Institute for Legal Studies, Hungarian Academy of Sciences, Telephone number: (36-1) 201 9717-886, E-mail: pap@jog.mta.hu
The question of ethnic and racial discrimination has became an extremely important point in legal discussions relating to the proliferation of new law enforcement authorizations employed in the "war against terrorism". The old desire within law enforcement to minimize human decision-making and to deploy automated investigating or screening processes brought a revitalization of formerly discredited measures such as profiling. Also, the fast development of informational technology enables the interconnection of various commercial and law enforcement databases, and also the melding of formerly independent immigration, criminal etc. data-sources. The IT sector in the security business brings lucrative opportunities, thus the private sector is no longer only an innocent victim of excessive government legislation, on, say, reporting requirements, but also a keen partner.
Due to spatial constraints this paper does not include a legal analysis of ethno-racial profiling, nor does it enumerate possible legislative remedies to its problems, its scope is limited to clarifying some of the crucial concepts and definitions.
There is not one universally accepted and utilized definition for profiling. Depending on the actual context, international organisations, academics and politicians use vastly differing criteria. Even in the narrowly defined context of racial profiling by law enforcement agencies, considerably different definitions have been used. For example, in 2002, the EU's Working Party on Terrorism
- 277/278 -
drew up recommendations for member states on the use of "terrorist profiling," and defined it as using "a set of physical, psychological, or behavioural variables, which have been identified as typical of persons involved in terrorist activities and which may have some predictive value in that respect."[2] In their Opinion on ethnic profiling,[3] the E.U. Network of Independent Experts on Fundamental Rights defines ethnic profiling as the practice of classifying individuals according to their 'race' or ethnic origin, their religion or their national origin, on a systematic basis, whether by automatic means or not, and of treating these individuals on the basis of such a classification. In its letter of 7 July 2006, the European Commission gave the following definition: racial or ethnic profiling encompasses any behaviour or discriminatory practices by law enforcement officials and other relevant public actors, against individuals on the basis of their race, ethnicity, religion or national origin, as opposed to their individual behaviour or whether they match a particular 'suspect' description'. According to James Goldston, the executive director of the Open Society Justice Initiative, by ethnic profiling we mean the use of racial, ethnic or religious stereotypes in making law enforcement decisions to arrest, stop and search, check identification documents, mine databases, gather intelligence and other techniques'[4]. In R. v. Richards[5] a case presented to the Ontario Court of Appeals, the African Canadian Legal Clinic defined racial profiling as criminal profiling based on race. Racial or colour profiling refers to that phenomenon whereby certain criminal activity is attributed to an identified group in society on the basis of race or colour resulting in the targeting of individual members of that group. ECRI's General Policy Recommendation No. 11 on Combating Racism and Racial Discrimination in Policing[6] uses the following definition: "The use by the police, with no objective and reasonable justification, of grounds such as race, colour, language, religion, nationality or national or ethnic origin in control, surveillance or investigation activities".
Profiling in the abstract sense refers to identifying information, making predictions and, finally, inference.[7] In any abstract profiling operation, three stages
- 278/279 -
may be identified: The first stage is "observation", often referred to as data warehousing, where personal or anonymous data are collated. If the data refer to an identifiable or identified individual, they will generally be anonymised during this stage. The collected data may be of internal or external origin. For example, a bank might draw up an anonymous list of its customers who are bad payers, together with their characteristics, or a marketing firm might acquire a list of the major supermarket chains' "shopping baskets" without the shoppers being identified. This first stage is followed by a second set of operations, usually referred to as data mining,[8] which is carried out with statistical methods the purpose of which is to establish, with a certain margin of error, correlations between certain observable variables. For instance, a bank might establish a statistical link between a long stay abroad and one or more missed loan repayments. The concrete outcome of this stage is a mechanism whereby individuals are categorised on the basis of some of their observable characteristics in order to infer, with a certain margin of error, others that are not observable. The third and last stage, known as "inference", consists in applying the mechanism described above in order to be able to infer, on the basis of data relating to an identified or identifiable person, new data which are in fact those of the category to which he or she belongs. Very often, only this last operation is referred to as "profiling", however, it is essential, to see this final stage as part of a process.[9]
The Consultative Committee of the Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data defined profiling as a computerised method involving data mining from data warehouses, which makes it possible, or should make it possible, to place individuals, with a certain degree of probability, and hence with a certain induced error rate, in a particular category in order to take individual decisions relating to them. This concept of profiling differs from criminal profiling, where the aim is to get inside and understand the criminal's mind, but is similar to behavioural analysis since the aim is not to understand the motives which lead or might lead an individual
- 279/280 -
to adopt a given behaviour, but to establish a strong mathematical correlation between certain characteristics that the individual shares with other "similar" individuals and a given behaviour which one wishes to predict or influence. As this approach does not depend on human intelligence, but on statistical analysis of masses of figures relating to observations converted into digital form, it can be practiced by means of a computer with minimum human intervention.[10]
Lee Bygrave of the University of Oslo proposed the following definition: profiling is the process of inferring a set of characteristics (typically behavioural) about an individual person or collective entity and then treating that person/entity (or other persons/entities) in the light of these characteristics. As such, the profiling process has two main components: (i) profile generation -the process of inferring a profile; and (ii) profile application - the process of treating persons/entities in light of this profile."[11]
The word "profile" (profil in French) was originally used in the artistic field. It denoted the outlines and features of a face seen from one side or, more broadly, the portrayal of an object seen from one side only. By extension, this term eventually took on the figurative sense of "all the characteristic features of a thing, a situation or a category of people". Where people are concerned, this term thus refers to all the characteristic features exhibited by a person or category of persons.[12] A profile is thus merely an image of a person based on different features: in the artist's profile, features are sketched, in profiling, data are correlated and in neither case can the profile be equated with the person him or herself. The core of the profiling-problem will lie exactly in the fact that it tends to reduce the person to the profile generated by automated processes which are liable to be used as a basis for decision-making. These processes carry several risk factors. The European Commission for example has brought attention to the fact that the registered data images of persons (their "data shadows") might eventually usurp the constitutive authority of the physical self despite their relatively attenuated and often misleading nature, along the fact that rampant automation of decision-making processes engenders a lack of involvement and abdication of responsibility on the part of "human " decisionmakers.[13] It is for this reason that Article 15 of the data protection Directive of the EU[14] deals explicitly with "automatedindividual decisions."[15]
- 280/281 -
Historically, the term "profiling" in law enforcement first came to prominence in connection with the training of crime profilers in the USA. In theory, these people are supposed to be capable of determining a criminal's personality type by analysing traces left at the scene of the crime. Recent developments in information technology, however, today make profiling activities increasingly easy and sophisticated, thus the possibilities offered by profiling are numerous and cover different areas of application. For example, in the USA, ATS (Automated Targeting System) has been developed in order to evaluate the probability of a given individual being a terrorist. Also, data mining is an extremely valuable tool in the area of marketing and customer management. It is one means of moving from mass marketing to genuinely personalised marketing. Likewise, profiling is widely used in the field of risk management, when determining the characteristics of high-risk customers. Such aims may include the adjustment of insurance premiums; prevention of arrears; aid to payment decisions where current account overdrafts exceed the authorised limit in the banking sector; use of a risk "score"[16] in order to offer individual customers the most appropriate loan or refuse a loan depending on the probability of honouring repayment deadlines and the terms of the contract, etc. Cable digital TV provides programme distributors with precise information regarding channel selection and channel hopping by viewers who receive television channels via the telephone cable by means of DSL technology. They can thus create and keep a perfectly accurate viewing profile for each user. It therefore becomes technically possible to tailor advertisements to the user's profile. Also, a simi-
- 281/282 -
lar methodology is used by Google's on-line advertising system, where user's click stream is monitored. In the age of strategic marketing, profiling and data mining is used in creating packages and special offers; designing new products and customer loyalty policy.[17]
Profiling based on data mining is, thus, the process by which large data bases of personal information are subjected to computerized searches using a set of specific criteria. In law enforcement, these criteria are generally based on the common characteristics of persons responsible for past offences and often include ethnicity, national origin and religion. The data is used to narrow down a set of targets for further investigation. Data mining has been explored with a specific interest in its potential as a tool to identify terror "sleeper cells" as was the case with Germany's data mining effort after the discovery of the Hamburg cell in 2001. Called the Rasterfahndung, this massive and costly data mining exercise failed to turn up a single terrorist. The experience has been similar in the US, where immigration data was used to identify tens of thousands of persons for scrutiny as potential terrorists, but in the end not one charge on terrorism offences resulted. It is noteworthy that Germany's constitutional court has ruled that, in the absence of a concrete danger, this technique constitutes an unwarranted intrusion on personal privacy.[18]
Let us collect some of the indispensable logical elements that pertain to the concept of profiling. Thus, profiling (i) can be applied in a number of contexts, that can vary from the commercial sector to the field of law enforcement; (ii) it is a mechanism where the task is to narrow down the circle of potential individuals that may fall within the scope of activities of a particular agent within the given field: it may involve identifying a group of customers or potential perpetrators; (iii) profiling will always include certain characteristics upon which the process relies; and (iv) there will always be a scheme of reasoning according to which these characteristics and the way of their employments is established.
As mentioned above, this paper cannot include a thorough legal analysis of profiling, however, it needs to be stated that each of the elements will evoke and include a set of legal rules or tests to apply, where a specific type of profiling, say profiling on the basis of ethnicity applied in border control by the
- 282/283 -
police, will involve a peculiar patchwork of legal norms. For example, if race or ethnicity is used within the process, stringent data protection regulations for the handling of sensitive data are applicable. Also, if the field of activity falls, say, within the scope of the Race Directive[19] and the screening mechanism has the consequences of disparate treatment (which may include denying service or over-policing), the legality of profiling under the directive will depend on whether the rationale behind is rational and efficient (appropriate and necessary).
Bearing these in mind, let us now focus our attention on profiling on the basis of ethnicity and race as it comes up in the context of various law enforcement activities, especially in counter-terrorist measures and border control.
Throughout this paper, the terms "police", "policing" and "law enforcement agents or agencies" will be used synonymously - including both (i) public administrative and (ii) criminal investigative functions. In this regard, the terminology set forth by the explanatory memorandum of ECRI's General Policy Recommendation N° 11 on Combating Racism and Racial Discrimination in Policing will be followed, where the term "police" refers to those exercising (or having by law) the power to use force in order to maintain law and order in society, normally including prevention and detection of crime - regardless of how such police are organised; whether centralised or locally oriented, whether structured in a civilian or military manner, whether labelled as services or forces, or whether they are accountable to the state, to international, regional or local authorities or to a wider public, including secret security and intelligence services and border control officials as well as private companies exercising police powers as defined above.[20] The term refers to control, surveillance or investigation activities. Acts that fall in this definition include: stops and searches; identity checks; vehicle inspections; personal searches; searches of homes and other premises; mass identity checks and searches; raids; surveillance (including wire-tapping); data mining/data trawling.[21]
According to Rebekah Delsol,[22] racial profiling refers to the use by the police of generalisations based on race, ethnicity, religion or national origin, rather than individual behaviour, specific suspect descriptions or accumulated intelligence, as the basis for suspicion in directing discretionary law enforcement actions such as stops, identity checks, questioning, or searches among other tactics. Racial or ethnic profiling is, thus, distinct from 'criminal profiling"
- 283/284 -
which relies on forms of statistical categorisation of groups of people according to identifiable characteristics, believed to correlate with certain behaviours. Specific definitions of racial or ethnic profiling vary along a continuum ranging from the use of race alone as the reason for the stop to those using race along with other factors as the reason for the stop.[23]
Using a narrow definition, racial profiling occurs when a police officer stops, questions, arrests and/or searches someone solely on the basis of a person's race or ethnicity. A broader definition acknowledges that race may be used as one of several factors involved in an officer's decision to stop someone. A stop is likely to be made on the confluence of several factors such as race or ethnicity along with age, dress (hooded sweatshirts, baggy trousers, perceived gang dress etc.), time of the day, geography (looking 'out of place' in a neighbourhood or being in a designated 'high-crime area'). This definition reflects the fact that racial profiling may be caused by the purposefully racist behaviour of individual officers, or the cumulative effect of the unconscious use of racist stereotypes, but may also result from institutional factors, such as the use of enforcement techniques and deployment patterns, which impact on ethnic groups unequally.[24]
Profiling can take place in other stops or contacts with the public by any type of law enforcement officer or other authorities such as traffic stops in cities as well as highways, stopping and questioning of pedestrians in public places in urban areas, sweeps of trains and buses, immigration status checks by immigration officials, and airport security and customs checks or searches. Patterns of profiling can also be seen in discriminatory treatment after a stop has taken place, such as black motorists being given traffic citations while white motorists are let off with a warning, or Latino/a youth, but not white youth, being cited for noise violations, mass controls in public places, stop and search and identity checks, data mining and raids on places of worship, businesses and organisations.[25]
Thus, ethnic or racial profiling, that is profiling that includes race and ethnicity as one of the characteristics involved in the process, is a practice that relies on the tenet that ethnicity in itself signals a certain type of criminal involvement, terrorist plotting or illegal border crossing as more likely, and this assumption serves as a sufficient and therefore legitimate basis for law enforcement (police, secret service etc.) suspicion. The peculiarity of profiling lies in the fact that it is not based on illegal behaviour, but is centred around the idea to collect
- 284/285 -
legal behavioural patterns or character-traits that may signal criminal behaviour - it is therefore based on an assumed correlation between criminality and the specified characteristics or behavioural patterns, a deduction based on retrospectively judged effectiveness, which is always assumed, rather than checked and confirmed.
Law enforcement profiling, which mostly takes the form of stop and search, was first developed in the U.S. for detecting drug couriers, and was later implemented in traffic control, and more recently in anti-terror procedures. Originally, the procedure of profiling was aimed at creating a description profile for suspects, in order to help the authorities in filtering out potential perpetrators based on certain sets of (legal) behaviour and circumstances. In the case of drug couriers, such a characterization might include short stop-overs between significant drug sources and the distribution location, cash paid for the airline ticket, and, based on criminal statistics, also ethnicity, sex and age. The inclusion of ethnicity in the profile was reasoned by the fact that gangs that play key roles in organized crime tend to be almost exclusively ethnically homogenous.
Thus, at the heart of ethno-racial profiling is the idea that the race or ethnicity of the perpetrator serves as a useful tool for the detection of criminality. Therefore, stops are not or not solely induced by suspicious or illegal behaviour, or by a piece of information that would concern the defendant specifically. Instead, a prediction provides grounds for police action: based on the high rate of criminality within the ethnic group or its dominant (exclusive) involvement in committing certain acts of terrorism, it seems like a rational assumption to stop someone on ethnic grounds. Measures are therefore applied not so much or solely on the basis of the (suspicious) behaviour of the individual, but based on an aggregate reasoning. This does not necessarily involve a discriminatory intent, since the goal is to make an efficient allocation of the limited amount of the available police and security resources. After all, one might argue that minorities (blacks, Roma, etc.) constitute a very large proportion of the prison population (vastly exceeding their ration in the overall population) and many, or most of the terrorists are presumably Muslim fundamentalists (mostly from Arab countries). Accordingly, appropriate restriction of the circle of suspects seems easily justifiable. For example, the EU Terrorism situation and trend report 2008 identifies five types of terrorist groups and movements, two of which: "Islamist terrorism" and "Ethno-nationalist and separatist terrorist groups" include classifications that deem such interconnections useful.
- 285/286 -
From the above stated it follows, that when considering the legality of certain specific profiling techniques, it needs to be scrutinized on the outset whether the inclusion of ethno-racial characteristics, or sensitive data handling would not deem the profiling mechanism facially illegal. Should a profiling scheme pass this first line of scrutiny, the legal qualification of the given measures will depend on its proportionality, which will need to include the scrutiny of the efficacy of the given mechanism. In other words, it is a necessary but not sufficient condition to be legal for a profiling mechanism that singles out individuals on the basis of their ethnicity or race to be efficient from the practical point of view: that is, it needs to be proved that criminal involvement is higher within the specific ethno-racial group and the hit rates (successful application of the profiling mechanism) need to be high. The objective measurement of these factors are logical and necessary conditions for the balancing tests that are applied by international judicial organs and constitutional courts, as these tests include the weighting of how intrusive certain means are in comparison to the ends-provided of course, that the ends are legitimate.
In addition to the handling of sensitive data and the efficacy of profiling, a third important point of scrutiny concerns the reasons and justifications (standards of suspicion or probable cause) based on which law enforcement action or sanctions, such as stop and search may be initiated. It is important to note that since law enforcement work, investigation, crime prevention is not built solely on actual illegal behaviour and suspicion, a core concept in this discussion is not always easy to codify.
The question can be rephrased as under what conditions might the police (or other law enforcement organs) initiate action? The standards will change according to how concretely specified the perpetrator is, what the degree of suspicion is, and in what capacity the law enforcement agent is acting: is it a random, voluntary encounter; consensual questioning that does not involve coercion, where in theory the citizen may disregard the question; stopping and questioning during an investigation; vehicle control; border control, etc. Along the way, we may well ask: is it justifiable to institute a roadblock obstructing everyone's way (and not just that of a specific ethnic group)? What kind of suspicion (if any) is necessary for such a measure? Are random checks acceptable?[26] In order to easily pinpoint to unwarranted ethnic motivation, it would
- 286/287 -
require that the court (or legislator) state that police action can be initiated exclusively on the basis of individual behaviour or suspect description. But courts throughout the world will uphold a number of types of general controls: roadblocks, raids, alcohol tests,[27] Courts have not been entirely consistent either. In the 1975 Brignoni-Ponce case, for example, the US Supreme Court did not accept the authorities' defence that in the vicinity of the U.S.-Mexican border, on a stretch of road that usually teems with illegal immigrants, someone could be stopped solely on the grounds that she looks Mexican.[28] Stark contrast is provided by a 2001 decision of the Spanish constitutional court, which stated that skin colour and foreign appearance could serve as a profound cause for determining whom the police might stop. A year after the Brignoni-decision, in US vs. Martinez-Fuerte,[29] based on a similar pretext (the large number of Mexicans), the American Supreme Court, too, found it acceptable to order a roadblock.
The American case law and jurisprudence provides good illustration for the case. As spelled out by a set of detailed court decisions, the law distinguishes four ways in which police action may rely upon ethnicity or race, applying different constitutional measures for each of them.
(a) The first, unproblematic scenario is when the victim or witness to a crime provides a description of a specific suspect which includes ethno-racial characteristics. In these situations, American courts have invariably found that it was legal to use such information - in search warrants, for example.
(b) A second, somewhat different scenario is one in which the description provided by the victim or witness contains very little concrete detail about the suspect beyond her race or ethnicity. In such cases, on several occasions, the courts' stance was that race and ethnicity can be operative in negative descriptions only; for example, if the informant identified the perpetrator as black, then that information can serve as basis for the police not to stop whites and Asians, but it would border on discrimination for them to start stopping blacks without any further reason for doing so beside their skin colour. This way, if we know only that the perpetrator is black, then the law enforcement syllogism
- 287/288 -
means the following: the perpetrator is black and the suspect is white; from these it follows that the perpetrator cannot be identical with the suspect; but it does not follow from the pair of claims that the perpetrator is black and so is the suspect (or the pedestrian or driver), that then the suspect is the perpetrator.[30]
(c) The third case is racial profiling, which relies on the tenet that ethnicity in itself makes criminal involvement more likely, and this assumption is not based on any specific or general information about a given, concrete individual.
(d) Finally, the fourth case, which features prominently in the war against terror, involves preventive measures that rely on official, written directives about certain racial, ethnic, national or citizenship-based considerations. In these cases, the application of ethno-racial profiles is no longer left to the discretion of the police, border guards and airport security personnel. Instead, ethnic profiling becomes an officially formulated prescription. For example, recommendation of the President of the Hungarian Financial Supervisory Authority No. 1/2004 on the prevention and impeding of terrorist financing and money laundering provides a vivid example for singling out Arab and Muslim countries by the very formulation of its due diligence and reporting requirements: "Transactions should primarily be examined in terms of whether they are related to individuals, countries or organisations contained in the specific international lists. ... Raised attention needs to be paid to electronically sent and received amounts, which are unusual for certain reasons, including especially the size of the amount, the beneficiary target country, the country of the customer placing the order, currency or the method of sending or receipt. .. If an activity does not fit in the registered and reported activities, if the origin of received funds is unclear, if an amount increases from unusual sources, the target country or addressee raises a suspicion, the financial service provider needs to analyse and evaluate them with special care, and the transaction should be reported to the authority even if the smallest suspicion arises."[31]
The question of how to, and how not to take ethno-racial classifications into consideration in policing is only one of the intricate issues involved in the legal assessment of racial profiling. Another question relates to general police powers and the issue of when police is authorised to perform coercive measures, including those to stop and search. In Delaware v. Prouse,[32] the Supreme Court has acknowledged the dangers of unconstrained police discretion. In Prouse, a police officer stopped a motorist, not because of any traffic violation or suspi-
- 288/289 -
cious activity, but simply to spot-check the driver's license and registration. The officer did not conduct this random spot check pursuant to any police department or state standards, guidelines, or procedures. Finding that the motor vehicle stop constituted a seizure, the Court proceeded to balance the interests of the State of Delaware against the intrusions on the motorist's Fourth Amendment (the clause in the constitution that guards against unreasonable stops and seizures) protections. The State asserted that a practice of discretionary spot checks promoted its interest in ensuring the safety of its roads. The Court found that the spot checks that were not suspicion-based only marginally contributed to roadway safety. Highlighting the limits of acceptable police discretion, the Court has contrasted the problematic discretionary spot checks in Prouse with the more neutral, though still not suspicion-based, checkpoints in cases like Michigan Department of State Police v. Sitz.[33] Here, motorists challenged sobriety checkpoints in which the police stopped all drivers, without individualized reasonable suspicion or probable cause, and briefly examined them for signs of driving under the influence of alcohol. A checkpoint advisory committee had created guide-lines regarding the selection and administration of the checkpoints. Unlike in Prouse, the officers who staffed the checkpoints in Sitz did not choose whom to stop. Instead, the checkpoints were selected according to predetermined guidelines, and the officers stopped every approaching automobile. Thus, there existed little opportunity for arbitrary and unconstrained exercise of discretion by law enforcement. Applying a balancing test, the Court ultimately upheld the state's use of the checkpoints because the state's interest in preventing drunk driving, and the extent to which the checkpoints could reasonably be said to advance that interest, outweighed the intrusions on the motorists. In Illinois v. Lidster,[34] the court upheld highway checkpoint to seek information on fatal hit-and-run accident committed one week before.
Taking into consideration the complexity of racial profiling, in order to tackle to problem efficiently, it is also important to scrutinize the question of individual suspect descriptions. The requirement of probable cause to initiate stop and search procedures only tackles one element of a potentially arbitrary policing practice. Since overt and covert racial prejudices are widespread in society, before including an ethno-racial element in a suspect description, police should be put under an obligation to thoroughly investigate the reliability of such descriptions. As Priyamvada Sinha points out, a useful suspect description is one that is complete, accurate, and reliable. However, several factors can impede the creation of a useful suspect description, including the circumstances of the
- 289/290 -
crime, the specific characteristics and capabilities of the witness and, also, the efficacy of the police interviewer. There are a number of relevant factors to a victim's or a witness's ability to perceive and to remember: poor or nonexistent lighting, the speed of the event, violence that renders the incident difficult to perceive and to remember by the witness, but also stress level, age, and bias. A witness may also not recognize or predict her own inability to perceive or to remember and may not be able to convey effectively to an interviewer her perceptions due to fear, anxiety, or limitations on verbal skills.
A landmark study regarding perception and memory demonstrates the strong influence of race on witnesses. White subjects were instructed to look at pictures of an African American man and a white man who appeared to be arguing, in which the white man was holding a razor. The subject was asked to describe the scene to a second person, who then related that description to another person, and so on. Researchers discovered that in the successive descriptions, the razor often migrated from the hand of the white man to that of the African American man. They concluded that this result was due to cultural biases or stereotypes of African Americans as being more prone to criminality and violence. Also, victims tend to mischaracterize or to overestimate the number of incidents with African Americans in crime victimization surveys. Records in Portland, Oregon revealed only a thirty-four percent agreement rate between the racial characteristics of suspects recorded in police data and those reported to researchers in subsequent surveys about crime. The victims also estimated a greater number of incidents involving African American suspects than did the police. The researcher suggested that such inaccuracies in crime victimization surveys might be examples of victims "project[ing] racial bias or prejudice into their perception of who committed the crime."[35]
American jurisprudence[36] provides a number of examples where courts found stops illegitimate when based on overbroad suspect descriptions. The issue was highlighted by the infamous 1969 Davis v. Mississippi case,[37] where a woman who had been raped described her assailant to the police as a male "Negro youth." The police subsequently questioned or fingerprinted between sixty-five and seventy-five young African American males at the police station, in school, and on the streets. There were numerous other cases: while trying to capture a serial rapist in a primarily white area of Ann Arbor, Michigan, the police identified over 700 African American men as suspects, on the basis of a description of a "six-foot black man." They took DNA samples from 160 African Ameri-
- 290/291 -
can men. The attorney who represented donors seeking the return of their DNA samples commented that the " [police] attempted to [test] every African American man in the city who vaguely met the description." In Philadelphia, eight women who were raped described their assailant as a "slender black male." The Philadelphia Police Department used this sparse description to stop and question many African American males ranging in age from twenty to forty, weighing 120 to 160 pounds, and from five feet four inches to five feet ten inches. A woman who was raped at a bus stop in Baltimore, Maryland reported to the police that the perpetrator was an African American man, in his early thirties, about five feet ten inches, and weighing 180 pounds. The police district commander issued a memorandum to officers that simply instructed: "Every black male around this Bus Stop is to be stopped until the subject is apprehended." In Brown v. United States[38], the District of Columbia Circuit Court held that an anonymous tip regarding a narcotics seller who was a "black male, approximately 5'6" in height, wearing a white shirt with dark writing on the front and blue jeans" was inadequate as a suspect description, absent any other indicia of distinctiveness. Citing another case,[39] the court noted that "d]escriptions applicable to large numbers of people will not support a finding of probable cause".
In time, American courts developed more and more articulated guidelines to annul such practices. In People v. Robinson,[40] a New York state court overturned a conviction in a case where the police had stopped, searched, and arrested suspects on the basis of descriptions of young African American men with medium builds and slightly varying heights. The court likewise noted the insufficiency of such general descriptions, which could have matched many people in the neighbourhood. In United States v. Jones,[41] a circuit court held that an anonymous tip that several Black males were drinking and causing a disturbance at specific intersection "was so barren of detail about the alleged culprits' physical descriptions" that, coupled with a lack of corroboration, it could not establish reasonable suspicion for stop. In Commonwealth v. Creek[42] the court held that holding physical description of "black male with a black [three-quarter] length goose" jacket was insufficient suspect description where the defendant was arrested in predominantly African American neighbourhood on a cold fall night. In Faulk v. State[43] the court held that description of an armed robber as "young black male wearing a multicoloured shirt" was too general to yield probable cause and noting that police officer who stopped ap-
- 291/292 -
pellant "had only one fact to connect the appellant to the armed robbery-that he was a young black male". In Brown v. State[44] it was found that the description in an armed robbery including race and approximate height and weight, coupled with absence of any incriminating conduct or circumstances surrounding appellants, failed to yield probable cause for arrest.[45]
Another important feature of profiling relates to the proliferation of biometrics[46] industry, which curtails the involvement of human participation and control in decision-making, along recent trends of outsourcing traditional state authorities to the private sector. It appears to be the case that industrial business interests go hand in hand with government policies to strengthen control mechanisms - often lacking proper efficacy and risk assessment and evading traditional democratic control-mechanisms. As a briefing paper[47] commissioned by the Directorate-General Internal Policies points out, governments across the EU increasingly advocate biometric enhanced identity documents. However, the mesmerising possibilities opened by inter-operability,[48] information and data exchange by civil and law enforcement agencies so far escape effective parliamentary scrutiny and control, and citizens are unlikely to have a choice as to whether or not they wish to supply their biometric details to government authorities or public policy agencies.[49]
- 292/293 -
Biometric data to verify the authenticity of an individual's claim to be the person named on, say, a travel or any other document are not new. What is new is (a) the speed with which biometric data can be collected, stored and exchanged by inter-operable data systems and data miners; (b) their conflation with information from which culture, behaviour and/or profiles may be inferred; and (c) apparent willingness by EU policy-makers to allow biometric data transfer on a one way basis to a third state.[50] However, Juliet Lodge argues that the use of biometrics raises similar concerns to those relating to the storage of DNA data. The trend is for biometrics (iris recognition, finger prints, thumb-prints, handprints, voice-prints, signatures, 3D facial images, face or vein heat pattern imaging) to be used as a password or key to authenticate an individual's identity and with it her entitlement to access public and commercial services, exercise political rights (for example, though e-voting), and engage in processes connected to living in modern societies.
At the same time, as the briefing paper points out, in all EU member states, the trend is for government agencies and corporate interests using biometricised documents to downplay the technical disadvantages (false positive and false negative matches of all biometric tools, costs, divergent or competing standards, problems with proper installation of systems, network security, durability, quality of images and documents, dangers of inter-operability, susceptibility to capture by ambient technologies, fraud, ID theft, ambiguities regarding the ownership of legal liability for systems, etc) and advocate their widespread roll-out as a way to extend the application of information and communication technologies (ICT's) to government. Especially in relation to certain problematic policy areas: to monitor migration, combat identity theft and fraud, cut costs, produce efficiency gains for administration, and enhance convenient access to government services for citizens. Juliet Lodge claims that generally, states play relatively little attention to the issues of digi-inclusion and digi-exclusion, mandatory versus voluntary enrolment in biometricised identity documents, the ability of users to pay for the initial and subsequent cards, and safeguards against malevolent insider fraud. The ideal of exchanging personal data (including biometric data) subject to the principle of informed consent seems understood but not necessarily well-articulated in legislation or implementation. Governments and commercial interests rather than citizens seem to be the drivers behind and beneficiaries of e-government delivery. This is partly because communication over the purposes of biometric ID cards, and accountability for their deployment is hazy, weak or non-existent, costs are poorly communicated in many states, with implausible claims in some as to how they will combat crime, illegal immigration and terrorism (the culture of fear) and
- 293/294 -
enable simple e-government in civil and critical infrastructure applications. The result is that legislation lags very far behind ICTprogress.[51]
Within the trend for biometric processing to become omnipresent lies a sever danger: for biometric data to be useful, data banks need to be, and technically speaking, can easily become inter-operable-both within member states' national administrations and across them. The needs of Frontex, Europol, Eurodac, VIS, SIS II, police, customs, migration and judicial cooperation on law enforcement, crime and immigration matters, as well as in relation to crossborder civil law issues, ranging from the procurement of goods and services to family law, driving licences and insurance all could be handled by one huge database. This would, of course generate a considerable degree of resistance, if communicated openly, however, it appears to be the case that soon a biometricised ID card (biometricised travel documents, eIDs, e-purse, and transaction cards) can easily become the only way of crossing quasi-borders in non-territorial spaces of e-governance, as well as at the virtual and drifting borders of the EU. These cards can be used for purposes other than the very limited ones of verifying a person's identity, e.g. to check that one has paid road tax, has insurance and so on. Inter-operable data bases are very useful to both private and public sectors.
Even though the voluntary subscription of citizens to use digi-IDs, verichips or smartcards seems to imply a greater degree of trust towards private authorities than to public ones, it may be misleading, since with the proliferation of security obligations burdening private entities, the line between public and private services blurs. It needs to be added that driven by business interests in the biometrics industry, such obligations to participate in state control might not even be so burdensome for private entities making profit from the deal. Juliet Lodge emphasizes that out-sourcing data collation, processing and checking on civil and commercial matters is risky and eludes effective, territorial scrutiny by existing political or regulatory authorities. Outsourcing aggravates the decline in parliaments' ability to hold governments accountable.[52]
It has been shown that proportionality and impact-assessment will be a central point in the legality of profiling. As Rebekah Delsol points out, Europe's rapidly-expanding immigration and border control data bases offer a new information resource for law enforcement and counter-terrorism as well as immigration control, and it may well be tempting to seek to exploit them through the
- 294/295 -
use of profiling. However, as border control systems rapidly create large volumes of data and law enforcement access to this data is permitted for counter-terrorism investigations as well as immigration control, it is incumbent on the European Union and its member states to rectify what has become an unintelligible system of standards with complex and overlapping data protection standards and complete gaps in anti-discrimination norms.[53]
Ahmed, Patricia - Feliciano, Cynthia - Emigh - Rebecca Jean: Ethnic Classification in Eastern Europe, http://www.sscnet.ucla.edu/soc/groups/ccsa/ahmed.PDF
Anderson, David A.: The Aggregate Burden of Crime, 42 J.L. & Econ. (1999)
Banks, Richard: Racial profiling and antiterrorism efforts, Cornell Law Review, July 2004,
Harris, David A.: Racial profiling revisited: "Just common sense" in the fight against terror?, Criminal Justice, Summer 2002
Banks, Richard: Beyond racial Profiling: Race, Policing, and the Drug War, Stanford Law Review, December, 2003
Bowett, Derek: The Impact of Security Council Decisions on Dispute Settlement Procedures, European Journal of International Law (1996)
Bowring, Bill - Korff, Douwe: Terrorist Designation with Regard to European and International Law: The Case of the PMOI, International Conference of Jurists, Paris, 10 November 2004, http://www.cageprisoners.com/articles.php?id=7822
Braber, Liam: Korematsu's Ghost: A Post-September 11th Analysis of Race and National Security, Villanova Law Review, 2002.
Brandl, Steven: Back to the future: The implications of September 11, 2001 on law enforcement practice and policy, or Ohio State Journal of Criminal Law, Fall, 2003.
Buerger, Michael - Farrell, Amy: The evidence of racial profiling: interpreting documented and unofficial sources, Police Quarterly, Vol. 5., No 3, September, 2002
- 295/296 -
Bygrave, L. A. « Minding the machine: Article 15 of the EC Data Protection Directive and Automated Profiling », Computer Law & Security Report, 2001, Vol. 17.
Cameron, Iain: (2003) European Anti-Terrorist Blacklisting, in Human Rights Law Review, Vol. 3, No. 2.
Cuéllar, Mariano - Florentino: Choosing Anti-Terror Targets by National Origin and Race, Harvard Latino Law Review, Spring 2003
Davies, Sharon: Reflections on the Criminal Justice System after September 11, 2001, Ohio State Journal of Criminal Law, Fall 2003
Delsol, Rebekah - Shiner, Michael: Regulating Stop and Search: A Challenge for Police and Community Relations in England and Wales, Critical Criminology (2006)
Delsol, Rebekah: Presentation to the LIBE Committee of the European Parliament, Brussels, 30 June 2008, http://www.europarl.europa.eu/document/activities/cont/200806/20080625ATT32712/20080625ATT32712EN.pdf
Dinant, Jean-Marc - Lazaro, Christophe - Poullet, Yves - Lefever, Nathalie - Rouvroy, Antoinette: Application of Convention 108 to the profiling mechanism Some ideas for the future work of the consultative committee, Consultative Committee of the Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data (T-Pd) 24th meeting 13-14 March 2008 Strasbourg, G01 (T-PD), Secretariat document prepared by the Council of Europe Directorate General of Human Rights and Legal Affairs, Strasbourg, 11 January 2008 T-PD(2008)01
Dolan Paul - Peasgood, Tessa: Estimating the Economic and Social Costs of the Fear of Crime, British Journal of Criminology, January 2007
ECRI General Policy Recommendation No. 11 on Combating Racism and Racial Discrimination in Policing
ECRI General Policy Recommendation No. 1 on Combating Racism, Xenophobia, Antisemitism and Intolerance
ECRI General Policy Recommendation No. 7 on National Legislation to Combat Racism and Racial Discrimination
ECRI General Policy Recommendation No. 8 on Combating Racism While Fighting Terrorism
Ellmann, Stephen: Racial Profiling and Terrorism, New York Law School Journal of Human Rights, 2003, No. 19.
- 296/297 -
Fassbender, Bardo: Quis judicabit? The Security Council, Its Powers and Its Legal Control, Review Essay, EJIL Vol. 11 (2000) No. 1.
Fraas, Michael: Sicherheitsrat der Vereinten Nationen und Internationaler Gerichtshof: Die Rechtmäßigkeitsprüfung von Beschlüssen des Sicherheitsrats der Vereinten Nationen im Rahmen des VII. Kapitels der Charta durch den Internationalen Gerichtshof. Frankfurt am Main, Peter Lang, 1998.
General recommendation XXXI on the prevention of racial discrimination in the administration and functioning of the criminal justice system (2005)
Goldston, James: Toward a Europe Without Ethnic Profiling, Justice Initiatives, Open Society Justice Intitiative, June 2005
Gross, Samuel R - Livingston, Debra: Racial Profiling Under Attack, Columbia Law Review, June 2002
Harris, David A.: New Risks; New Tactics: An Assessment of the Re-Assessment of Racial Profiling in the Wake of September 11, 2001, Utah Law Review, 2004
Harris, David A.: The Stories, the Statistics, and the Law: Why "Driving While Black" Matters, Minnesota Law Review, December 1999
Harris, David A.: Using Race as a Factor in Assessing the Reasonableness of Fourth Amendment Activity: Description Yes, Prediction No, Mississipi Law Journal Special Edition 2003,
Hayes, Ben: A Failure to regulate: Data protection and Ethnic Profiling in the Police Sector in Europe, Justice Initiatives, Open Society Justice Initiative, June 2005
Herbst, Jochen: Rechtskontrolle des UN-Sicherheitsrates. Frankfurt am Main, Peter Lang, 1999.
Humphreys, Stephen: The Case for Monitoring Ethnic Profiling in Europe, Justice Initiatives, Open Society Justice Initiative, June 2005
Joo, Thomas W.: Presumed Disloyal: Executive Power, Judicial Deference, and the Construction of Race Before and After September 11, Columbia Human Rights Law Review, Fall 2002
Lianos, Michalis: Hegemonic Security Discourse: Late Modernity's Grand Narrative, Tuesday 6 September 2005
Loader, Ian: Policing, Securitization and Democratization in Europe, Monday 18 April 2005, http://www.libertysecurity.org/article209.html?var_recherche=policing%2C%20securitization
- 297/298 -
Lock, Peter: Anti-terrorism and Effects on Freedom of Movement- Assessing the Concept of Progress in the Fight against Terrorism, Wednesday 20 July 2005, http://www.libertysecurity.org/article318.html?var_recherche=lock%20peter
Lodge, Juliet: Trends in Biometrics, Directorate-General Internal Policies, Policy Department C, Citizens Rights and Constitutional Affairs, Briefing Paper, 4 December 2006.
Mikos, Robert A: "Eggshell" Victims, Private Precautions, and the Societal benefits of shifting crime, Michigan Law Review, November 2006
Miller, Joel - Balogi, Anna - Bezlov, Tihomir - Gounev, Philip - Pap, András L - Pernas, Bengona - Simonovits, Bori - Vargha, Lili - Wagman, Dani: Racism and Police Stops. Adapting US and British Debates to Continental Europe, European Journal of Criminology, 2008:5
Miller, Joel - Bernát, Anikó - Dencső, Blanka - Gounev, Philip - Pap, András L - Pernas, Bengona - Simonovits, Bori - Wagman, Dan: "And I can stop and search whoever I want": Police stops of ethnic minorities in Bulgaria, Hungary and Spain, Open Society Justice Initiative, New York, 2007
Nojeim, Gregory: Aviation Security Profiling and Passengers' Civil Liberties, Air and Space Lawyer, Summer 1998
O'Rawe, Mary: Ethnic Profiling, Policing, and Suspect Communities: Lessons from Northern Ireland, Justice Initiatives, Open Society Justice Initiative, June 2005
Opinion 2006/4 of the Network of Independent Experts on Fundamental Rights on Ethnic profiling
Osler, Mark Capone and Bin Laden: The failure of government at the cusp of war and crime, Baylor Law Review, Spring 2003.
Padieu, R. (2000) «Mobiliser les données existantes: enjeux et conditions", in L'utilisation des sources administratives en démographie, sociologie et statistique sociale, Dossiers et recherches No. 86, INED
Pap, András L.: Constitutional Ambiguities Regarding Anti-Terrorist Financial Enforcement Measures - The Case of Hungary, Acta Juridica Hungarica, 47, No. 4, (2006)
Pap, András L.: Criminal Justice and Ethnic Data Collection in Hungary, Acta Juridica Hungarica, Hungarian Journal of Legal Studies, 2006/2 (Vol. 47., No. 2.)
Pap, András L.: Ethnic discrimination and the war against terrorism - The case of Hungary In: Gábor Halmai, (ed) Hungary: Human Rights in the Face of Terrorism, Special English Edition of Fundamentum, Human Rights Quarterly, Vandeplas Publishing: Human Rights Series 1, USA, 2006
- 298/299 -
Pap, András L.: Human Rights and Ethnic Data Collection in Hungary, Human Rights Review, Vol. 9, Issue 1, March 2008
Pap, András L.: Police ethnic profiling in Hungary - Lessons from an international research, Regio, A review of Studies on Minorities, Politics, Society, 2007, Vol. 10., pp. 117-140.
Ramirez, Deborah - Hoopes, Jennifer - Quinlan, Tara Lai: Defining Racial Profiling in a Post-September 11 World, American Criminal Law Review, Summer 2003
Recommendation (2001) 10 of the Committee of Ministers to Member States on the European Code of Police Ethics, Part V. Para 16 and 18.
Recommendations on Policing in Multi-Ethnic Societies by the High Commissioner on National Minorities
Simon, Patrick: "Ethnic" statistics and data protection in the Council of Europe countries. Study Report, Institut National d'Études Démographiques, Strasbourg, 2007
Sinha, Priyamvada: Police Use of Race in Suspect Descriptions: Constitutional Considerations, 31 New York University Review of Law and Social Change (2006)
Stone, Christopher: Preparing a Fresh Assault on Ethnic Profiling, Justice Initiatives, Open Society Justice Intitiative, June 2005
Stuntz, William J.: Local Policing After the Terror, Yale Law Review, June 2002
Thompson, Anthony Stopping the Usual Suspects: Race and the Fourth Amendment, New York University Law Review, October 1999
Racial and ethnic discrimination is an issue of key importance in the discourse on human rights. The paper discusses ethnic profiling and related law enforcement strategies as well as issues in data mining. Profiling means selecting suspects on the basis of alleged racial and ethnic characteristics, and establishing links between ethnic affiliation and potential criminal behaviour. Profiling can prompt the enhanced law enforcement inspection of certain
- 299/300 -
national, ethnic and religious groups. Thanks to the expanding body of European Community law, those combating terrorism and illegal migration can today rely on a growing arsenal of legal and administrative instruments. It is high time to reconsider the question to what extent can police or national security action be taken against certain individuals based on outside characteristics and based solely on the fact that the unidentified subjects concerned appear to be members of certain religious or ethnic groups. (These groups are statistically overrepresented among, or account for the dominant majority of the perpetrators of certain types of crime.) Thanks to modern information technology - at least as claimed by the advocates of those techniques -, automated procedures can be increasingly used to identify potential offenders with reliance on the above-mentioned criteria. Indeed, law enforcement agencies could hardly wait the advent of such equipment as it can carry out selection on a large scale without human interference.
The scope of the essay does not enable the author to offer an exhaustive examination of ethnic profiling. It has to put up with describing the most basic information on data mining and profiling. Among the issues discussed are the classification of ethnic groups in the course of law enforcement proceedings and the constitutionality of the causes that can serve as basis for launching related law enforcement measures.
Im Menschenrechtsdiskurs ist die Frage der Diskriminierung auf Grund der Rasse, bzw. der Ethnizität eine besonders Wichtige. Die Studie untersucht innerhalb dieses Themas die Problematik der Polizeistrategien, die auf der Auswahl auf ethnischer Grundlage und der Profilerstellung auf ethnischer Grundlage basieren, sowie die Problematik des "Data Mining". Das Thema der Studie ist die Untersuchung der Polizeiverfahren, die auf den vermeintlichen Zusammenhängen des "Profiling", also der Auswahl auf Grundlage der Kriterien der Rasse, bzw. der Ethnizität, d.h. der ethnischen Zugehörigkeit und der potentiellen Kriminalität basieren, und zu vermehrten Behördenkontrollen einzelner nationaler, ethnischer, bzw. religiöser Gruppen führen. Das sich erweiternde
- 300/301 -
Instrumentarium des Auftretens gegen den Terrorismus und die illegale Migration (auch als Ergebnis der Gesetzgebung der Europäischen Union) rückt folgende Frage in ein neues Licht: In wie weit kann diejenige bloße Tatsache als Grundlage von polizeilichen oder Verfahren der nationalen Sicherheit dienen, dass die einer Maßnahme unterzogene Person Mitglied einer (auf Grund äußerlicher Merkmale erkennbaren) religiösen oder ethnischen Gruppe ist, aus deren Mitgliedern in bedeutend höherer Zahl, eventuell in überwiegender Mehrheit, Täter bestimmter Verbrechen hervorgehen. Die moderne Informationstechnologie ermöglicht nämlich die Herausbildung und Verbreitung von automatisierten Verfahren, nach denen die Strafverfolgungsorgane bereits seit Langem streben, die eine menschliche Entscheidungsfindung ausschließen oder zumindest stark einschränken und die - zumindest den Befürwortern dieser Verfahren zufolge - in hohem Maße zur Identifizierung der potentiellen rechtsverletzenden Personen beitragen können.
In Anbetracht der Umfangsschranken beschäftigt sich die Studie - anstelle der umfassenden Analyse der auf ethnischer Grundlage ausgeführten Auswahl -lediglich mit der Bestimmung der Begriffe des Data Mining und der Profilerstellung. Im Rahmen dessen beschäftigt sie sich mit der Frage der ethnischen Klassifikationen im Laufe von Polizeiverfahren, sowie der Verfassungsmäßigkeit der Gründe und Motive, die als Grundlage der Einleitung dieser Behördenverfahren und -maßnahmen dienen. ■
NOTES
[1] The paper was written under the aegis of the Bolyai Research Scholarship of the Hungarian Academy of Sciences as part of the 68361 OTKA Research Grant.
[2] Ben Hayes, A Failure to regulate: Data protection and Ethnic Profiling in the Police Sector in Europe, Justice Initiatives, Open Society Justice Initiative, June 2005, p. 37.
[3] 2006/4.
[4] Ethnic Profiling and Counter-Terrorism: Trends, Dangers and Alternatives, June 2006
[5] R. v. Richards (1999), 26 C.R.(5th) 286 (Ont. C.A.). 15
[6] Adopted on 29 June 2007
[7] For a detailed analysis, see Jean-Marc Dinant, Christophe Lazaro, Yves Poullet, Nathalie Lefever, Antoinette Rouvroy: Application of Convention 108 to the profiling mechanism Some ideas for the future work of the consultative committee, Consultative Committee of the Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data (T-Pd) 24th meeting 13-14 March 2008 Strasbourg, G01 (T-PD), Secretariat document prepared by the Council of Europe Directorate General of Human Rights and Legal Affairs, Strasbourg, 11 January 2008 T-PD(2008)01
[8] Data mining can be defined as the application of statistical, data analysis and artificial intelligence techniques to the exploration and analysis with no preconceived ideas of (often large) computer data bases in order to extract fresh information that may be of use to the holder of these data. In other words, the value of data mining is that it is an IT tool which can "make the data talk". Generally speaking, the methods on which data mining is based can be divided into two categories: some are descriptive and others predictive, depending on whether the aim is to explain or predict a "target" variable. Descriptive methods are used to bring out information that is present but hidden within the mass of data, while predictive methods are used to exploit a set of observed and documented events in order to try and predict the development of an activity by drawing projection curves. This method can be applied to management of customer relations in order to predict a customer's behaviour. The aim is for example to determine the profile of individuals with a high purchasing probability or to predict when a customer will become disloyal. Id. p. 8-9.
[9] Id. p. 3.
[10] Id. p.5.
[11] L. A. Bygrave, « Minding the machine: Article 15 of the EC Data Protection Directive and Automated Profiling », Computer Law & Security Report, 2001, vol. 17, pp. 17-24, available on-line at http://www.austlii.edu.au/au/journals/PLPR/2000/40.html.
[12] Application of Convention 108., p. 5.
[13] See Id. p. 13.
[14] Directive 95/46/EC
[15] 1. Member States shall grant the right to every person not to be subject to a decision which produces legal effects concerning him or significantly affects him and which is based solely on automated processing of data intended to evaluate certain personal aspects relating to him, such as his performance at work, creditworthiness, reliability, conduct, etc. 2. Subject to the other articles of this Directive, Member States shall provide that a person may be subjected to a decision of the kind referred to in paragraph 1 if that decision: (a) is taken in the course of the entering into or performance of a contract, provided the request for the entering into or the performance of the contract, lodged by the data subject, has been satisfied or that there are suitable measures to safeguard his legitimate interests, such as arrangements allowing him to put his point of view, or (b) is authorized by a law which also lays down measures to safeguard the data subject's legitimate interests.
[16] For example, the Ombudsman in Cyprus - which delivers opinions on issues of equality and discrimination- delivered an opinion on a particular case of ethnic profiling on 23 June 2005. The Ombudsman decided that the policies of insurance companies to exclude Pontian Greeks from getting insured and/or have them insured at a higher rate, is discriminatory. The insurance companies argued that the reason why it is more difficult for Pontian Greeks to get insured is because of their past practice which has shown that they are more prone to accidents and usually do not cooperate well with the insurance companies. The Ombudsman noted that "such a policy is based on the racial or ethnic origin of the insured and is unacceptable as it violates the Equal Treatment (Racial or Ethnic) Origin Law." See the E.U. Network of Independent Experts on Fundamental Rights Opinion 2006/4 on ethnic profiling.
[17] Application of Convention 108., p. 4-5 and 10-11.
[18] The Federal Constitutional Court decided on April 4th, 2006, that the Rasterfahndung was not conform to the individual's fundamental right of self-determination over personal information. See Rebekah Delsol, Presentation to the LIBE Committee of the European Parliament, Brussels, 30 June 2008, http://www.europarl.europa.eu/document/activities/cont/200806/20080625ATT32712/20080625ATT32712EN.pdf
[19] Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin
[20] Para 22.
[21] Para 36.
[22] Op cit.
[23] Id.
[24] Id.
[25] Id.
[26] For example, in the 1979 Delaware v. Prouse case (440 US 648, (1979) the Supreme Court found random stops and checks to be unconstitutional. (See for example Anthony Thompson, Stopping the Usual Suspects: Race and the Fourth Amendment, New York University Law Review, October, 1999, pp. 973-974.) In the 2000 City of Indianapolis v. Edmund case (531 US 32 (2000)), the Court still found it unacceptable to have a road check following a roadblock, with the involvement of drug-searching dogs. But the Court upheld a roadblock in the context of a 2004 investigation concerning a hit-and-run accident, when during the time when the crime was committed, the road was blocked and without using further coercive measures the police politely asked motorists about the case, showing them photographs. (Illinois v. Lidster, (000 U.S. 02-1060 (2004)), see also Thompson, p. 920. The question is, of course, whether the Court's position would be similar if the roadblock were put in front of a mosque or a Middle-Eastern grocery store...
[27] For alcohol tests ordered for railroad workers involved in an accident see Skinner v. Railway Labour Executives Assn, (489 US 602, (1989)); for sobriety tests around nightclubs see Police v. Sitz (496 US 444, (1990)), and for alcohol tests prior to after-hours extra-curricular school events see for example, Veronia Sch. Dist. 47J v. Acton (515 US 646 (1995)), Bd. of Educ. v. Earls (536 US 822 (2002))
[28] US v. Brignoni-Ponce, 422 US 873 (1975)
[29] 428 U.S. 543 (1976)
[30] See Sharon Davies: Reflections on the Criminal Justice System after September 11, 2001, Ohio State Journal of Criminal Law, Fall, 2003, p. 66.
[31] See, http://www.pszaf.hu/english/start.html
[32] 440 U.S. 648 (1979)
[33] 496 u.s. 444 (1990).
[34] 540 u.s. 419 (2004)
[35] Priyamvada SINHA: Police Use of Race in Suspect Descriptions: Constitutional Considerations, 31 New York University Review of Law and Social Change (2006), pp. 144-45, 149. and 155.
[36] See Sinha, pp. 135-139.
[37] 394 U.S. 721 (1969)
[38] 590 A.2d 1008 (D.C. 1991)
[39] Commonwealth v. Jackson, 331 A.2d 189, 191 (Pa. 1975)
[40] Robinson, 507 N.Y.S.2d at 269-70
[41] 242 F.3d 215, 216, 218-19 (4th Cir. 2001)
[42] 5 9 7 N.E.2d 1029, 1031 (Mass. 1992)
[43] 5 7 4 S.W.2d 764, 765-66 (Tex. Crim. App. 1978)
[44] 481 S.W.2d 106, 110-12 (Tex. Crim. App. 1972)
[45] There are other, more recent cases that include a different context for stop and search: in Buffkins v. City of Omaha, 922 F.2d 465, 467, 470 (8th Cir. 1990) the court held that a tip that a black person or persons arriving on a flight from Denver would be importing cocaine to the Omaha, Nebraska, area before 5:00 p.m. on March 17, 1987, was not sufficient to justify stopping a Black woman carrying a toy animal. In United States v. Grant, 920 F.2d 376, 388 (6th Cir. 1990) the court found no reasonable suspicion where Border Patrol agents in Detroit, Michigan airport stopped defendant because he was man of colour with dreadlocked hair and his flight originated in Los Angeles, California, an alleged "drug source" city. In Orhorhaghe v. INS, 38 F.3d 488, 497 (9th Cir. 1994) the court held that the sole basis for suspicion was racial background or national origin, assumed from defendant's "foreign-sounding" surname, which did not satisfy reasonable suspicion for seizure. In United States v. Tapia, 912 F.2d 1367, 1368, 1371 (11th Cir. 1990) the court held that Mexican ancestry, possessing few pieces of luggage, being visibly nervous, and travelling with out-of-state license plates is not enough to yield reasonable suspicion for additional detention after initial traffic stop.
[46] A biometric is a measurement and biometric identifiers are supposed to uniquely identify and reliably confirm an individual's identity.
[47] Juliet Lodge: Trends in Biometrics, Directorate-General Internal Policies, Policy Department C, Citizens Rights and Constitutional Affairs, Briefing Paper, 4 December 2006.
[48] Inter-operability is a term used to refer to the capacity of databases to be linked or interrogated to enable deeper searches for information on a given individual subject.
[49] Lodge, op.cit.
[50] As illustrated, for example, by the PNR arguments before and after 30 September 2006. Lodge, id.
[51] Lodge, op cit.
[52] Id.
[53] Rebekah Delsol, Presentation to the LIBE Committee of the European Parliament, Brussels, 30 June 2008, http://www.europarl.europa.eu/document/activities/cont/200806/20080625ATT32712/20080625ATT32712EN.pdf
Lábjegyzetek:
[1] Institute for Legal Studies, Hungarian Academy of Sciences, Telephone number: (36-1) 201 9717-886, E-mail: pap@jog.mta.hu
Visszaugrás
