For weeks, William Celli posted threatening comments against the Richmond Muslim community on his Facebook page. He pledged his allegiance to Donald Trump and posted a photo of a possible homemade pipe bomb. At one point, he even stood outside the local Richmond mosque and shouted, “I’m going to kill you all!” as congregants were leaving their Friday afternoon prayers, according to the Council on American-Islamic Relations, a national nonprofit that tracks anti-Muslim hate.
The outspoken Donald Trump supporter was arrested in Richmond on Dec. 20, 2015 after threatening Muslims at the local mosque.
But in the California Department of Justice’s data on hate crimes in the state, no incident matching this description exists, and this case isn’t the only one missing from the data, experts say.
For 2015, the state reported 40 anti-Muslim hate crimes, based on police reports, not convictions. But the omission of Celli’s case is worrisome, said Hamza Mehter, the imam at the Islamic Society of West Contra Costa County, which was the target of Celli’s threats.
“If it’s really the case that things are not being reported as they are, then there’s something wrong,” Mehter said.
Celli took a plea deal, admitting to a charge of attempting to violate the free exercise of civil rights but a charge of making criminal threats with a hate crime enhancement was dropped. He received three years probation and a 90-day county jail sentence.
“Many of our congregants wished that it had been prosecuted to a greater extent,” Mehter said. “But I feel like we have moved on from it. Hopefully Mr. Celli has reformed.”
Lt. Felix Tan of the Richmond Police said that a department computer system crash in November 2015 may be the reason Celli’s case is missing from the data.
“The crash caused a lot of problems, and some of the data was never fully recovered,” Tan said.
Difficulty with data storage and transmission from local departments to the Department of Justice is only one of the reasons why anti-Muslim hate crimes might not make it into the state’s annual report.
Linguistic barriers, trust barriers and training barriers keep hate crimes from being reported and recorded properly, said Brian Levin, a former New York Police Department officer who now serves as a criminal justice professor at California State University San Bernardino, where he is also the director of the Center for the Study of Hate & Extremism.
“The problem is many victims don’t report. And sometimes when they do, the crimes are not adequately analyzed to be properly recorded by law enforcement,” Levin said. “A lot of police aren’t as well-trained as we would like.”
In response to a request for all anti-Muslim motivated crimes dating back to 2011, the San Francisco Police Department returned four cases. Two cases in 2011, one case in 2014 and one in 2015.
The San Jose Police Department also returned four cases in response to the same request. All four occurred in 2016, meaning that for the previous five years, no anti-Muslim crimes had been recorded in the city at all.
“It is very difficult to determine just how many anti-Muslim incidents are occurring in San Jose. SJPD can only provide data about incidents that are reported,” Sgt. Enrique Garcia of the San Jose Police Department, said in an email.
Assigning bias motivation is the most difficult part of investigating hate crimes, Garcia said.
“In order to be called a hate crime, the investigation has to clearly show that the suspect’s bias was the motivation for the crime…,” Garcia said in the email. “In most cases, the bias can’t be proven.”
“The determination of hate crime is sometimes not as easy as people expect it to be,” said Ameena Jandali of Islamic Networks Group, a nonprofit focused on countering bigotry and protecting religious freedom. “If law enforcement doesn’t immediately come out and call something a hate crime, it can get lost in the shuffle.”
The Department of Justice defines a hate crime as “a criminal offense committed against a person, property, or society that is motivated, in whole or in part, by the offender’s bias against a race, religion, disability, sexual orientation, or ethnicity/national origin.”
Offensive acts that can fall under the definition of a hate crime include “verbal slurs,” “annoying telephone/fax,” “graffiti” and “damage to vehicle,” among other offenses.
While the state reported just 40 of these offenses in California in 2015, the most recent year available, data from the Council on American-Islamic Relations (CAIR) showed 126 anti-Muslim hate incidents.
CAIR data showed that in 2015, there were 30 cases of anti-Muslim motivated assault and vandalism, 96 cases of hate mail or other communication and 60 other hate incidents in California.
A similar pattern was evident in 2014. That year in California, CAIR reported 14 cases of anti-Muslim assault and vandalism, 60 cases of hate mail/telephone call or other form of communication, and 24 other hate incidents, totaling 98.
That’s compared to 18 cases reported by the California DOJ in 2014.
The disparity in these numbers can be attributed to the fact that CAIR data is based on self-reporting by individuals rather than relying on the hate crime designation by police. In addition, CAIR uses the term ‘hate incidents’ to include reports of people being yelled at or targeted in a way that is discriminatory but does not go so far as to break the law.
But even though CAIR’s data includes more cases, it doesn’t claim to capture the full scope of the problem.
“I worry that both our numbers and the DOJ numbers underrepresent the severity of the issue,” said Zahra Billoo, executive director of CAIR’s San Francisco Bay Area chapter.
Underreporting: A culture of silence
Nonprofit organizations that track Islamophobia, such as CAIR and the Southern Poverty Law Center, claim there has been a noticeable increase in the number of anti-Muslim incident reports received since the recent presidential election.
Billoo said that in addition to official complaints, she’s also heard multiple stories of victimization from community members who did not report their experiences to police.
“I will see stories on Facebook that never make it into an official report, or I will hear from community members who will say: many months ago, this thing happened, and I didn’t know who to call,” Billoo said.
“Other examples that we get are ‘I didn’t know help was available. I was worried about making it worse. I didn’t know it was that bad, and I know you guys are busy so I didn’t want to bother you,’” Billoo said. “And the list goes on.”
Jandali, of the Islamic Networks Group, estimates that for every report that is made, 10 more go unreported.
“In recent years, both the rise in ISIS and terrorist attacks and extreme political rhetoric has really made Islamophobia mainstream,” said Jandali, who speculates that the anti-Muslim rhetoric of Donald Trump’s campaign and presidency has contributed to the recent spike in anti-Muslim hate crimes.
The impact that these acts of hate have on the community is hard to quantify, Jandali said.
“Hate crimes don’t just impact that one person. They amplify the fear and anxiety that the whole group is experiencing, causing collective victimization.”
The FBI and California Department of Justice will not release data for 2016 until the middle of this year. Jaimie Tackett, of the California Department of Justice, said the data goes through multiple levels of manual and electronic review before it can be made public.
“As a result of these processes, information that is initially entered — i.e., individual data points — often changes, and when information is changed it is overridden such that the DOJ does not retain prior drafts of that information,” Tackett said in an email.
Because of the limits of the Department of Justice’s data collection, other organizations have tried to gather data on anti-Muslim hate crimes. But often, these collections are flawed as well.
Georgetown University conducted a study that counted anti-Muslim hate incidents from January 2015 to March 2016 and became one of the most widely cited research reports on the subject of Islamophobia during the recent presidential election cycle.
But the Georgetown data “did not meet the reasonable accepted standards within the criminological community,” even though its numbers were reported by multiple media outlets, said Levin, with the Center for the Study of Hate & Extremism. “They conflated homicides that had Muslim victims, which had nothing to do with anti-Muslim hate crime, and then said there were twelve Muslims murdered in anti-Muslim attacks.”
For example, the study counted the following: “On Dec. 7, in Miami-Dade, Fla., someone shot a Muslim store clerk.” But according to media reports, the shooting was part of an attempted robbery, and police said there was not enough evidence to support any other motive.
The Georgetown study also tracked “Islamophobic political rhetoric by each presidential candidate” in relation to these crimes.
“People have a thing that they want to drop everything in Trump’s lap, when they should only drop some of it,” Levin said.
Engy Abdelkader, Georgetown faculty member and the study’s lead author, said in an email: “We’re not law enforcement officials, and our job was not to conduct hate crime investigations into each and every act or threat of violence we came across in media reports etc. Rather, the report provides data and analysis about the Muslim-American experience in contemporary America.”
With this broader scope, the Georgetown report did include the case of Celli in Richmond.
The news organization, ProPublica, spearheaded another initiative to establish a national database for reporting hate crimes as well as hate incidents. The project, called “Documenting Hate,” allows victims to upload information about their own experiences.
ProPublica is still in the data collection and verification phase, said Rachel Glickhouse, the partner manager of “Documenting Hate,” but the organization hopes to make the database available for use by journalists, researchers and civil rights organizations later this year.