The Usage of Data Creep and Data Doubles in Police Databases

In Canada, 420,000 individuals who have never been convicted of a crime have their names and information in police databases (Carlson, 2014). Some of these individuals are listed for mental health reasons, such as attempted suicide, violence, or mental instability, while others are in the database due to being charged with a crime, but never convicted (Carlson, 2014). These police databases are routinely searched when background checks are performed on individuals, such as when applying for work, or sometimes when crossing the border to the US (Carlson, 2014). Unfortunately, there are no rules and legislation as to what information and when the police can release this information, and this has made it difficult for some individuals to attain employment (Carlson, 2014). An investigation by the Toronto star shows the difficulties of a withdrawn charge on a subsequent job search – Diane was wrongly accused of assault by her ex-spouse and almost lost her job as a counsellor when a background check was performed and the withdrawn assault charge was revealed to her workplace (Carlson, 2014). Luckily, after months of appeals, the police force agreed to remove the charge off the Canadian Police Information Centre (CPIC) Database. However, most individuals are not successful at having this information removed from the database, and meanwhile, individuals convicted with a crime have a clear pardon legislation that allows them to have their conviction record expunged (Carlson, 2014). criminologists argue that it’s not beneficial to preserve non-conviction records, as the CPIC database’s objective is to keep information about individual’s criminal history, which one can have only if they were convicted of a criminal charge, but not if just charged with a crime. Furthermore, these may violate the constitutional rights of an individual to defend against false records and allegations against them (Carlson, 2014).

According to Haggerty and Ericson (2000), surveillance technologies used to be discrete and not linked with one another, however, we now have what the authors call ‘surveillant assemblage’. These surveillant assemblages is a combination of various surveillance techniques that break down the individual into discrete information flows, which are then reassembled into ‘data doubles’, by classification criteria that was established by the institutions that are capturing the data (Haggerty & Ericson, 2006). Each individual has a data double, which is a profile of their digital footmarks left in today’s age of electronic information. These data doubles are pure information, and they can be used by the government and corporations in order to differentiate between categories of individuals for marketing and institutional agenda purposes (Haggerty & Ericson, 2000). Furthermore, the government can and has profited from selling data found in official records, and corporations can exchange services for the personal information of those using the services.

Another worrisome feature of today’s surveillance techniques includes ‘function creep’, a process where laws and information gathering procedures that have been promoted and justified for a particular application are eventually being used for purposes that these were not initially designed for (Haggerty & Ericson, 2006). For example, black boxes in cars were originally used for airbag deployment during an accident, but have since been used in criminal investigations because these boxes record the speed of the car and whether a seatbelt was worn. Function creep is especially problematic because it is often used in an ad hoc way, where new monitoring possibilities can be found in a system that was not originally utilized for that particular manner, and most proponents of a new surveillance system cannot predict in advance how that system might be used in the future. Often, these function creep changes are done by bureaucratic reforms that are not made available to the public (Haggerty & Ericson, 2006).

When analyzing the Toronto Star investigation, we can see that although the police argues that CPIC notations are useful for officer and public safety when responding to a call, it is possible the police is keeping track of these records for the purposes of data doubles. For example, the police may want to categorize people who are mentally unstable or violent but have not been charged or convicted of a crime. Each individual’s data double could be used for the purposes of surveillance of protests, predicting public opinion regarding a particular public interest issue, knowing who to monitor when there is an expected protest or civil disobedience, etc. Furthermore, we can predict that this database may be used for different purposes in the future by the usage of data creeping. Data creeping occurs when a system designed for a particular function is used for more than it was originally mandated to do (Haggerty & Ericson, 2006). For example, like when the RCMP released information to Kinder Morgan regarding individuals protesting the pipeline. Other uses could include using information to refuse social or government services to individuals, for example, on the basis that violent or unstable individuals should not get access to firearms, social welfare, or other services and incentives. The police can also profit from selling this information to marketers – maybe mentally unstable individuals might be interested in pharmaceutical interventions to help with their instability, or violent individuals could be interested in martial arts courses. This information can further be used to substantiate a claim that someone has a history of being mentally unstable or violent when charged with a crime in the future. Overall, knowledge is power, and the police holds onto this knowledge as one day it may be used to have power over someone, utilizing their own discretion (or lack thereof) in making these disclosures about individuals.


Carlson, J. (2014, May 24). 420,000 in police database never convicted: Analysis. Toronto Star. Retrieved from:

Haggerty, K. D., & Ericson, R. V. (2000). The surveillant assemblage. British Journal of Sociology, 51(4), 605-622.

Haggerty, K. D., & Ericson, R. V. (2006). The new politics of surveillance and visibility. Toronto: University of Toronto Press.


1 Comment

Filed under Contributor Post

One response to “The Usage of Data Creep and Data Doubles in Police Databases

  1. This is an interesting and effective post. You do a good job of describing the practices in question and identifying the links between these practices and the concepts we are studying.

    One quibble: You note that “it is possible the police is [are] keeping track of these records for the purposes of data doubles.” I wonder: Is there any other explanation for this practice? It seems evident that these databases exist so that police have stable and long-term access to more complex data doubles than would be available through a system based on criminal records alone.