Edward Martin was convicted in 2001 of assault with intent to kidnap, assault with intent to rape, and assault and battery. These charges arose out of a 1994 attack on a fifteen year old girl while she walked home from a beach in Yarmouth. On appeal, the defendant raises several grounds for reversal. He argues that the motion judge should have excluded the complainant’s one-on-one identification of the defendant as overly suggestive and the subsequent in-court identification as tainted; that the trial judge, who was not the same judge as the motion judge, should have sanitized a mug shot of the defendant before allowing the jury to see it, and should have given a limiting instruction to the jury regarding that mug shot; that the Commonwealth failed as matter of law to prove the intent elements of the crimes of which the defendant was convicted; and that the defendant was denied his right to a speedy trial. We conclude that the identification procedure was overly suggestive and therefore reverse the convictions. Although the case can be resolved on this issue alone, we address the defendant’s other arguments to avoid needless appeals in the event of a retrial. See Fabre v. Walton, 436 Mass. 517, 522 n.8 (2002); Commonwealth v. Mottola, 10 Mass. App. Ct. 775, 781-782 (1980) (value of avoiding future appeals and preserving judicial resources).
1. Facts. The complainant was attacked on July 20, 1994, while she walked from the beach to her grandparents’ West Yarmouth home on a path in a wooded area next to a cranberry bog. She was carrying a small purse. According to her testimony, she heard someone running up behind her. Before she could turn around, she was grabbed from behind, forcibly turned around, and thrown to the ground on her back, whereupon her assailant began punching her in the face while covering her mouth with one hand, telling her to “shut up.” Soon after the attack began, other people in the area who heard the complainant’s screams came to the scene. As these people drew nearer, her assailant picked her up and attempted to drag her off the path into the bog, telling her to come with him. Unsuccessful in *589carrying the complainant away, the assailant fled into the woods as bystanders arrived. The entire attack lasted a few minutes.
After the attack, the complainant reported the incident to the police and gave a description of her attacker, from which a composite photograph was made. The complainant described her attacker as a white man between thirty to forty years old, tall, thin, with thinning brown and gray hair, clean shaven and with tan skin. She did not mention any distinguishing marks on her assailant’s face. The complainant also told police her attacker had been wearing a yellow canvas hat, a blue shirt, shorts, and sneakers with white socks. She examined a number of mug shots at the Yarmouth police department that day, including a photo of the defendant from two years earlier, but she did not identify the defendant as her attacker at that time.
For the next four days, the complainant, her father, and local police searched for the perpetrator. Yarmouth police officers drove around the area with the complainant in a police car throughout each day, stopping when they saw an individual who seemed to fit her description of her assailant and asking the complainant whether he was the attacker. The complainant testified that during this period she looked at hundreds of people. On at least three occasions prior to the complainant’s positive identification of the defendant, the police actually stopped individuals so the complainant could look at them. Other officers also stopped additional individuals in the area and took their photographs, which the complainant subsequently reviewed during the course of the days after the attack. The complainant’s father searched the area independently on his bicycle.
On the fourth day after the attack, the complainant’s father saw the defendant and thought he might be the attacker. The complainant’s father called Yarmouth police, and several officers arrived and detained the defendant in a beach parking lot while another police car brought the complainant to the scene. When the complainant arrived, the defendant was standing next to several uniformed police officers and the complainant’s father was standing nearby. It was the first and only occasion on which the complainant was asked to look at a possible suspect with her father present. She looked at the defendant from inside the police car, asked the police that he be brought closer to her, and *590then stated that he was the man who had attacked her, saying that she recognized a mark on the defendant’s forehead. The defendant was then arrested. Later that day the complainant returned to the police station, where police showed her a photo array containing eight photos, one of which was a picture of the defendant taken shortly after she identified him earlier that day. The complainant selected the defendant’s photograph from the array. At trial, the complainant was asked whether “the individual that attacked [her] that [she] identified at Veteran’s Beach” was present in the courtroom. She identified the defendant.
Police searching the area near where the attack took place discovered what appeared to be a campsite. At this site, they found various documents, including job applications, bearing the defendant’s name. They also found a series of notes written on scraps of paper, which were similar to notes found in the defendant’s possession when he was arrested. These notes were disjointed and sometimes incoherent, suffused with derogatory descriptions of women, and often took the form of lists of brief phrases or statements. Several notes suggested a plan to attack and rape a woman (“ice the bitch”; “crack the head and skull”; “I want to, I want to, sex with you in five minutes”), including notes on a woman’s activities (“two o’clock, blond bitch from Yarmouth, other direction . . . 12:00 to 1:30. Brown hair, tied back. Went to lunch. 1:35, returned. There at 2:30. Left at 3:30. Security guard?”), and notes about what to do after an attack (“take time, look around, one good around, take one good look around, cut bitch or slice, all less, specify, speak the English, get the fuck moving”).
Police were not able to determine when the notes were written, but the notes appeared weathered at the time of the defendant’s arrest. A handwriting expert called by the Commonwealth testified that the notes appeared to be in the same handwriting as job applications found at the campsite, which were filled out with the defendant’s name.
2. Identification of defendant. After the motion judge decided the question of identification in favor of the Commonwealth, pursuant to a motion to suppress and a motion in limine, the *591defendant failed to renew his objection at trial. There is some question as to what standard of review should apply in this setting. A motion in limine, which seeks a pretrial evidentiary ruling, is not sufficient to preserve appellate rights, absent a further objection at trial. Commonwealth v. Whelton, 428 Mass. 24, 25 (1998). A motion to suppress, however, is reviewable even if no further objection is made at trial. Id. at 25-26. In the instant case, the defendant’s argument was advanced by means of a motion to suppress, which rested on constitutional principles. There was no need for him to renew his objection after the pretrial motion hearing and the issue is properly before us in this appeal. Ibid.
The defendant argues that the one-on-one identification procedure was impermissibly suggestive and therefore violative of his due process rights guaranteed under the Federal and State Constitutions. One-on-one identification procedures, or showups as they are commonly known, raise serious questions of both constitutionality and reliability. See Stovall v. Denno, 388 U.S. 293, 302 (1967); Commonwealth v. Botelho, 369 Mass. 860, 865-866 (1976). For this reason, they are generally disfavored. See Commonwealth v. Moffett, 383 Mass. 201, 213, (1981); Commonwealth v. Thompson, 427 Mass. 729, 735, cert. denied, 525 U.S. 1008 (1998). Nevertheless, courts over the years have carved out a significant class of cases in which showups, although suggestive, are not impermissibly suggestive. See Commonwealth v. Austin, 421 Mass. 357, 361 (1995). We can discern a certain set of judicial rules from these cases concerning the admissibility of showups: the defendant has the burden of proving, by a preponderance of the evidence, that the contested identification was unnecessarily suggestive. Commonwealth v. Santos, 402 Mass. 775, 781 (1988). Once the defendant has met this burden, the Commonwealth is barred from introducing the identification, notwithstanding other indicia of reliability. Commonwealth v. Johnson, 420 Mass. 458, 465 (1995). Showups are acceptable when conducted in the immediate aftermath of a crime or in exigent circumstances, Commonwealth v. Barnett, 371 Mass. 87, 92 (1976), cert. denied, 429 U.S. 1049 (1977), but haste or pressing exigency need not be present, Commonwealth v. Connolly, 356 Mass. 617, 623-624, cert. denied, *592400 U.S. 843 (1970). Showups are evaluated in light of their general purpose, which is to clear innocents of suspicion quickly while keeping police from pursuing fruitless avenues of investigation. Commonwealth v. Barnett, 371 Mass. at 92.
In practice, this body of case law has led to the admission of more showup identifications than it has excluded. Rather than focus on what special circumstances allow an otherwise disfavored practice, the courts have tended to build their analysis around the presence or absence of “special elements of unfairness.” Commonwealth v. Leaster, 395 Mass. 96, 103 (1985). This is the approach that the Commonwealth urges us to take in this case. However, the showup in this case falls completely outside the realm of those approved in the past. There are no exigent circumstances present: the police could just as easily have taken a photograph of the defendant and had the complainant view that photograph in an array, as they did with other individuals on several occasions before the defendant was identified. The showup did not take place within the immediate aftermath of the crime: it occurred four days later. Nor can we say that by conducting the showup, the police were seeking to clear an innocent person of suspicion or make sure their investigation was moving in the right direction. Although different cases have indicated that none of these special factors, standing alone, is determinative or required, the absence of any of them militates against the admissibility of the showup. We are aware of no case in which a showup identification was admitted on facts that, like these, suggest no special need. Although it has created certain exceptions, the Supreme Judicial Court has not, as yet, indicated that one-on-one identification procedures have ceased to be “disfavored.”
In addition to a lack of special factors to commend it, the showup in this case contains certain elements of unfairness. First and foremost, the complainant’s father was present and the complainant testified that she realized before identifying the defendant that her father was the person who spotted the defendant and contacted the police. The motion judge found, and the Commonwealth urges us here to consider, that the complainant, despite her age, was especially self-confident, *593poised, and immune to any coercive influence from her father’s presence. See Commonwealth v. Austin, 421 Mass. at 362 (“The police could take into consideration that the percipient witnesses appeared unusually reliable and nonsuggestible [when deciding whether a showup was warranted]”). In essence, this is the approach to showup identifications that the United States Supreme Court approved in Manson v. Brathwaite, 432 U.S. 98, 112-114 (1977), whereby a court can allow an impermissibly suggestive identification procedure if it determines that the identification was nevertheless reliable. The Supreme Judicial Court expressly rejected this approach in Commonwealth v. Johnson, 420 Mass. at 465-472, holding that art. 12 of the Massachusetts Declaration of Rights requires a per se exclusion of unnecessarily suggestive identification procedures without recourse to a secondary analysis of witness certainty or reliability.1
The holding in Johnson requires us to conclude that the presence of the complainant’s father presents a real risk of suggestiveness. “Considering the complexity of the human mind and the subtle effects of suggestive procedures upon it, a determination that an identification was unaffected by such procedures must itself be open to serious question.”2 Id. at 469, quoting from State v. Leclair, 118 N.H. 214, 218 (1978).
*594The passage of time between the crime and the showup is another factor that cuts against the admission of the complainant’s identification of the defendant. Four days cannot be characterized as within “the immediate aftermath of [the] crime.” Commonwealth v. Barnett, 371 Mass. at 92. The general view that showups should occur only immediately after the crime is to minimize the witness’s suggestibility. “To have the witness view the suspect while his recollection or mental image of the offender is still fresh, before other images crowd in or his attempts to verbalize his impressions can themselves distort the original picture, provides the witness with good opportunity for an accurate identification.” Ibid. Four days, during which the complainant was asked to view several men on the street and a series of mug shots, may have detracted from her ability accurately to identify her attacker. Compare Commonwealth v. Levasseur, 32 Mass. App. Ct. 629, 636-637 (1992), cert. denied, 506 U.S. 1053 (1993).
We conclude that it was error to admit the showup identification. From the jury’s questions during deliberations, we can infer that the identity of the complainant’s assailant was the crucial question, and one over which the jury struggled. Absent the admission of the showup identification (and taking into account our conclusion, infra, on the lack of an independent basis for the complainant’s in-court identification of the defendant), it is difficult to say whether the defendant could have been convicted. The defendant’s right to due process was violated, and the convictions must be reversed.
3. Independent basis for in-court identification. The judge also ruled that the complainant’s in-court identification of the defendant had a sufficiently independent and untainted source, based on her view of her assailant during the attack. We conclude that the judge’s finding was erroneous as matter of law.
The judge appears to have applied the multifactor analysis set forth in Commonwealth v. Botelho,3 369 Mass. at 869, to determine whether the later in-court identification had a source independent of the previous suggestive identification. Botelho, *595which adopts the test established in United States v. Wade, 388 U.S. 218, 241 (1967), requires that the Commonwealth prove4 by clear and convincing evidence that any in-court identification is not tainted by an intervening illegal identification procedure. Commonwealth v. Botelho, 369 Mass. at 866-868. The test lists six factors: “(1) The extent of the witness’ opportunity to observe the defendant at the time of the crime; prior errors, if any, (2) in description, (3) in identifying another person or (4) in failing to identify the defendant; (5) the receipt of other suggestions, and (6) the lapse of time between the crime and the identification.” Id. at 869, citing United States v. Wade, 388 U.S. at 241.
The judge characterized the complainant’s description of her attacker as possessing “remarkable accuracy.” The judge, however, apparently failed to consider the fact that the complainant did not mention a distinguishing mark on the defendant’s forehead, but claimed to remember the mark when she identified the defendant at the showup. Since the complainant’s description of her attacker was somewhat general, her omission of the unique and distinguishing feature of a mark on her attacker’s forehead calls into question the “remarkable accuracy” of that description.5
Significantly, the judge did not consider the showup itself as a suggestive factor that could taint the subsequent, in-court identification. “Because of the dangers inherent in subsequent suggestions, however, the judge should . . . hear evidence upon, and take into account in [her] conclusion, what influence, if any, the illegal pre-trial confrontation between the witness and the defendant has had upon the witness’s prior recollection of the defendant from the time of the crime” (citations omitted). Commonwealth v. Mendes, 361 Mass. 507, 511-512 (1972). Here, the judge considered only statements of the police to the complainant and the presence of police and the complainant’s *596father at the showup. The judge did not consider whether the confrontation itself affected the complainant’s ability to identify the defendant at trial based solely on having seen him prior to the confrontation. Nor did the judge consider the photo array shown to the complainant at the Yarmouth police department later on the day of the showup, from which she selected a photo of the defendant taken at the time of the showup, in which the defendant was in the same location and wearing the same clothes as he was at the showup. We think the subsequent photo identification can only have served to cement the complainant’s belief, right or wrong, that the defendant was her assailant. See United States v. Crews, 445 U.S. 463, 472-473 (1980) (“This is not to say that the intervening photographic and lineup identifications — both of which are conceded to be suppressible ... — could not under some circumstances affect the reliability of the in-court identification and render it inadmissible as well. Indeed, given the vagaries of human memory and the inherent suggestibility of many identification procedures, just the opposite may be true”).
While no one of the Botelho factors is dispositive, we think the motion judge’s failure to consider the effect of the showup and subsequent photo array on an in-court identification, especially in light of the lapse of more than seven years between the crime and the in-court identification, misapplies Botelho and ignores its purpose. This is especially so in a case in which the jury appears to have struggled with the identification element. See Commonwealth v. Wynter, 55 Mass. App. Ct. 337, 344 (2002) (error of law as to suppression of identification evidence was grounds for reversal where jury was initially deadlocked).
Any one of these considerations, alone, might not be grounds for reversal. Taken together, however, they do not support the judge’s ruling. The judge did not properly consider two crucial factors in the Botelho test: (1) she ignored a significant error in the complainant’s initial description of her attacker, which the complainant subsequently changed; and (2) she failed to consider the highly suggestive nature of the showup itself as well as the suggestive photo identification that occurred within hours of the showup. This provides an insufficient foundation for a conviction in which the jury struggled with the identifica*597tian element. “The inquiry cannot be merely whether there was enough to support the result, apart from the phase affected by the error. It is rather, even so, whether the error itself had substantial influence. If so, or if one is left in grave doubt, the conviction cannot stand.” Commonwealth v. Peruzzi, 15 Mass. App. Ct. 437, 445-446 (1983), quoting from Kotteakos v. United States, 328 U.S. 750, 765 (1946).
4. The mug shot. After she was attacked but before the showup, the complainant viewed photo arrays at the Yarmouth police station, one of which included mug shots of the defendant, taken two years earlier. The complainant failed to identify the defendant from his mug shots. At trial, the mug shots, a front view and a profile view printed on a single piece of paper, were submitted to the jury. Printed on the back of the mug shots were the words “YARMOUTH POLICE DEPARTMENT,” along with information about the defendant, including name, age, alias, physically descriptive factors (height, weight, etc.), and a date, “7/21/92.” The defendant argues that the mug shots should have been sanitized to exclude information about their provenance before being submitted to the jury.
The admission of a defendant’s mug shot in evidence is “laden . . . with potential for characterizing the defendant as a careerist in crime . . . .” Commonwealth v. Smith, 29 Mass. App. Ct. 449, 451 (1990). Such admission is subject to a three-part test: (1) the prosecution must show some need to introduce the mug shot; (2) the mug shot should, as much as possible, not indicate a prior record; and (3) the mug shot should not call attention to its origin or the implications of that origin. Id.; United States v. Fosher, 568 F.2d 207, 214 (1st Cir. 1978).
Here, the Commonwealth needed to show that the complainant failed to select the photos because the defendant’s appearance had changed. However, the information on the back of the mug shots served only to remind the jury that the Yarmouth police had the defendant’s picture in their possession prior to the attack the complainant. See Commonwealth v. Rodriguez, 378 Mass. 296, 309 (1979). See also Commonwealth v. Lockley, 381 Mass. 156, 165-166 (1980) (dates prior to incident should have been removed from mug shot).
The Commonwealth suggests that the jury already knew that *598the police had the defendant’s mug shots, so there was no need to sanitize them. This argument misses the mark. The purpose of the rule regarding mug shots is not to keep the jury from knowing where the photos came from (since most jurors probably know what a police mug shot looks like), but to minimize the prejudicial impact of the photographs — “[t]he manner of introduction at trial must be such that it does not draw particular attention to the source or implications of the photographs” (emphasis added). United States v. Fosher, 568 F.2d at 214. Here, the mug shots could easily have been sanitized simply by covering the reverse side of the sheet on which they were presented.
Witness identification appears to have been the most difficult issue for the jury in this case. “On the Commonwealth’s evidence, any reasonable doubt by the jury about the ability of [witnesses] to identify the defendant as one of their assailants might have been resolved upon the basis of his prior involvement with the . . . police.” Commonwealth v. Gee, 36 Mass. App. Ct. 154, 161 (1994). In the event of a retrial, the mug shots should have any reference to their provenance removed. We leave to the discretion of the trial judge whether to sever the front and side views in the mug shots and whether to give a special instruction to the jury regarding the mug shots. See ibid.
5. Speedy trial. The defendant argues that he was denied a speedy trial under Mass.R.Crim.P. 36, as amended, 422 Mass. 1503 (1996). We review rulings of speedy trial motions de nova. For although we give deference to the findings made by the judge below, we may make our own determinations. Commonwealth v. Vasquez, 55 Mass. App. Ct. 523, 526 (2002). While it is indisputable that the defendant suffered an inordinately long period of pretrial detention, the majority of that period is not countable under rule 36, either because the defendant was responsible for delays, because he consented to continuances, or because he had been found incompetent to stand trial. After careful review of the record, we conclude that the analysis of the motion judge on this question was not error.
6. Sufficiency of the evidence. The defendant argues that as matter of law, there was insufficient evidence to prove an intent to rape or an intent to kidnap. We disagree. In reviewing the *599denial of a motion for required finding of not guilty, we look at the evidence in the light most favorable to the Commonwealth and ask whether a rational jury could have reached a guilty verdict. Commonwealth v. Ruci, 409 Mass. 94, 96 (1991).
In arguing the absence of proof of intent to rape, the defendant makes much of the fact that there was no apparent sexual touching or language used during the attack on the complainant. Were there no other evidence, this argument might be persuasive. However, the violent sexual content of the notes found in the defendant’s possession and at his campsite is very specific as to his intent, including some details of how the crime would have been accomplished. While the Commonwealth never established when, precisely, these notes were written, it introduced testimony that they appeared old four days after the attack. There was also testimony that the complainant was carrying a small purse at the time of the attack, but no testimony or other indication that it was stolen. See Commonwealth v. Mahar, 21 Mass. App. Ct. 307, 314 (1985) (intent to rape could be inferred from absence of evidence of intent to commit larceny or other crime). Taken together, these facts, although circumstantial, form sufficient basis for a rational jury to find intent to rape. See Commonwealth v. Rojas, 388 Mass. 626, 629 (1983) (circumstantial evidence sufficient to establish guilt).
The defendant’s argument with regard to intent to kidnap is equally unavailing. There was testimony that the assailant told the complainant to “shut up,” attempted to drag her off the path toward his campsite, and told her to come with him. These facts are sufficient for a rational jury to find an intent to kidnap. Accordingly, the ruling of the trial judge denying the defendant’s motion for required findings of not guilty was not error.
For the reasons cited, supra, we conclude that the defendant’s right to due process was violated, requiring reversal of the convictions.
Judgments reversed.
Verdicts set aside.
The Commonwealth cites authority for the proposition that we should defer to the factual findings of the motion judge. The question, though, is not one of factual accuracy but of application of the law.
The Commonwealth also argues that the repetition of the showup procedure over the days preceding the complainant’s identification of the defendant reduces the suggestiveness, and that the complainant’s failure to choose one of the other suspects shown to her proves her reliability. Like the question of the complainant’s relative maturity and poise, this line of reasoning is speculafive and unprovable. While the Commonwealth’s argument is logical and compelling, we can just as easily imagine a scenario in which a fifteen year old girl feels significant pressure simply because she has spent four days with police officers without identifying a suspect. Similarly, the fact that the complainant asked that the defendant be brought closer to her before she positively identified him need not be seen as an indication of certainty and lack of suggestiveness. Just as we cannot know the true nature of the effect of her father’s presence, we cannot know whether subtle cues from the police offleers driving the complainant around Yarmouth, combined with the presence of her father, pressured her to pick the defendant. This is precisely the sort of unsupported speculation in which we ought not and need not engage.
While not citing Botelho, the judge’s analysis tracks its analysis closely, and the Commonwealth relies on it on appeal.
Once the defendant has proved by a preponderance of the evidence that the initial identification was overly suggestive, the burden shifts to the Commonwealth to prove the admissibility of an in-court identification. See Commonwealth v. Botelho, 369 Mass at 866-868; Commonwealth v. Johnson, 420 Mass. at 463-464.
The Supreme Judicial Court, in a case presenting similar facts, has held that such a disparity is significant when determining whether an identification is independent or tainted. Commonwealth v. Johnson, 420 Mass. at 471-472.