💾 Archived View for gem.sdf.org › jquah › news › news2021.gmi captured on 2022-01-08 at 14:03:28. Gemini links have been rewritten to link to archived content

View Raw

More Information

-=-=-=-=-=-=-

Cherrywood Ln and Long Branch

Playing the roles of courtroom reporter and armchair psychologist

Friday 10 January 2020

I was pretty naive to have expected a final decision within an hour of starting the proceedings. Today's

sentencing hearing spanned the entire morning and afternoon, only wrapping up at 5:40pm. Judge Hazel gave everyone a two-hour lunch break while he weighed the intricate arguments presented in the morning session, so the actual duration of the courtroom proceedings was more like four hours. In those four hours we heard a substantial amount of witness testimony and disputes over the text of the law.

The first part of the sentencing hearing brought back the three ATF special agents we heard from during trial. Chris Szokolczai, one of the executors of the search warrant, testified that the flashlight parts and cylindrical plugs in one of the photos could easily have been made into a silencer for one of the guns. He also walked us through the browser history on the Samsung tablet, which showed visits to online merchants selling silencer parts, and search queries on how to make a silencer, going back to April 2018. Szokalczai then walked us through the photos of drug manufacturing paraphernalia recovered during the search warrant, including Pyrex cookware with cocaine residue, vials full of PCP, a milk frother, and cutting agents typically mixed with these controlled substances. On cross-examination, Mr. Faison got Szokalczai to confirm that the smaller quantity of PCP found in Faison's bedroom was packaged similarly to the larger stash of PCP found in a garbage bag in Larry Newman, Jr.'s bedroom.

The next fact witness was special agent Matthew Leonard, who repeated his conclusions about the two firearms recovered from Mr. Faison's person. The AR-type pistol had no manufacturer markings, leading Leonard to conclude it was custom-assembled at someone's home. That much was not new to anyone who paid attention during trial. What intrigued me about Leonard's testimony was the new information on the Kimber pistol, which we thought was purchased by Larry Newman Jr. (or someone else without a felony conviction) on 9/1/2018 at Fred's Outdoor Sports, based on the photo of the receipt. Leonard reported having traced the Kimber pistol to a batch that got stolen during a 2008 heist in Virginia, so either the Fred's receipt was not for that weapon, or somehow the stolen Kimber gun made its way onto the shelves of Fred's store.

The third fact witness was special agent Jamie Rohsner, another executor of the search warrant on Mr. Faison's bedroom. Rohsner walked us through the chemical analyses performed by the PG County Forensics Lab, which positively identified the residue on the glass saucepan as cocaine, and the contents of the vials as PCP. On cross-examination, Mr. Faison got Rohsner to say she had no knowledge of any fingerprint analysis conducted on the Pyrex cookware or other drug manufacturing paraphernalia.

Then we heard Mr. Patel from the federal defender's office, whose testimony concerned the pre-sentencing report's recommendation that the offense level should be enhanced based on the connection of the firearm possession to possible violations of the controlled substances act (which violations were not part of the original indictment). Mr. Patel argued that the commentary attached to a prior 4th-circuit ruling should not be applied in this case, because it goes beyond a strict reading of the relevant statute in USC Section 21. Prosecutor Wright was allowed to present a short counterargument before the lunch break, and the debate continued after the court reconvened at 3:00pm. Judge Hazel used the lunch break to read through the prior court decisions and anticipate how the legal arguments would play out, so he had already made up his mind about which precedent to follow when interpreting the sentencing guidelines for Mr. Faison. Judge Hazel ended up siding with the federal defender's office; he ruled that the [1]court precedent in Dozier need not be followed in this case.

The next part of the hearing gave both sides a chance to make closing arguments. Prosecutor Wright went first, addressing a question posed by Judge Hazel and then summarizing the government's position that a ten-year sentence is appropriate. She cited four possible grounds for enhancement of the offense level, and received pushback from Judge Hazel on about two and a half of them.

1. The Kimber pistol turned out to be stolen. The government's position on "strict liability" puts the onus on the possessor to verify the origin of any firearm in possession, which would enhance the offense level by two points. Judge Hazel pushed back by saying that any previously convicted felon now charged with possession would obviously have obtained the firearm in the illegal market, so this provision of the law doubly punishes the repeat offender for the murky provenance of most weapons in that market. The purpose of enhancements, in Judge Hazel's view, is to distinguish between defendants charged with the same crime, ranking them according to how much of a danger they pose to society. The provenance of the Kimber pistol does not help make these distinctions, so Judge Hazel applied a downward variance to this enhancement.

2. The possession of the firearm was connected to another intended felony: either conspiracy to manufacture and distribute controlled substances (based on the newly presented evidence of cocaine and PCP in the basement bedrooms), or first-degree assault (based on the jail call where Mr. Faison said "I was gonna KILL HIM"). The prosecution wanted to apply a four-point enhancement to the offense level on these grounds. Judge Hazel remarked that the government had a stronger case for the connection to first-degree assault, but the evidence for a connection to drug distribution was pretty flimsy. Indeed, the weapons found on Mr. Faison's person were retrieved from the trailer parked outside, but the shotgun in the closet next to Larry Newman, Jr.'s bedroom would, on its own, provide enough protection for the garbage bag full of PCP vials.

3. The defendant perjured himself by fabricating a story on the witness stand and coercing his family members to corroborate his made-up story. For this "obstruction of justice", the prosecution sought a two-point enhancement of the offense level. Judge Hazel pushed back strongly on this point, saying that the nature of any trial is for two competing stories to be told around the same set of facts, and the choice of a defendant to serve as his own advocate in the courtroom should not lower the bar for an accusation of perjury.

4. Violating the terms and conditions of supervised release, by attempting to build firearms and silencers within a year of being released from prison, would merit another enhancement of the offense level, according to the government. This point was not seriously contested by Judge Hazel, but Mr. Faison objected that other people in the house had access to the Samsung tablet and to his bedroom, which leaves unproven his personal involvement in such activities (even at the lower standard "preponderance of evidence"). Prosecutor Wright responded by saying that Larry Newman, Jr. would not have left such incriminating supplies in plain view in Mr. Faison's bedroom, unless he regarded Mr. Faison as a willing accomplice to these activities.

In his closing argument, Mr. Faison repeated his claim that the government is good at mischaracterizing snippets of jail calls. He asked the court to consider the totality of the circumstances, and the character witnesses he would have called if their health permitted the stress of a courtroom cross-examination. He sought sympathy for his difficulty of transitioning back to life as a free man, after spending the majority of his 47 years of life in prison and learning maladaptive responses to perceived aggressions. He objected to being held responsible for the poor choices of his younger brother, a grown man in his own right who leads a life independently of Mr. Faison and can be expected to spill over some elements of that life into the adjoining bedroom (if Mr. Faison is frequently away from that house with the mother of Mr. Faison's new baby). Lastly, Mr. Faison asked the court to consider the victim's conduct (the pursuit through neighborhood roads by the driver of the burgundy vehicle), without which Mr. Faison would have continued his claimed 20-year abstention from handling firearms.

After taking into consideration all these adjustments to the offense level, rejecting some and allowing others to stand, Judge Hazel calculated the recommended range of prison time as anywhere from 77 to 83 months. Taking into account the violation of supervised release, the final sentence turned out to be 91 months, to be served in Fort Dix. A more complete explanation of Judge Hazel's reasoning will be entered into the court record shortly, along with the letter of support that the court received from one of us jurors on behalf of Mr. Faison, and the intricate untangling of prior court decisions filed by the federal defender's office.

Friday 17 January 2020

Currently Reading:

"What's the Matter With Kansas?" by Thomas Frank

The puzzle that Frank attempts to unravel in this book is how the formerly socialist-leaning midwestern states abandoned their nineteenth-century allegiances and became die-hard supporters of the conservative party in recent decades (Frank was writing in the aftermath of the Bush v. Gore election). His explanation hinges on a shared sense of victimhood in the cultural sphere, which is a mantle donned by high-earning business executives just as easily as by their struggling employees. All references to economic disparities and class distinctions are suppressed in this world view. Shifting social norms, such as the prohibition on "Merry Christmas" greetings in a federal workplace, or more welcoming stances on reproductive rights and same-sex relationships, are the sole basis for "red state" voters feeling oppressed.

The Republican party's successful appeals to this sense of victimhood, by themselves, only partially explain the high degree of polarization observed today. The Democratic party's strategy of "triangulation" during the Clinton administration combined a more business-friendly economic policy with firmly liberal stances on social issues, in an attempt to secure the high-dollar campaign contributions of the socially-liberal "creative classes" centered in New York and Los Angeles. By forgetting to actively court votes in "red states" based on real differences in economic policy, however minor, the Democratic party conceded a substantial amount of territory to Republicans, who never stopped fomenting the backlash in their red state campaigns. The Democratic party's pandering to the "creative classes" receives further elaboration in Frank's more recent book Listen, Liberal, while the need to campaign in red states is the thesis of at least two chapters in Bernie Sanders' essay collection Where We Go From Here.

The crux of Thomas Frank's thesis, a shared sense of victimhood and a backlash against the imposition of foreign values, has echoes in the larger grievance culture of the United States. From our peer groups and our elected officials, we learn the habit of shifting blame to external agents, never letting responsibility fall on the cherished members of our in-group. When the US president blames China's currency manipulation for the dearth of decent-paying US manufacturing jobs; when college administrators blame lack of county funding for not delivering the negotiated faculty salary increases; when academic deans justify the sweeping cancelation of classes by citing the registration choices of prospective students; in all these instances I hear echoes of the backlash mentality that underpins Thomas Frank's thesis.

As a teacher I regularly encounter the "attribution error" among my students. Their performance on an assessment is the result of multiple factors: what background knowledge they brought to the class, how I designed each lesson, how they absorbed those lessons, how they reflected on those lessons in the time leading up to the assessment, what specific questions appear on the assessment itself, and the random circumstances of their lives around the time of the assessment. When the contributions from intrinsic factors are downplayed, and the contributions from external factors embellished, a backlash response to perceived aggressions is often the result. Even in the absence of a reinforcing media environment (as documented by Thomas Frank for the conservatives who see themselves as cultural victims), a self-serving story repeated over many years to an uncritical audience of close friends can make the correct attribution difficult to accept.

The simplest template for an outcome with multiple causes is "outcome = baseline + random noise", as Daniel Kahneman reminds us in his chapter on regression to the mean in Thinking, Fast and Slow. The second term L of this decomposition (L for luck) contains all the "unknown unknowns" that are not incorporated into a predictive model. The baseline term is often broken down further into a term K we can control ("knowns"), and a term U we cannot control (AKA "known unknowns"). When the variance from the U and L terms dwarfs the variance from the K term, then the backlash mentality is on firmer empirical ground. But if the "known" term provides most of the variance, the backlash mentality would appear to be unjustified.

Retrospective analysis of the factors that contributed to an outcome starts by trying to establish the baseline prediction, using either the historical record of similar scenarios or some other source of unbiased data. When students in my college classes compare their past performance to how they did on the latest exam, a substantial difference between the historical average and the recent performance spurs them to update their variance estimates for U, L, and K, with most of the additional variance tacked on to U and L. Rarely do they use the feedback of an exam to update their estimate for K, even after they have been informed that studying strategies that worked in high school will need to be revised for college-level classes.

In this week alone, I found three more examples of attributions that distribute the variability among the three components K, U, and L in strikingly different ways.

1. As a job applicant with hundreds of rejections and non-responses from employers, I observe in myself the thought pattern: "The reviewers of my resume didn't make enough of an effort to see how my experience aligns with the job description." This instance of the attribution error deliberately plays down the factors under my control, e.g., how I tailor the resume to each job, and the personal choices that brought me to the current position of teaching part-time with ever-diminishing prospects for in-house promotion.

2. At this week's Democratic primary debate, Senator Warren read into the electoral losses of her male opponents a lesson about those candidates' national prospects, as if the failures to win statewide races in fiercely conservative or independent Indiana or Vermont tell us more about the candidate than about the priorities of the voters in those states. Meanwhile a win in deep-blue Massachusetts is brandished by Senator Warren as her own personal accomplishment, with no mention of the conditions on the ground that all-but-assured her victory.

3. Falling asleep tonight to the Amanda Palmer album There Will Be No Intermission, I noticed another example of allocating blame in the song Judy Blume. Here Amanda Palmer takes refuge in a universe of fictional adolescents, whose preoccupations are on those topics that adults deem off-limits. The outcome of not feeling like a valued member of adult society is again decomposed into K+U+L, but with K contributing an insignificant part of the total variance by virtue of the sentiments ubiquitous among Judy Blume's characters. (This assumption is analogous to the weather modeling practice of filling the main diagonal of the covariance matrix with the lab-established discrepancies among copies a given instrument, rather than estimates of the variances out in the field.) Taking a longitudinal view from the perspective of someone whose adolescence is long past, the reader would recalculate K with the additional knowledge of life's trajectory. Indeed, the mismatch between adolescent thoughts and adult expectations is entirely within the error bars of a baseline prediction from this vantage point, leaving almost no unexplained variance for the U and L terms to pick up. Until that vantage point is reached, though, the typical middle-schooler looks to the sentiments expressed in the song to justify the self-preserving tactic of minimizing contributions from U and L.

I can't remember the details of seventh grade
All I remember is lying and being afraid
But I don't forget Katherine and Michael going all the way
Steph on the scale in the bathroom alone that day
Karen pretending to puke so her dad would stay
Margaret arguing with God while she masturbated
All of them mixed up in my head like a love letter
All of them saying "Amanda, you know better,
You are not to blame
The world's a frightening place
So go on and think how you want
You will not be alone with your thoughts
Well you will, but you won't in a way
'Cause a girl thought it too in a book that the library bought."

Monday 29 March 2021

The shift to remote teaching has put on hold some curriculum innovations proposed within the MC math department in recent years. Adapting our existing pedagogy to a remote environment is challenging enough. Nobody was eager to compound the challenge with a rapid adoption of partially-vetted OERs.

Eschewing the clean UI of a publisher's companion website (and the college-licensed suite of Office365 apps) in favor of open educational resources like RStudio and Jupyter Notebooks is probably too much of a lift for the students in Weather and Climate who just want a general education credit. Therefore I only demonstrated the familiar Excel interface in a live class session, leaving the Python/pandas tutorial as an optional reading assignment. (The college offers an Office365 subscription to all enrolled students, so requiring Excel is not an additional expense that would violate the terms of a "Z course".)

My own assumptions this semester about the typical student's ability to scale the Excel learning curve were mostly accurate. I did have a few students who manually constructed a time series using a rather coarse temporal resolution, but at least they chose the correct type of graph when going through the "Insert > Chart" dialog.

The proliferation of different Excel interfaces is more bothersome to those of us with muscle memory from the era when software releases were more infrequent, than it is to the students who are accustomed to smart phone apps getting trivial facelifts every month. The greater challenge that comes from using Excel is its poorly-documented timeseries functionality.

One of my students in a previous semester ended up employing substring extraction functions (LEFT, RIGHT, or MID) on the datetime cell contents in order to get the hour and minute of each set of observations. These commands can easily generate garbage if Excel has its own ideas about the format of the cells being referenced. In my demo this semester, I steered my class toward alternatives that would force Excel to treat the cells as integers rather than strings. It occurred to me while doing this demo that students at this level might never have seen practical uses for the FLOOR and MOD functions, and so would not have thought to use them in a formula.

In contrast, the pandas module in Python has a more self-explanatory set of commands for these kind of conversions, all fully-documented in the help files for `pd.to_datetime`. The `to_datetime` command can be invoked to tell Python exactly how the timestamps are formatted. You can also add an offset so that timestamps are placed in the correct month and year, rather than in January 1900. These features make Python/pandas almost as user-friendly as the weatherData package in R (for the purpose of creating meteograms).

Sunday 11 April 2021

I'm heading out to get my first jab of the Pfizer vaccine. Wish me luck!

Checklist:

Monday 24 May 2021

One annoying aspect about Zoom meetings under dwm is the popup notification that takes over half my workspace just to tell me who the new host is. If the main room of the Zoom meeting is just a reception area anyway (with the real work happening in breakout rooms), I'd much prefer to accomplish some other business in a separate workspace while letting the Zoom audio keep me apprised of any walk-in appointments. If at any time I cared to know who the host was, I could just return to the Zoom workspace and scan the participants list. A popup alert stealing focus from my text editor (demoting it from full-screen to the narrow stacking area) is an unwanted interruption.

What does this UI annoyance mean when interpreted in light of Rabbi Hillel's maxim, "what is hateful to you, do not do unto others"? Was I right to refrain from weekly email reminders to my math students last semester, on the theory that they might find it hateful to see their inboxes cluttered by the same alerts they could have generated themselves by reading the syllabus on day one and making the appropriate cronjob? At least one respondent to the course evaluations didn't think so. Such respondents apparently believe that a reminder from me might have saved them from missing the quiz that tanked their average.

From the virtual educator summit last summer, a tip that continues to echo in my mind is to be "professor of the obvious." It's not safe to assume that our students have the same mental bandwidth that we do. Abundance of free time might be the new reality for professors (thanks to the elimination of a lengthy commute), but the students' workdays are filled with many more responsibilities besides school (e.g., searching for jobs that no longer exist). Incessant emails from professors are not likely to be interpreted as nagging, but rather as the only means of cutting through the noise and refocusing their attention on the obligations they signed up for.

The apparent conflict between Rabbi Hillel's maxim and the advice to be "professor of the obvious" can be resolved by carefully unpacking the pronoun in "hateful to you". The "you" in this maxim need not be your present self, but might be regarded more usefully as your remembered self (from a time when your habits of self-regulation were still at a novice level).

Cognitive scientist Daniel Willingham points out the qualitative differences between the thinking patterns of experts and the thinking patterns of novices. Experts quickly identify the deep structure of a problem in their domains of expertise, while novices pay more attention to the surface features. This change in thinking emerges only after sufficient "time on task": at least 10,000 hours of practice, to cite a popular figure. Unpromising branches of the decision tree are quickly lopped off by the expert's trained mind, while the novice entertains all possible routes to a solution and is paralyzed by the plethora of choices. This strain on the novice's mental bandwidth makes it more likely that outreach from a concerned professor will be welcomed with gratitude (as a relief from indecision, not an intrusion on a smoothly-running workflow).

In later years the former novice might forget how much mental bandwidth used to be consumed by such plodding navigation of the decision tree. The intrusion into a smoothly-running workflow by an undesired popup alert would then merit the description "hateful", whereas the present-day inbox intrusion by a professor's reminder email would merit the opposite description.

In Norton Juster's classic children's book The Phantom Tollbooth, the character Alec Bings comes from a race of people whose vantage point remains unchanged throughout life. Their legs and torso grow down to eventually touch the ground, while their eyes stay always at the same height. Alec Bings viewed humans (represented by Milo in the story) as encumbered with a significant handicap, in having to adjust the cues from their visual field according to the size of their bodies. A counterpart to this fantasy might replace the visual field with moral and aesthetic judgments, and for a race with constant vantage point in these dimensions, the phrase "hateful to you" would not require such convoluted parsing.

Back to my gemlog