Why SAPS Crime Intelligence is a hot mess

1 Feb 2021

Among the most important reasons Crime Intelligence is in such disarray is because it is, to all intents and purposes, unregulated. It is impossible not to conclude that the political class made a deliberate decision to keep regulation loose to make it more susceptible to manipulation.

The Crime Intelligence Division of the South African Police Services (SAPS) is a hot mess. The instability in the top leadership appears to be never-ending, and the instances of abuse rival those of the State Security Agency (SSA).

Crime Intelligence failed to detect the Guptas and their criminal activities. Like the SSA, it maintained the former president’s power base and acted as a Praetorian guard of sorts for him. Like the SSA, intelligence projects were established as vehicles for disinformation campaigns and even a propaganda war in favour of the former president and against his perceived enemies.

When former head of the Independent Police Investigations Directorate (IPID), Robert McBride, attempted to rein it in, it obstructed IPID investigations by over-classifying documents that were necessary for its investigations.

While it busied itself with meddling in the ruling party’s factional battles, organised crime spiralled. Small wonder that in his book, The President’s Keepers, journalist Jacques Pauw described Crime Intelligence as having “little honour, no moral compass, a complete lack of integrity”, as well as being rudderless and dysfunctional.

So why is Crime Intelligence such a hot mess? One of the most important reasons is because it is, to all intents and purposes, unregulated. It is impossible not to conclude that the political class made a deliberate decision to keep regulation loose to make it more susceptible to manipulation.

The division has a threadbare mandate. Its powers and functions are barely set out in law and controls on its spying powers are practically nonexistent. Legislators have been sleeping on the job. No one should be surprised that it has run amok.

The SAPS Act is silent on Crime Intelligence, leaving it to be regulated by general provisions. This is not appropriate as police spying uses extremely invasive methods that need to be defined clearly and controlled carefully. Internal directives do not count as adequate controls as they can be set aside.

The only law that sets out what the division is meant to do is the National Strategic Intelligence Act (NSIA), but that is only in passing and these provisions have not been revised for more than two decades.

According to the NSIA, the division gathers, correlates, evaluates, coordinates and uses crime intelligence in support of the police’s objectives as set out in the Constitution. It can also institute counterintelligence measures within the SAPS and supply crime intelligence relating to national strategic intelligence to the coordinating body for intelligence, the National Intelligence Coordinating Committee, or NICOC.

That’s the sum total of Crime Intelligence’s mandate in law. To say that it’s paper-thin is an understatement.

These acts are also out of date. The SAPS Act dates from 1995, although it has been amended along the way. The NSIA dates from 1994.

But, at least these were post-apartheid acts. Things are even worse when it comes to Crime Intelligence’s controversial secret services account, which it uses to fund covert operations. The act that governs this fund dates from apartheid’s darkest days.

The regime established a special account to fund the activities of the notorious Bureau for State Security in 1969. Then in 1978, the regime promulgated the Secret Services Act in 1978 with a much broader remit to fund its dirty tricks against the liberation movements. This pithy one-page act established a secret fund which more government departments could draw on.

Significantly, the act allowed for any department to apply for funding to fund secret operations. Abuse of these funds was a near certainty given that the only criterion they had to meet was that funded activities be in the national interest. No definition was provided for national interest, which gave departments of state the latitude to stretch the term to include whatever they saw fit to protect the regime.

Threats are too easy to talk up to justify access to the funds. A threat can be anything that a Crime Intelligence operative says it is, and secrecy makes it difficult to challenge them. Once they invent threats successfully, then it’s game on.

Under pressure to curb the massive human rights abuses during the 1980s, then-president FW De Klerk reformed the Secret Services Act in 1992. An amendment required the funds to be approved by an evaluation committee, which promised to bring greater transparency to the process.

However, there were serious flaws with this committee. It included only one member who was independent of the executive. The committee only had to consider whether activities to be funded were in the (still undefined) national interest.

To this day, these self-evidently inadequate provisions regulate the spending of Crime Intelligence’s secret services account, which draws from the fund. Internal controls are built on quicksand if the primary legislation is so weak, as they can be easily set aside. Furthermore, it is unclear whether the evaluation committee has even been functioning.

In other words, the secret services account rests on governance arrangements that were made for abuse. It is a shocking dereliction of duty that legislators have allowed these arrangements to persist in a democracy.

In the circumstances, all manner of tricks to fleece the public become possible. It is too easy for a Crime Intelligence operative to withdraw money as a cash advance to pay a source, for instance, claim that they were robbed in the street and then pocket the money.

Sources can be invented, as financial controllers may never get to see the actual source. A crooked police officer can simply present them with false identifiers, such as an ID number and receipts. They can pay off bonds on their houses by letting them to the division as safe houses at hugely inflated prices.

Sources, too, have a vested interest in maintaining threats long after their sell-by date, as they are paid for information and their livelihoods may depend on that income.

Threats are too easy to talk up to justify access to the funds. A threat can be anything that a Crime Intelligence operative says it is, and secrecy makes it difficult to challenge them. Once they invent threats successfully, then it’s game on.

There is an established intelligence cycle for strategic intelligence, or intelligence about the longer-term threats of special concern to strategic decision-makers. NICOC sits at the apex of this process as it is responsible for coordinating intelligence and developing a national intelligence estimate. These are used to forewarn policymakers about possible threats and assist the Cabinet to develop national intelligence priorities. Or at least they’re meant to.

In reality, though, the quality of NICOC products is questionable. Part of this has to do with the secrecy of this environment, as its products are a result of the collective efforts of the government spy agencies only. Estimates and priorities aren’t declassified and tested in public debate on a regular basis, turning the intelligence community into an echo chamber susceptible to confirmation bias.

Getting the right people into leadership positions is important, but it’s not the whole story. Clearly, brave people are attempting to clean up the division, but even the best people are likely to be undermined in such an ill-defined environment.

Also not helping matters is the fact that the mandates of the various intelligence agencies are not clearly delineated, especially those of Crime Intelligence and the SSA. Mandate and function creep have been the unfortunate result, fuelling competition rather than cooperation. Serious crime, for instance, receives the attention of both agencies, with little clarity on who does what.

Given these weaknesses, it is hardly surprising if unscrupulous individuals in Crime Intelligence overstate or manufacture security threats to justify access to the secret services account.

For instance, according to the 2016-2017 SAPS annual report, during the #Feesmustfall protests, SAPS said in relation to these protests and community protests more generally, that they investigated “…an increase in threats to the authority of the state”. This intention stretched far beyond suspected criminality and into legitimate protest action.

Consequently, on McBride’s account, members of the division used the protests as a pretext to purchase social media monitoring equipment at hugely inflated prices, although apparently the equipment was never installed. This was from a company that was being investigated for laundering money to buy votes for former president Jacob Zuma.

These abuses are likely to intensify as SAPS becomes more intelligence-led. SAPS is embracing this policing model — which is based on the assessment and management of risk — with vigour. Intelligence-led policing is proactive in that it attempts to predict and respond to crime before it arises, rather than waiting for a crime to happen.

To an extent, this policing model makes sense because it allows the police to deploy scarce resources more strategically to areas of highest risk. But, it can also lead to policing in the absence of a criminal predicate, which makes room for speculative policing of nebulous, ill-defined threats. The model can also lead (and in other countries has led) to racially-charged profiling of certain social groups that are perceived to be more likely to commit crime.

Crime Intelligence relies far too much on trust in its operations. It assumes that the people who oversee the division will be of high integrity, and that is its greatest weakness.

Getting the right people into leadership positions is important, but it’s not the whole story. Clearly, brave people are attempting to clean up the division, but even the best people are likely to be undermined in such an ill-defined environment.

The Police Act should recognise that there are particular dangers in Crime Intelligence that need to be mitigated, given the high levels of secrecy. In addition to its powers and functions, the act should spell out operating principles for the division.

For instance, the act should make it clear that the division should have a bias towards overt intelligence methods as a general principle, with covert methods being the exception. As things stand, the business model of intelligence leans it towards covert methods because they are more profitable than overt methods.

The division should rely on open source intelligence as the source of first resort, relying on publicly available sources. Covert collection, on the other hand, should be authorised only when gaps exist in open source intelligence and to complete the intelligence picture.

Once covert collection is reduced to a minimum, then the calls on the secret services account will automatically decline as it can only be used to fund covert activities.

As intelligence is often broader, more secretive and subjected to less scrutiny than conventional policing work, it is important that safeguards are included at the beginning. Record keeping is a case in point. The SAPS Act should guard against the dangers of improper collection and storage of information.

For instance, the bill should state that the division can target people or organisations only when a criminal predicate or nexus is present, and information that fails to meet this threshold should be expunged.

There needs to be a proper diagnostic assessment of what has gone wrong in Crime Intelligence, and what steps need to be taken to undo the damage. Otherwise, any efforts to address the hot mess are going to be cosmetic and easy to reverse. Recent attempts to reform public order policing have shown just how resistant the SAPS is to change.

The act should also contain principles around the processing of personal information and state that Crime Intelligence information should not include any information that has been obtained unlawfully.

People should not be targeted on the basis of their support for an unpopular cause, or for their membership of a particular racial group, or for their religious or political beliefs only. Political groups should be investigated only when there is reasonable suspicion of criminal activity.

Targeting should be authorised by an internal panel, which should prepare annual reports open to the public, covering requests for surveillance and the number of requests granted. As acting Director-General of the State Security Agency, Loyiso Jafta, said at the Zondo Commission last week, in relation to the SSA, compliance assurance of intelligence was generally after the fact.

In the SSA, not enough attention was paid to the management of projects before they were executed because the emphasis was on getting things done. The inference of Jafta’s comments was that projects were pushed through with little discussion or debate. Crime Intelligence appears to suffer from similar problems.

National intelligence estimates should be declassified and released publicly on an ongoing basis. National intelligence priorities should be debated in an open sitting of Parliament and closed sittings should be the exception rather than the rule. These measures should allow manufactured intelligence to be challenged, even if it is, to an extent, after the fact.

Counter-intelligence is particularly susceptible to abuse. The term is defined broadly in the NSIA as measures taken to impede and to neutralise the effectiveness of foreign or hostile intelligence operations and to counter any threat or potential threat to national security.

As a 2008 Commission of Inquiry into abuses in the SSA’s predecessor, the National Intelligence Agency, argued, words such as “neutralise”, “impede” and “counter” were vague and could potentially cover a range of illegitimate actions. Yet, more than a decade down the line, this problem still has not been addressed.

Clearly, the Secret Services Act needs to be repealed and replaced with something that spells out proper criteria for the accessing of secret services funding and puts in place measures to prevent abuse.

Ways also need to be found to reduce the amount of hard cash circulating in Crime Intelligence, as it is an Achilles’ Heel of the division and a recipe for continued corruption. From using e-wallets to blockchain and cryptocurrencies, methods could be devised of holding operatives accountable for monies received and journaling transactions, while retaining legitimate operational secrecy.

The Civilian Secretariat for the Police has put forward amendments to the Police Act. These have included a section on Crime Intelligence, which should be welcomed. This section of the draft bill covers the reporting lines of the division, its functions, and the process of obtaining security clearance. However, the section doesn’t even begin to touch on the issues raised here and relies too heavily on outdated and inadequate legislation.

There needs to be a proper diagnostic assessment of what has gone wrong in Crime Intelligence, and what steps need to be taken to undo the damage. Otherwise, any efforts to address the hot mess are going to be cosmetic and easy to reverse. Recent attempts to reform public order policing have shown just how resistant the SAPS is to change.

Happily, though, the Zondo Commission is putting huge amounts of information into the public domain about Crime Intelligence and state intelligence activities generally. This airing of the spies’ dirty linen provides the country with a rare opportunity to come up with much more decisive interventions to fix the mess. DM

Jane Duncan is a professor in the Department of Journalism, Film and Television at the University of Johannesburg. She is the author of Stopping the Spies: Constructing and Resisting the Surveillance State in South Africa (Wits University Press, 2018).

Categories:

South Africa

Tags:

DM is much better if you are signed in...

South Africa