigen, 2:21, 23.

⁸⁰ Bergh et al., Den hemmelige krigen, 2:93.

⁸¹ Unless violating other laws. Relations to foreign states could still be recorded as part of personal security clearance processes.

⁸² Anders Bratholm and Magnus Matningsdal (ed.), Straffeloven med kommentarer: Anden del: Forbrydelser (Oslo: Universitetsforlaget, 1995).

⁸³ Bergh et al., Den hemmelige krigen, 2:533–537; and Synstnes, Den innerste sirkel, 280–281.

⁸⁴ NOU 1998:4 Politiets overvåkingstjeneste, ch. 3.4.

⁸⁵ Lov om Etterretningstjenesten, 20 March 1998; Lov om forebyggende sikkerhetstjeneste, 20 March 1998; Synstnes, Den innerste sirkel, 285–287.

⁸⁶ Bergh et al., Den hemmelige krigen, 2:533.

⁸⁷ Ibid., 535.

⁸⁸ Ibid., 533–534.

⁸⁹ On Norway, see Hallvard Notaker, Høyres historie 1975–2005: Opprør og moderasjon (CappelenDamm, 2012).

⁹⁰ Bergh et al., Den hemmelige krigen, 2:534; and Synstnes, Den innerste sirkel, 326–327.

⁹¹ NOU 1999:27 ‘Ytringsfrihed bør finde Sted’, 197.

⁹² E.g., NOU 2003:18 Rikets sikkerhet, 15–16.

⁹³ NOU 1998:4, 7.9; see also St.meld. no. 26 (2003–2004).

⁹⁴ NOU 2003:18, 29.

⁹⁵ Norsk forening for kriminalreform (KROM) [Norwegian Association for Criminal Reform]. Ot.prp. no. 8 (2007–2008).

⁹⁶ St.meld. no. 26 (2003–2004).

⁹⁷ Ot.prp. no. 8 (2007–2008), ch. 7.3.1.1; cf. NOU 2003:18; St.meld. no. 17 (2001–2002) Samfunnssikkerhet.

⁹⁸ Ot.prp. no. 8 (2007–2008), ch. 7.12.2.6.

⁹⁹ O.tid. 12.02.2008, 263; L.tid. 28.02.2008, 12 (Parliamentary record).

¹⁰⁰ Hoffman, Conflict in the 21st Century; Daniel Kreiss, Taking Our Country Back (Oxford: Oxford University Press, 2012).

¹⁰¹ Daniel Dobrowolski, David V. Gioe and Alicia Wanless, ‘How Threat Actors Are Manipulating the British Information Environment’, RUSI Journal 165/3 (2020), 22–23; Bradshaw et al., Industrialized Disinformation.

¹⁰² Adjusted for inflation; to be treated as an estimate, because the numbers come from more than one source. Sigurd Allern, Når kildene byr opp til dans (Oslo: Pax, 1997), 99; Statistics Norway, Business statistics, table 12910.

¹⁰³ Ole Janszo (ed.), Bransjerapporten 2019 (Oslo: Teft, 2019).

¹⁰⁴ Rodríguez, Information Laundering; Kalsnes, Falske nyheter, 57–74.

¹⁰⁵ F. J. Zuiderveen Borgesius et al., ‘Online Political Microtargeting: Promises and Threats for Democracy’, Utrecht Law Review 14/1 (2018), 82–96.

¹⁰⁶ Flemming Splidsboel Hansen, Trying to Get What You Want: Russian Influence Operations (Copenhagen: Danish Institute for International Studies, 2018); Sasha Issenberg, The Victory Lab (New York: Broadway, 2013).

¹⁰⁷ For police purposes, see Politiregisterloven; for the NIS, see Lov om etterretningstjenesten; for the civilian sector, see the General Data Protection Regulation (GDPR), interpreted for political communication in Norway in Datatilsynet, På parti med teknologien (Oslo: Datatilsynet, 2019), https://www.datatilsynet.no/globalassets/global/dokumenter-pdfer-skjema-ol/rettigheter-og-plikter/rapporter/pa-parti-med-teknologien.pdf; see also Borgesius et al., ‘Online Political Microtargeting’.

¹⁰⁸ Henrik Twetman and Gundars Bergmanis-Korats, Data Brokers and Security (Riga: NATO Strategic Communications Centre of Excellence, 2021).

¹⁰⁹ Henrik Skaug Sætra, ‘When Nudge Comes to Shove: Liberty and Nudging in the Era of Big Data’, Technology in Society, 59 (2019), 2.

¹¹⁰ Twetman et al., Data Brokers and Security, 25; ‘Når mobilen blir fienden’, nrk.no, 18 May 2020 https://www.nrk.no/norge/xl/norske-offiserer-og-soldater-avslort-av-mobilen-1.14890424.

¹¹¹ Njord Wegge and Thorsten Wetzling, ‘Countering Hybrid Threats through Signals Intelligence and Big Data Analysis?’, in Intelligence Relations in the 21st Century, ed. Tom Røseth and John Michael Weaver (Cham: Palgrave Macmillan, 2020); and Alexander Babuta, Marion Oswald and Ardi Janjeva, Artificial Intelligence and UK National Security: Policy Considerations. (London: RUSI, 2020), 27.

¹¹² Twetman et al., Data Brokers and Security, 35.

¹¹³ Not only domestic law would apply here – in 2020, the Norwegian government postponed the implementation of parts of the Intelligence Service Act, awaiting analysis of decisions by the European Court of Human Rights and the Court of Justice of the European Union (Ministry of Defence, ‘Utsetter ikrafttredelse av to kapitler i e-loven’, regjeringen.no, 10 November 2020 https://www.regjeringen.no/no/aktuelt/elovkap7og8/id2784463/).

¹¹⁴ ‘Commitment to Enhance Resilience’, North Atlantic Council, Warsaw, 8 July 2016 https://www.nato.int/cps/en/natohq/official_texts_133180.htm.

¹¹⁵ Dobrowolski et al., ‘How Threat Actors Are Manipulating the British Information Environment’, 33.

¹¹⁶ UK Cabinet Office, Global Britain in a Competitive Age: The Integrated Review of Security, Defence, Development and Foreign Policy (London: UK Cabinet Office, 2021), 74–75; see also Intelligence and Security Committee of Parliament, Russia (London: Parliament, 2020), 33, 36.

¹¹⁷ Folketingstidende 2018–19, 1. samling, tillæg A, L95 som fremsat, 1 (Danish Parliamentary records).

¹¹⁸ Folketingstidende 2018–19, tillæg B, betænkning over L95.

¹¹⁹ Folketingstidende 2018–19, 1. samling, tillæg A, L95 som fremsat, 2.

¹²⁰ In the Norwegian legislative process, public consultation precedes the government’s finalisation of a bill before presentation to Parliament.

¹²¹ ‘Høringsnotat om endringer i straffeloven mv. – påvirkningsvirksomhet’, 12 May 2021 Ministry of Justice, 3, horingsnotat-om-endringer-i-straffeloven-mv.-pavirkningsvirksomhet.pdf (regjeringen.no).

¹²² NOU 2019:13, ch. 11.5.1; see also PST to Justis- og beredskapsdepartementet, 20.01.20, https://www.regjeringen.no/no/dokumenter/horing—nou-2019-13-nar-krisen-inntreffer/id2666172/; Ann Karin Larssen and Gjert Lage Dyndal, ‘Om handlingsprinsippet, “gråsonekriser” og korona’, in Larssen et al., Strategisk ledelse i krise og krig.

¹²³ MoD, Introducing the Integrated Operating Concept, 17.