Introduction

Identifications of persons1 represents an important police task in the early stages of the investigation of the committed crime when police agencies are trying to identify suspects (National Academy of Sciences, 2013). Also, the results of identifications may be used as evidence in criminal proceedings in many countries, including Serbia. Therefore, there is a need for serious consideration of the validity of relevant criminal procedure rules regarding identification of persons, not only for the actual perpetrators but also for innocent suspects.

According to the "Innocence Project"2, a well-known advocacy group engaged in exonerating the wrongly convicted through DNA testing and reforming the criminal justice system to prevent future injustices, witness misidentification is the leading cause of wrongful convictions. Given that some missidentifications may be caused by the way in which law enforcement handles identification procedures, it is important to adopt policies and procedures on witness identification that are based on research by experimental psychologists and others (National Academy of Sciences, 2013). For several decades, scientists have conducted research on the factors that affect the accuracy of witness identification procedures and have identified two main categories: (a) factors over which the criminal justice system has no control (usually referred to as "estimator variables", e.g. the age of a witness, the degree of stress or trauma a witness experiences, viewing conditions, a presence of a weapon during the crime, etc.) (Wilford & Wells, 2013, p. 24) and (b) factors over which the criminal justice system has control (usually referred to as "system variables", e.g. lineup format, lineup size, lineup presentation, etc.) (Mickes, 2015, p. 93).

The use of various procedures for conducting identification of persons is determined by the provisions of the Criminal Procedure Code of the Republic of Serbia (hereinafter: the CPC) (Criminal Procedure Code of the Republic of Serbia, 2011-2021), so it can be assumed that the quality and precision of procedure rules will affect both the legality and accuracy of the conducted identifications. In order to prevent the exclusion of identification of persons as evidence, it is important to determine compliance of the main procedure rules' contents with the most important scientific findings on factors that affect the identification accuracy over which authorities have control. Given that crime investigators'3 behaviour in preparing and administering lineup identification procedures is often challenged in Serbia by requests for protection of legality submitted to the Supreme Court of Cassation of the Republic of Serbia (hereinafter: the SCC), it is important to determine how the case law treats identification evidence. These requests are relevant, given that they represent an extraordinary legal remedy that an authorized person (the Republic public prosecutor, the defendants, and their defense counsels) may submit against a final decision of the court or for a violation of provisions of the procedure which preceded its issuance (Art. 482, para. 1 of the CPC).

Given that the court did not exclude the identification evidence in the previous proceedings, the aim is to determine the validity of the assertions made in these requests regarding the legality of conducted police identification procedures, and then to provide recommendations regarding improving the normative framework that would enable establishment of up to date and well informed standards for the collection, preservation and presentation of identification evidence.

The normative basis and criminal procedure rules in the identification of persons in The Republic of Serbia

Identification of persons is prescribed by the CPC, under the provisions which prescribe an evidentiary action regarding the questioning of witnesses (Art. 100 of the CPC), but also an interrogation of the defendant (Art. 90 of the CPC). A normative basis regarding the identification of persons is primarily established by the provisions which prescribe the conduct of the identification procedures with witnesses. Therefore, the witness identification shall be conducted if it is necessary to establish whether a witness can identify a certain person or object, or their characteristics as he or she had previously described them (Art. 100, para. 1 of the CPC). Regarding the manner in which identification procedures shall be conducted in the pre-investigation proceedings and during the investigation, there is a request that competent authorities must prevent the person possibly being identified from seeing the witness, and the witness from seeing that person before the formal identification procedure (Art. 100, para. 2 of the CPC). Further, the legislator prescribed that the identification of person may be conducted in the pre-investigation proceedings and during the investigation in the presence of the public prosecutor (Art. 100, para. 3 of the CPC), which indicates to the conclusion that the identification of persons conducted by the police will have procedural value only if the public prosecutor was present (Ilić et al., 2013).

Although the legislator did not explicitly authorize the police to conduct the identification procedure, there are two reasons for interpreting that the police represent the competent authority conducting identification procedures. The first reason is that the public prosecutor and the police represent the main figures of the pre-investigation proceedings and investigation (Bošković & Kesić, 2020). Also, the public prosecutor may refer conduct of certain evidentiary actions (even the identification of persons) to the police (Art. 299, para. 4 of the CPC). By prescribing that the public prosecutor shall be "present" during the identification procedure, it may be concluded that the main activities in preparing and administering such procedure shall be on the police. Also, Art. 287 of the CPC gives police the authority to conduct an evidentiary action during the pre-investigation proceedings but must inform the public prosecutor thereof without delay and conduct it in accordance with the CPC. The second reason concerns the competence of the police to conduct identification, both in terms of criminalistic knowledge and in terms of preconditions needed for conducting the identification procedure (e.g., lineup construction, providing the premises with special mirrors for identification). However, such a legal solution neglects the provisions of Art. 288, para. 2 of the CPC, which prescribes that the police may not question a citizen in a capacity of defendant, or in a capacity of witness or expert witness, except in the case referred to in Art. 289 of this Code (Škulić, 2020). Therefore, an exception to this rule refers only to the interrogation of the suspect.

Further details regarding the manner of conducting the identification procedure are prescribed by Art. 90 of this Code, under the chapter regarding the evidentiary action of interrogation of the defendant, because he/she may also identify the other defendants or other persons. Given that Art. 100 para 1 of the CPC prescribes that the witness identification shall be performed in accordance with Art. 90 of this Code, all procedure rules regarding the defendants' identification apply also to the witness identification.

Based on the provisions of Art. 90, para. 1 of the CPC, the person to be (possibly) identified, whose basic characteristics are similar to those that the witness has previously described, will be shown to him/her together with other persons not known to him/her. The requirement of 'other persons' is important because it advocates against the potential use of show-ups in which one person who is suspected of having committed the crime is offered to the witness with the option to identify this person as the perpetrator or not (Colloff & Wixted, 2020; Lawson & Dysart, 2012). Finally, it is important to note that the competent authorities are obligated to ask a witness to state whether he/she can identify that person either with full certainty or with a degree of certainty, and, if so, to point to the person thus identified (Art. 90, para. 2 of the CPC).

Besides in-person identifications, the CPC prescribes the possibility of conducting identifications based on the photographs, if the person possibly being identified is not available/accessible. In that case, the authorities may conduct the identification by showing the photograph of that person together with other photographs of persons whose basic characteristics are similar to those previously described (Art. 90, para. 3 of the CPC). Finally, identification of a person may also be conducted on the basis of his/her voice (Art. 90 para. 4 of the CPC).

The present article considers current procedure rules and their application in collecting the identification evidence in the light of the case law of the Supreme Court of Cassation of Serbia and the main scientific findings on relevant factors that affect the accuracy of witness identification over which authorities have some control (usually referred to as "system variables") (Wells, 2018).

Criminal procedure rules regarding identification of persons in the light of the case law of the Supreme Court of Cassation

The present article analyzes 33 judgments issued by the Supreme Court of Cassation in the period from 2013 to 2021, regarding the request for protection of legality submitted by a defense counsel. In 30 analyzed requests, the defense counsels disputed the legality of identifications of persons conducted by police in the pre-investigation proceedings and investigation. In the remaining three requests, the identification of persons was not conducted, but the defense counsels disputed the testimony of the witnesses regarding the perpetrator's identity (because of the improper behaviour by the crime investigators).

In-person and photo-identifications

Based on relevant procedure rules, the police should conduct in-person identifications by using live lineups with one suspect (whenever a suspect is accessible). This implies that identification based on the photographs shall be conducted only if the person (possibly) being identified is not available/accessible, but this must be in line with all procedural rules referred to in-person identifications. Analyzed SCC case law indicates that Serbian police conduct predominantly in-person identifications. Out of 30 analyzed requests submitted to the SCC in which an identification procedure was conducted - 25 requests referred to in-person identifications (83.3%), while only 5 requests referred to the identifications based on the photographs (16.7%).

The conduct of in-person identifications has its advantages, because the witnesses may observe the lineup members in their entirety (i.e. their faces, full bodies, walking, and talking). Since the witnesses have more information, there is a belief that it is likely to lead to more accurate identifications (Price et al., 2019). On the other hand, live lineups are not always achievable, because they require the lineup members-the suspect and fillers (i.e. known innocent persons which are shown together with the suspect) and other parties (e.g. witnesses, police personnel, public prosecutor, etc.) to be physically present for the identification procedure (Fitzgerald et al., 2018). Given that, there is a need to determine the legal preconditions for conducting identifications based on photographs.

According to the CPC, the identifications based on the photographs shall be conducted only if the person (possibly) being identified is not available/accessible. If photo-identification procedure was conducted in line with all procedural rules regarding in-person identifications, the court's decision may be based on this evidence. A SCC judgment issued regarding the request for protection of legality supports this. According to the opinion of the SCC made regarding this request "the injured party made positive identification of the suspect based on five photographs, in the presence of the public prosecutor and authorized police officers, thus, it can be considered as evidence collected under the CPC" (Supreme Court of Cassation [SCC], 2020b4).

Given that photo-identification may be evidence, there is no need to conduct a live identification procedure when the suspect becomes available. This is important because provisions of the previous CPC did not stipulate photo-identification as an evidentiary action, so defense counsel could not challenge the legality even though it preceded the live lineup identification. Rather, the prior showing of photographs was deemed an operational activity, which had no evidentiary value. The intent of the actual CPC is to have such photo-identification as an evidentiary action when the police know the identity of the suspect, but he/she is not available/accessible.

The case law of the SCC demonstrates that police conduct photo-identifications as well when the suspect is available/not available. Based on the assertions of the defense counsels made in the three requests, photo-identifications were conducted when the suspect was not available, which was in accordance with the provisions of the Art. 90, para. 3 of the CPC (SCC, 2021e5; 2020b6; 2019e7). Therefore, the SCC assessed these assertions as unfounded and dismissed these requests.

However, in two requests submitted to the SCC, the defense counsels disputed the legality of conducted photo-identifications, because the suspects were arrested and brought to the police premises, and thus were accessible for in-person identification (SCC, 2021c8; 2019d9). The SCC also dismissed both requests as unfounded. Given that mentioned requests were based on quite the opposite assertions, it is questionable why the court made the same decision (to dismiss these requests), by assessing that these challenged photo-identifications were conducted in accordance with the CPC.

Nominal lineup size - the number of persons presented in a lineup

One of the first issues which crime investigators should consider is the way of constructing a lineup as a method for establishing that the suspect and the culprit/perpetrator are or are not the same person (Brewer & Doyle, 2021). According to Wells et al. (2020), the lineup should contain at least five appropriate fillers for every one suspect, who do not make the suspect stand out in the lineup (based upon physical appearances or other contextual factors, such as clothing or background). Many police agencies have applied this recommended minimum, so a typical police lineup contains six members - one suspect plus five similar-looking fillers (i.e. known-innocent persons) (Collof et al., 2021; Wooten et al., 2020).

The CPC did not define the number of fillers to be shown together with the suspect, which may cause different lineup sizes. By interpreting the provisions of Art. 90, para. 1 of the CPC, which prescribes that the suspect shall be shown to the witness together with "other persons", it could be concluded that a lineup must comprise at least two fillers. It is interesting that the CPC/200610 (Criminal Procedure Code of the Republic of Serbia, 2011-2021) set a minimum of five and a maximum of eight fillers who shall be shown together with the suspect to a witness. It is questionable why the current CPC did not retain this research informed legal solution, given that certain legal solutions regarding the witnesses (e.g. procedural rules referred to the witness protection) contained in that Code were retained and still apply (Škulić, 2007). Thus, since the current CPC did not prescribe a precise number of persons in lineups, the crime investigators may construct lineups comprised of two or three fillers, and such behaviour could not be legally challenged.

The results of analyzed SCC case law supports this. Out of 30 analyzed requests submitted to the SCC regarding conducted identifications, information regarding the fillers number was available in 17 requests. In most requests (10), the witnesses were presented with the lineups in which the suspect was placed among four fillers, which indicates that a typical police lineup in Serbia is not in line with the above mentioned minimum of at least five appropriate fillers for every one suspect. In two requests, police lineup comprised only three fillers. Only in three requests the number of fillers was in line with the above mentioned minimum-in one request, the lineup comprised 7 persons (6 fillers plus the suspect), but this identification was conducted in the Republic of Srpska, while in another two requests the lineup comprised 6 persons (5 fillers plus the suspect).

Collecting description of the perpetrator and filler selection strategy

One of the principal activities of crime investigators regarding constructing a lineup is to select appropriate fillers, i.e., persons among which the suspect will be placed in a lineup. There are two main strategies for fillers selection: (a) a description-matched strategy (i.e. choosing fillers because of their match to a description of the culprit given by a witness), and (b) a suspect-matched strategy (i.e. choosing fillers because of their match to the suspect) (Fitzgerald et al., 2013). Given that the police conduct an identification in order to establish whether a suspect and a culprit/perpetrator are or are not the same person, there is a possibility that the culprit is not present in a lineup that will be presented to the witness. Therefore, a description-matched strategy is thought to be the best to protect the innocent suspect in a culprit-absent lineups (Wells et al., 2020). However, it is not always possible to follow a strict description-matched strategy in practice, so when the description of a perpetrator is very vague, or such description differs significantly from the suspect's appearance (Colloff et al., 2021), it may be acceptable to use a suspect-matched strategy.

According to the Art. 90, para 1 of the CPC, a suspect will be shown to the witness together with other persons not known to him/her whose basic characteristics are similar to those he/she has previously described, which indicate that the police should choose fillers based on the description of the perpetrator previously given by the witness. But is that really the case, or do the police actually use a suspect-matched strategy? In order to answer this question, it is necessary to mention several issues.

First, the above mentioned provision of the CPC obligates crime investigators to obtain the description of the perpetrator before conducting the identification procedure. Thus, this description shall be part of the record of the witness identification procedure. The question is whether the record only repeats the description given by the witness (e.g. during the prior interview (i.e. collecting information) often immediately after the commission of the crime) or is the witness obliged to provide a description again?

As a rule, the witness shall describe the perpetrator just before the beginning of the identification procedure, in front of the public prosecutor. It is quite possible that this description will differ from the one that was obtained prevously, given that numerous research studies generally suggest poorer memory performance with an increase in the interval between witnessing an event and being examined on it (usually referred to as 'retention interval') (Lin et al., 2019; Wixted et al., 2018). At the moment of collecting description from the witness, just before the beginning of the identification procedure, crime investigators will have already constructed the lineup, which implies that they have already selected fillers (possibly according to the characteristics of the suspect). Another fact in favour of the notion that police in Serbia use a suspect-matched strategy is that crime technicians shall photograph the lineup before conducting the identification procedure in order to substantiate the similarity of the fillers with the suspect.

Regarding documenting the description given by witnesses, in one request submitted to the SCC, the defense counsel alleged that the witness did not describe the perpetrator before the identification procedure. The SCC dismissed this request, by explaining that "the CPC does not stipulate an obligation on the authority to question the injured party as a witness before identification, so there is thus no need for documenting the description in this manner" (SCC, 2015c11).

In another request, the defense counsel also alleged that witnesses did not give a description of the perpetrator who shot at them, but were asked to state whether they can identify the perpetrator among the persons presented in the lineup (SCC, 2014c12). The SCC deemed that the identification was conducted in line with the CPC and dismissed this request. In another SCC judgment issued regarding similar assertions, the court deemed that “the witness should describe the perpetrator before identification, but this step is not actually necessary, nor does it affect the legality of the examined evidence or its probative value” (SCC, 2020e13).

The selection of fillers is an important issue, which is why it is necessary that CPC clearly specify which basic characteristics of lineup members must be similar (e.g., gender, age, hair, complexion, body constitution, etc.). The current CPC did not specify a standard of similarity of fillers in the lineup, so it is possible that a suspect may be physically distinctive from the fillers in lineups. The reason fillers may not be "sufficiently similar" could be the fact that crime investigators who are organizing the lineup find the fillers just before the beginning of identification, by choosing them among the police officers who are accessible at the time.

The case law of the SCC actually supports this. Based on the asertion made in one request submitted to the SCC, the suspect was shown to the witness together with the persons who were not similar in build, height, appearance, and age, one of them was of Roma ethnicity and others were police officers involved in the investigation, familiar to the witness (SCC, 2020f14). The court dismissed the request. In another request, a defense counsel alleged that witness stated at the main trial that persons in lineup shown to him were "both lower and higher and black hair and all sorts", but the SCC deemed that the standard of basic characteristics was fulfilled because the lineup comprised five people who were similar in age (SCC, 2015c15). The SCC also dismissed yet another request, by using the argument that the persons in the lineup were similar enough because they were essentially the same age (38–44 years), even though the defense counsel alleged that persons in the lineup differed in their height and build (SCC, 2020h16). The SCC dismissed also another two requests regarding the filler similarity (SCC, 2021b17; 2015b18). In one of these requests, the defense counsel made a remark regarding the biased lineup during the identification procedure, but the SCC dismissed the request by noting that it still did not affect the legality of this action (SCC, 2015b19).

However, the SCC accepted two requests in which the defense counsels alleged the police presented to the witnesses the lineups, which comprised several persons suspected of committing a crime. The SCC case law indicates that such behaviour violates provisions of criminal procedure (SCC, 2019b20; 2019c21). Therefore, if there are several suspects, the police must conduct an identification for each of them.

Prohibition of repeated identfications

The use of repeated identifications, i.e. testing the witness's memory on two separate occasions, with the suspect as the only person to appear both times, represents one of the most controversial issues in conducting identifications in Serbia. The results of the analyzed SCC judgments indicate that repeated identifications are very present in practice, in a form of exposing witnesses only to the photographs of the suspects or 'mugshots' (i.e. a series of photographs of the registered perpetrators), before the identification in live lineups.

Based on relevant research (Steblay & Dysart, 2016; Steblay et al., 2013) there are serious grounds for being concerned that such behaviour will unduly effect witnesses' identification of the suspect in a later lineup. This is because it may introduce a bias, so (а) the witnesses may select the suspect from the lineup due to the misplaced familiarity from prior seeing the suspect's photographs (not from having committed the crime), and (b) witnesses may tend to stay committed to their initial positive identification even if it is wrong (Lin et al., 2019). The use of "mugshots" represents relatively routine activity in the early stages of investigation when police try to establish the identity of the suspect. This is also the case in Serbian practice, since the use of "mugshots" may be treated as an operational activity which police may implement in order to locate the perpetrator of the criminal offense (Art. 286, para. 1 of the CPC). However, the above mentioned experimental studies suggest that viewing 'mugshots' detrimentally influences identification accuracy in subsequent lineup tasks.

The prohibition of repeated identification may also be considered regarding Art. 100, para 2 of the CPC, under which the witness should only see the suspect at the formal beginning of the identification procedure. That implies that the witness will enter the special premises which prevent the witness from seeing the suspect (and vice versa) until the lineup has been formed. Thus eliminating the danger to the witness himself/herself that may arise from the person being identified, but also prevents the possibility of influencing the witness to identify a certain person as the perpetrator (Ilić et al., 2013). But does this provision prohibit showing the witness a photograph of the suspect before conducting the identification procedure? The case law of the SCC indicates that there is no clear criterion for establishing the difference between using photographs as (a) operational and (b) evidentiary action.

In three requests submitted to the SCC, the defense counsels disputed the witnesses’ testimonies regarding the identity of the perpetrator, because of their prior viewing of the suspects’ photographs (SCC, 2021a22; 2020c23; 2014a24). Regarding two of these requests, the SCC deemed that there was no identification conducted according to Art. 90 and 100 of the CPC, but that the witnesses only testified about the manner in which they found out the identity of the perpetrator, which is why it represented a part of their testimonies that can be used as evidence in criminal proceedings. The SCC repeated this in yet another dismissed request (SCC, 2014e25).

In another dismissed request, the witness stated at the main trial that the police showed him the photographs of the suspect before the identification in the lineup, but after the intervention of the public prosecutor and the judge he changed his statement that he is not sure when that happened (before or after a lineup identification) (SCC, 2019f26). In another request, the defense counsel also disputed such behaviour of investigators, but the SCC assessed these assertions as unfounded, by noting that “the identification was conducted according to the CPC and that the witness identified the suspect in the lineup with full certainty“ (SCC, 2020g27). The same assertions were made in another three dismissed requests, but the SCC deemed that prior showing the suspects’ photographs did not affect the legality of the identification, but only challenged the validity of the witness’s testimony (SCC, 2015a28; 2015c29; 2014b30).

In yet another request, the defense counsel alleged that the identification procedure was unlawful because the witness found and saw the suspect’s photographs on social media Facebook before the identification procedure. The SCC accepted the assertion that the witness identified the suspect in the mentioned photographs and that her defense counsel delivered them to the police before the lineup identification. However, the SCC dismissed this request by deeming that “this did not affect the witness, because she found these photographs based on her own knowledge about the appearance of the perpetrator” (SCC, 2020i31).

Similar assertions were observed in another two requests – in one, the witness saw the suspect’s photographs taken from a surveillance camera and the social media (SCC, 2021d32), while in another request, the suspect’s photographs were published in the daily newspapers (SCC, 2016a33). The SCC dismissed both requests by noting that these circumstances did not affect the legality of the conducted lineup identification, but only challenged the factual basis established in the previous proceedings.

In another request, the defense counsel alleged that showing the suspect’s photograph to the witness by investigators violated the provisions of Art. 100, para. 2 of the CPC, which prescribe (as mentioned above) to conduct identifications in a way that prevents the witness from seeing a suspect before the formal identification. It is interesting that the SCC dismissed this request by explaining that Art. 90, para 3 of the CPC (which prescribes photo-identification) allows authorities to show a photograph of the suspect (if he/she is not available) to the witness, together with other photographs of similar-looking persons (SCC, 2020a34).

The SCC repeated this in another judgment, by explaining that "the identity of the suspect was not known at that time, and thus unavailable, and that the witness identified the suspect by seeing photographs of different persons, not only the photograph of the suspect" (SCC, 2018a35). The court deemed that such behaviour did not affect the legality of the identification conducted in a subsequent lineup, and dismissed the request.

Although in several judgments (as mentioned above) the SCC deemed that the provisions of the CPC which prescribe photo-identification (as evidentiary action) allow prior showing the suspect's photographs to the witness, in the following two judgments, the SCC deemed that "such behavior represents an operational activity undertaken in order to locate the perpetrator, which does not affect the legality of the identification conducted in a subsequent lineup” (SCC, 2020h36; 2018b37).

Documenting the witness decision and a degree of certainity

The justice system relies on witness certainty as a measure of the validity of identification evidence, so it represents a significant determinant of whether the court will admit the testimony of the witness as probative evidence. Therefore, it is important to secure a statement of the witness's certainty at the time of the identification (before certainty can be inflated). Given that witness's certainty may be influenced by the procedures used to collect identification evidence (Wilford & Wells, 2013), it is useful to consider the SCC judgments which contain information regarding the degree of witness's certainty.

Acording to the provisions of Art. 90, para. 2 of the CPC, authorities are obligated to ask a witness to state whether he/she can identify anyone with full certainty or with a degree of certainty and to make a record of what the witness says. The analized SCC judgments indicate that there is a significant number of witnesses who expressed full certainity in their identifications-out of 33 analyzed SCC judgments, witnesses expressed full certainty in their identifications in 18 judgments, while the remaining judgments did not contain such information. Accordingly, the number of such witnesses could be even higher.

In one request submitted to the SCC, there were no suggestive behaviours that could affect the witness’s certainty (SCC, 2019d38), while assertions made in another two requests were referring to the absence of the defense counsel during the identification (SCC, 2021e39; 2020b40). Assertions made in the remaining 15 requests referred to one or more suggestive behaviours that could affect the witnesses’ certainty.

The witnesses saw the suspect before the identification (13 requests)

The analyzed requests submitted to the SCC indicate that showing of suspects' photographs to the witness represents the most common suggestive behaviour which precedes the formal identification procedure. Given that a defense counsel disputed such behaviour in 13 requests, it could be concluded that many fully certain witnesses were exposed to so-called "repeated" identification.

In nine requests, the defense counsels alleged that the crime investigators showed photographs of the suspect to the witnesses (SCC, 2020a41; 2020g42; 2020h43; 2019a44; 2018a45; 2018b46; 2015a47; 2014b48; 2014d49). All these witnesses identified the suspect in later lineups and expressed full certainty in their identifications – 100%, except two witnesses who expressed 99% and 90% of certainty.

Based on the assertions made in another two requests, the witnesses saw the suspect’s photograph on social media/from a surveillance camera, after which they identified the suspect in a lineup with full certainty (SCC, 2021d50; 2020i51). In another two requests, the defense counsels alleged that witnesses saw the suspect in person before a lineup identification (SCC, 2021d52; 2014b53). The SCC dismissed these requests by noting that identification was conducted in line with the CPC, inter alia, regarding the request which prohibits the viewing of the person being identified before the formal identification procedure.

Suggesting/preparing the witness to make a positive identification (3 requests)

In two requests, the defense counsels alleged that the crime investigators showed a photograph of the suspect to the witnesses and suggested to them to make a positive identification, which they did in a later lineup (SCC, 2015a54; 2014e55). Based on the assertions made in the third request, the witness stated at the main trial that the investigators brought and showed him a photograph of the suspect while he was in the hospital and suggested to him “they have been looking for that perpetrator for a long time”, after which he concluded that the person on the photograph is the perpetrator (SCC, 2014e56). Although this witness was not sure about the identity of the perpetrator, he identified the suspect in a later lineup with full certainty. Тhe SCC also dismissed these requests.

Biased lineups (7 requests)

In three requests submitted to the SCC, the defense counsels alleged that the police used biased lineups in which the suspects stood out because the fillers were not similar to them in basic characteristics (SCC, 2020d57; 2020h58; 2014b59). Based on the assertions made in another request, the suspect stood out because she had visible injuries (SCC, 201760), while in another request, the lineup was biased, given that four other members in a lineup were of different ages (significantly younger), physical appearance (of stronger build and different hair color), and clothes from the suspect (only the suspect was wearing customs officers’ uniform) (SCC, 2020i61). In another request, the witness identified the suspect with 90% of certainty in the lineup in which only one member was of the Roma ethnicity, which was a distinctive feature of the suspect (SCC, 2016b62). In yet another request, the witness identified the suspect with 100% of certainty in the live lineup in which the fillers were not similar in height, which was the only distinctive feature of the perpetrator described by the witness (SCC, 2014e63). Тhe SCC also assesed all these assertions as unfounded and dismissed requests.

Discussion and conclusion

According to the above analysis of relevant criminal procedure rules and case law of the Supreme Court of Cassation in Serbia regarding the identification of persons, several recommendations may be made. The first recommendation is that there is a need to set a minimum of at least five appropriate fillers for every one suspect. The second recommendation refers to establishing the standard of similarity of persons in the lineup, by stipulating that the suspect will be shown together with persons who are similar in basic characteristics, such as gender, age, build, hair and other. This issue is important because in all our analyzed SCC requests in which the lineups were biased (i.e., they were not in line with this recommendation), the court ruled that there was no violation of the CPC. By prescribing the recommended standards for fillers would prevent the police from determining physical appearance of fillers in lineups at its discretion.

It is also important to mention that there are no efficient mechanisms to prevent the use of repeated identifications, given that the SCC treats the prior showing of photographs of the suspect as an operational activity or legitimizes such behaviour based on the argument that the CPC provisions related to photo-identification allow it. The analyzed SCC requests indicate that such suggestive behaviour is very present and does not represent a violation of the provisions of criminal procedure, although it may inappropriately increase the number of witnesses who identified the suspect with full certainty in a later lineup identification.

Concerning the opportunities for motions to suppress identification evidence as one of the legal safeguards to prevent mistaken identification, the case law of the SCC indicates that there is little guarantee that it will provide the intended protection in Serbia. The fact that out of 33 analyzed requests, the SCC dismissed 31 of them supports this contention. By accepting the remaining two requests as founded, the SCC verified only one research informed recommendation, which refers to the use of lineups with one suspect. Accordingly, if there are several persons suspected committing of a crime, competent authorities must conduct an identification procedure for each of them. Otherwise, a violation of the provisions of the criminal procedure will exist.

The present study confirmed Latin legal phrase that not everything that is legally permitted is honest (Non omne quod licet honestum estin), given that the current criminal procedure rules allow various inappropriate behaviours before and during the identification procedures, even though they may affect its accuracy, and that the SCC treats them as lawful. Accordingly, it is controversial for the court to use the identification evidence in the criminal proceedings as a measure for determining the validity of the witnesses' statements, even though the manner in which such evidence is collected is not valid, i.e. does not provide a valid basis for such assessment. Given that there are clear and strong scientific findings (as mentioned in the current article) supporting the contention that currently there is a lack of appropriate, research informed standards for the collection, preservation, and presentation of identification evidence in Serbia, it is justified to initiate the amendments to the CPC provisions regarding the action of identification of persons.

Dodatak

Practical application

The results obtained have strong implications for police and court practice regarding the action of identification of persons by the witnesses conducted by police in pre-investigation proceedings and investigation and its probative value in the criminal proceedings.

Acknowledgments

The authors would like to express their gratitude to the reviewers for their comments that greatly improved the manuscript.

Graphical abstract

Figure 1

Download as PPT slide
Download as image