Five Eyes: Blurring the Lines Between Intelligence and Policy

The public aura around the decades-old “Five Eyes” intelligence sharing partnership between Australia, the U.S., U.K., Canada, New Zealand has expanded rapidly since the name was first publicly acknowledged. In 2014, an Australian prime minister publicly referred to the “Five Eyes” for the first time. In 2016, the name first appeared in an Australian Defence White Paper. And in 2020, meetings between ministers from the five countries for, respectively, Defence, Foreign Affairs and Treasury, were first identified as “Five Eyes”. 

But labelling these meetings as “Five Eyes” is mistaken and possibly counterproductive. It unnecessarily limits their membership and risks blurring the critical distinction between intelligence and policy.

Intelligence sharing works when both sides can trust that information shared is raw and uninfluenced by policy preferences.

To be clear, members of the Five Eyes should keep coordinating policy. Ministers for Homeland Security and Immigration have been doing this through the “Five Country Ministerial” for several years. This cooperation can be especially beneficial when action depends on intelligence insights, for example when it comes to the attribution of malicious cyber behaviour. 

The Joint Australia-Canada-U.K.-U.S. statement on Hong Kong was hailed by many commentators as auguring a new era for Five Eyes (despite New Zealand’s absence). The collective approach added weight. But it would have had much more weight with the addition of a non-Anglo ex-colony.

Periodic proposals to expand the Five Eyes – by adding countries such France, Japan, Germany or South Korea – are missing the point and engendering disappointment. Five Eyes countries need to share more intelligence with trusted partners. And according to some reports, larger groups have been formed for particular purposes. But seeking to graft new members onto the existing Five Eyes agreement is simply unrealistic.

Characterising policy summits as Five Eyes gatherings also risks blurring the policy-intelligence distinction. The objectivity that is essential to good intelligence can be compromised when intelligence agencies get involved in policy debate. That’s why the directors of the CIA and U.S. National Intelligence don’t vote in U.S. cabinet meetings. Australia’s 2017 Independent Intelligence Review reaffirmed that “the need for intelligence assessments to be independent of policy-making … remains an indispensable requirement”.

That separation of intelligence and policy is as important internationally as it is domestically. Intelligence sharing works when both sides can trust that information shared is raw and uninfluenced by policy preferences. It works best when the agencies sharing are seen to be policy-neutral and able to engage frankly. 

While diplomatic meetings typically aim to find areas of agreement, intelligence meetings are just as often seeking disagreements. Disagreement produces better judgments. And separating analytic differences from policy disputes can better focus policy engagements. 

Forging a common approach to a problem as complex as 5G technology has been extremely difficult. It would be even harder if the governments involved suspected that the technical and intelligence assessments they shared were being shaped by policy goals.

As a new consensus on China takes shape in capitals, the ability of Five Eyes intelligence agencies to speak truth to power – and if necessary, to challenge that consensus – will become ever more important. To compete effectively with China, countries such Australia need an objective and nuanced view of the threat, as well as frank assessments of the impact of Western policy choices – especially when those assessments don’t gel with policy of the day.

Ben Scott has over twenty years’ experience in diplomacy, intelligence and international development. He joined the Lowy Institute from the Office of National Intelligence, where he was a Senior Middle East Analyst before representing ONI at the Australian Embassy in Washington DC from 2016-2020.

This article appeared originally at Lowy Institute's the interpreter.