This week, the Intercept released Part 1 of its interview with Kevin Urick, the prosecutor at Adnan’s 1999 and 2000 trials. Part 2 of that interview had been scheduled to be released on Thursday, but due to what appear to be some editorial issues over at the Intercept, it has yet to be published. Since I am tired of waiting for it to come out, however, I am going to go ahead and address Urick’s claims in the first part of the interview now.
Urick describes Adnan’s case “as pretty much a run-of-the-mill domestic violence murder.” When asked about whether he harbored any doubts as to the outcome of the case, he told the Intercept,
Urick: No. The reason is: once you understood the cellphone records–that killed any alibi defense that Syed had. I think when you take that in conjunction with Jay’s testimony, it became a very strong case.
Right away, we have a problem. “Once you understand the cellphone records,” Urick says, Adnan’s guilt becomes clear — but Urick himself never understood the cellphone records he used to convict Adnan of first-degree murder. His belief in Adnan’s guilt is based upon a false understanding (or, perhaps, a false professed understanding) of what the cellphone records shows in this case. Although he made numerous factual errors throughout Adnan’s first trial, perhaps the three of the most significant were: (1) his claim that Jay’s statements matched the cellphone records before the call logs and location data were shown to him; (2) his claim that Adnan “checked his voicemail” at 5:14 pm; and (3) his claim that the 7:09 and 7:16 pm calls are proof that the phone was in Leakin Park.
Let’s start with the corroboration issue. Here is what Urick said in the Intercept interview:
Urick: The problem was that the cellphone records corroborated so much of Jay’s testimony. He said, ‘We were in this place,’ and it checked out with the cellphone records. And he said that in the police interviews prior to obtaining the cellphone evidence.
That is simply a false representation of fact. As I discussed extensively in my previous post on the evolution of Jay’s perjured testimony, Jay’s statements to the police did not match the cellphone records before he was shown them. Flat out did not match, full stop. What Urick said in the Intercept interview is not true, as anyone can see for themselves, if they go through Jay’s first and second police statements. Particularly in the first interview, before Jay was ever shown either the call log or the tower data, when not a single call “check[s] out” with his statement.
Okay, fine, that is a slight overstatement. Jay did manage to get one call right — he (correctly) said Adnan had called him in the morning “about 10:45.” That’s the only one he gets, though, and that one has nothing whatsoever to do with the murder. No other calls he describe correlate with the cellphone records.
(In fact, even at trial, it did not get much better, as Jay’s testimony failed to match the vast majority of the phone calls made that day. But that is for another post.)
Urick: Now the thing about the cellphone records [is that they] corroborate Jay, his statements that he got a call around 2:45 p.m. or around that time from Adnan to come pick him up. And the cellphone records show that there was an incoming call around that time. So there’s corroboration of Jay’s statements to the police and the cell records.
It is not surprising Urick got this one wrong, because he got it wrong over and over again at Adnan’s trial too. Because do you know who never ever once made a statement that he got a call from Adnan to pick him up at 2:45 pm? Jay. Jay said over and over again that Adnan’s call for a pick up came at 3:40 pm, and he has never claimed, in a single recorded statement, that the “come-and-get-me” call from Adnan came at 2:45 (or 2:36 pm) that day. Yes, there was an incoming call to Adnan’s cellphone at 2:36 pm, but there was never any evidence introduced at Adnan’s first trial to support the claim that it was actually the “come-and-get-me” call — and there quite a lot of evidence to contradict it. The 2:36 pm call was and always has been a figment of Urick’s imagination.
Urick: So, many of the material facts were corroborated through the cellphone records including being in Leakin Park.
The Intercept: Syed never testified. What would you have asked him if he had?
Urick: I would’ve gone through the cellphone records. You called this person at this time. Jay talked to this person at this time. And my very last question would be: What is your explanation for why you either received or made a call from Leakin Park the evening that Hae Min Lee disappeared, the very park that her body was found in five weeks later? . . . A lot of what [Jay] said was corroborated by the cellphone evidence, including that the two of them were at Leakin Park.
First of all, it needs to be said: objection as to the form, assumes facts not in evidence. What a wasted “last question” that would have been, because you cannot ask a witness a question phrased that way. Urick is trying to make a rhetorical point, but that approach would never fly in the courtroom.
In any event, however, Urick is simply wrong. His claim about the cellphone records establishing that the phone was in Leakin Park at 7:09 and 7:16 pm is not based on any sort of established fact. It is simply conjecture — that could be what the cellphone data is showing, but there is no reason to believe that is more likely than alternative explanations.
I will admit that I too previously bought in to the story about the 7:09 and 7:16 pings definitively proving that the phone was in Leakin Park at around the time of those calls. It made sense, based on the data we had, and it seemed like Serial bought it too.
But once this claim is compared to the actual phone records, it falls apart. The “Leakin Park calls” that Urick’s case was so dependent upon do not even show by a preponderance of the evidence that the cellphone was in Leakin Park at 7:15 pm that night, let alone show it beyond a reasonable doubt.
Rabia was kind enough to indulge my curiosity about the cellphone data, and sent me a copy of the subpoenaed cellphone records from the police file. The records show that on February 18, 1999, the Baltimore Police Department subpoenaed Adnan’s cellphone records. On February 22, 1999, the cellphone records were faxed over to Detective Ritz by AT&T, along with a copy of the cell tower locations showing the addresses of each tower that Adnan’s cellphone had pinged that day.
And on the very first page of the fax from AT&T, the following message was displayed:
In case it is not legible, here’s what the second to last paragraph says:
“Outgoing calls only are reliable for location status. Any incoming calls will NOT be considered reliable information for location.”
Well. That’s something I sure wish they had thought to mention in the podcast. Or, you know, at Adnan’s trial.
Because you know those two “Leakin Park calls” that Urick keeps mentioning over and over again? Both of them are incoming calls. There are no outgoing calls that ping the Leakin Park tower/antenna. Which means that the phone calls that Urick’s entire case relied upon to show that the phone had been in Leakin Park were “NOT [ ] considered reliable information for location.”
As it turns out, though, it is not just the incoming calls we have to be skeptical of. The subpoenaed cell phone records demonstrate that the fact that a call pings L689B is not proof the cellphone was in Leakin Park at the time of the call, even for outgoing calls. Because here are a series of calls that took place two weeks after Hae’s murder:
The important calls here are shown at lines 29, 28, and 27. At 4:44 pm, an outgoing call is made that pings L689B — the Leakin Park tower. This is the same tower and antenna as the 7:09 and 7:16 pm calls on January 13th, which Urick claims as proof the phone was in Leakin Park. Then, a mere 74 seconds later, at 4:45 pm, another outgoing call is made — and it pings L653C.
As a refresher, to help show the relevance of this data, below is an “idealized” map of L689B’s range:
And here is the same with respect to L653C:
So these two towers are close to one another, yes. But the implication of a call pinging L689B vs. L653C would appear to be very different. L653C covers the Edmondson Avenue area, facing towards several areas where Jay and Adnan were throughout the day on January 13. It is the same tower/antenna that was pinged for by the 5:38 and 8:05 pm calls that day — two times when neither Urick nor Jay have ever claimed that the phone was in Leakin Park.
But as the calls from the cell records above demonstrate, it was entirely possible for two phone calls made one minute apart from one another — at 4:44 pm and 4:45 pm — to hit both the “Leakin Park tower” (L689B) and the Edmondson Avenue tower (L653C). Which means that any assumption that a call pinging L689B must have been in Leakin Park is based on a false premise.
Because even though the phone was in the exact same location at the time of both the 4:44 and 4:45 calls (or within 100 yards thereof), the location data provides a false location for one of the two calls. Which means we cannot assume a call that pinged L689B was in Leakin Park at the time it was made, as it is every bit as likely, based on the data we have, that a phone that pings L689B is sitting instead out on Edmondson Avenue, or Jenn’s house, or the McDonalds, or Neighbor Boy’s house, or Patrick’s house, or… you get the idea.
Part of the issue here is the way L689 is set up awkwardly compared to nearby towers. For L689B, the territory that it is closest to it and to no other tower is very small and constrained, covering a very narrow strip. But its area of overlapping range with neighboring towers is much larger than what is shaded in on the maps. So if the phone is at a location south of Leakin Park and north of Edmondson — or even to the west of both — it is in a territory that is well within the overlapping ranges of both L689 and L653.
So let’s recap:
- Incoming calls are not reliable indicators of location, according to AT&T.
- Calls that ping L689B (Leakin Park) are also capable of pinging L653C (Edmondson Avenue, Jenn’s house, etc.).
Which means Urick’s glib confidence that the 7:09 and 7:16 calls are proof that Adnan murdered Hae is entirely misplaced.
Now, moving on to the next problem with Urick’s prosecution of Adnan, take a look again at the cover page from AT&T — specifically at the third paragraph from the bottom:
When “SP” is noted in the “Type” column and then the “Dialed #” column shows “#” and the target phone number, for instance “#7182225555”, this is an incoming call that was not answered and then forwarded to voice mail. The preceeding row (which is an incoming call) will also indicate “CFO” in the “feature” column.
What this shows is that Urick discredited Adnan’s alibi based on a false representation to the jury. Because the 5:14 pm call was not Adnan checking voicemail, as Urick repeatedly asserted at trial — it was someone leaving a voicemail message. Urick’s claim that Adnan must have had his cellphone at 5:14 pm was therefore untrue and deeply prejudicial to Adnan.
That the 5:14 call was someone leaving a voicemail message is something Urick should have known. He had the cover page from AT&T; he knew that when the cell records showed two calls in a row, with the same time and duration, and the second call showed Adnan’s cell phone number with a pound sign in front (“#4433539023”), what was being shown was a voicemail being left on Adnan’s cell phone. And yet he told the jury that it was Adnan checking his messages.
Here are some examples to demonstrate that what the cell records show is indeed a voicemail being left by someone who called Adnan’s phone. On the subpoenaed cellphone records, unanswered incoming calls to voicemail are displayed just as the cover sheet describes, as two calls with the same time and duration, with the first call labeled as “incoming” and the second call labeled as “#4433539023,” which is Adnan’s cell phone number. It looks like this:
In contrast, calls made to check voicemail show up as just a single call to Adnan’s cellphone, without a pound sign in front, like this:
So why does all this matter? Because on January 13, 1999, someone called Adnan’s cellphone and left a voicemail message at 5:14 pm. It looks like this on the cell phone records:
Two calls, at same time, lasting the same duration, first call to “incoming,” second call to “#4433539023.” This is plainly someone leaving a voicemail message on Adnan’s cellphone, as the cover page from AT&T clearly explains.
At Adnan’s first trial, however, Urick informed a witness — Krista — precisely the opposite. He informed her that this showed Adnan checking his cellphone’s voicemail, and asked her to testify that Adnan was checking a voicemail left by her:
KU: Now look at the line immediately beneath that — line 18. And as you look at 18 and 19, you’ll see line 18 says #443-xxx-xxxx. Now if you look up at the top, you’ll see that that is the cellular phone number for Adnan Syed [indiscernible] and right beneath it, it says incoming call. If you go over to the time area they both occurred at 15:14:07 a[s] they both list 1:07. Now that is the means that AT&T uses to record someone checking their voicemail. . . .
Is that possible that when the Defendant is checking his voicemail at 5:14, that he’s checking the message that you left?
Krista: It could be possible. I don’t believe it is because usually I don’t arrive home from work till about 5:20 so it’s not likely that I would’ve been home on the phone at 5:14 — that early.
KU: If you check the entire evening you’ll see he never — there’s no other listing like that where the Defendant checks his voice mail. (12/13/99 Tr. 286-87.)
Urick is either a ridiculously sloppy attorney, or else a liar. It could go either way, but it is inexcusable regardless. He falsely represented to a witness what the phone records showed, in an effort to coax her testimony into fitting his narrative.
This was not an error Urick fixed in the second trial, either. Here is what he said in opening arguments, the second time around:
And you’ll see that there is a voice mail where the voice mail is checked at 5:14. Christy Myers will tell you that she called, left a message for the defendant to call her.
Well, the cell phone records indicate the voice mail was checked. And the very next call, which occurs at 5:38 is to Christy Myers. She’s not home at that time. Her voice — her answering machine comes on. You’ll see it’s a two second call, as soon as the answering machine comes on the person — defendant hangs up. (1/27/00 Tr. 108.)
Urick doubled down on his factual misrepresentation, wrongfully informing the jury that Adnan had been in possession of his phone at 5:14 pm, when in reality the cellphone records suggest just the opposite conclusion.
But when properly understood, in accordance with the clear directions from AT&T, the cellphone records support Adnan’s story, not Urick’s. Adnan says he was at track practice that day, and track practice ended at 5:30 pm. There is a 5:14 call to Adnan’s cellphone that goes unanswered (maybe from Krista, maybe from someone else), and at 5:38, there is a call (2 seconds, apparently unanswered) to Krista.
And what is this consistent with? It is consistent with a scenario in which Adnan did not have the cellphone at 5:14 pm — when the voicemail was left — but was picked up from track at 5:30 by Jay, and had the phone at 5:38 to make a call to one of his friends.
This is significant for another reason, as well. At trial, Jay testified that he received a call from Adnan asking to be picked up in “half an hour”:
Jay: Me and Mark went back downstairs. We played video games for like almost another hour. His sister came in and I got another phone call from [Adnan].
KU: What did he say at that time?
Jay: He said he was leaving school and wanted me — he wanted me to pick him up in about a half an hour. (12/14/99 Tr. 192.)
Jay claimed that this call came before the Best Buy trunk pop, but that does not fit the cell phone records at all. What it does fit, however, is a scenario in which the call Jay was waiting for was to pick up Adnan from track.
If track ended at 5:30 pm, as it usually did, then this would mean Adnan called Jay at 5:00 pm to tell him to pick him up “in about a half an hour,” when track ended. While there is no 5:00 pm call in the call records, there is a call at 4:58 incoming call that is otherwise unaccounted for. So this call would appear to be Adnan — at track, like he has said all along — calling Jay to tell him to get him in “half an hour.” At 5:30 pm, Jay picked Adnan up from track, and at 5:38, Adnan had his phone again, and made a call to Krista.
However, since Urick falsely claimed at trial that Adnan was ‘checking his voicemail’ at 5:14 pm, the jury would have reasonably believed that Adnan’s alibi was a lie, and that he was not at track until 5:30 pm that day. While it is possible that Urick made an honest mistake when he misinformed the jury about the significance of the 5:14 pm call, that does not make the error any less significant. The prosecution discredited a defendant’s alibi based on information that it knew or should have known to be false, and an error of that magnitude cannot simply be shrugged away.
p.s. If you like Serial and/or non-proliferation law, check out Arms Control Wonk’s latest podcast episode, in which they invited me on for a guest appearance to discuss Serial and the relevance of geospatial data.