The story of SignSynth

Leave a comment
Genny, the avatar for SignSynth, poses with hands in front of her in the G handshape (index finger extended, other fingers bent). She has a blocky, cartoonish appearance, made mostly of cylinders and rectangular solids, with a sphere for her head.

The beginning

Every morning in the fall of 1997 I would wake up at 6AM and immediately jump out of bed.  I went straight to my Fujitsu laptop in the home office, leaving my then-girlfriend sleeping in the bedroom.  I didn’t have anywhere to be; nobody was paying me, and this wasn’t for a class.  I was working on a new project: a sign-language synthesis prototype.

This project was the culmination of several threads in my life.  I had been inspired to go back to grad school by spending time talking about language with my girlfriend, a linguist who had just earned her PhD.  For the past two years I had been making my living in information technology, working my way up from Microsoft Office trainer to LAN operations tech.  I had also been taking night classes at the American Sign Language Institute and participating in discussions on SLLING-L, a sign linguistics email list, and the SignWriting List.

I had chosen the linguistics doctoral program at the University of New Mexico because I knew it was strong in sign linguistics, and I had just arrived in August.  I was taking a course in psycholinguistics with the sign linguist Jill Morford, and a course in ASL.

Several people had asked me, “So you’re interested in language, and you’re interested in computers.  Is there a way you could combine these interests?”  I replied, “Well, there’s speech synthesis,” but that technology was already pretty well established.  I realized that there wasn’t much in the area of technology for sign languages.  What would sign synthesis look like?

When I arrived in Albuquerque I mentioned this idea to Sean Burke, a linguist and programmer I had made friends with on an earlier visit.  Sean suggested using Virtual Reality Modeling Language, a standard for describing three-dimensional objects and their movements.  People could install a plugin in their web browsers, and point it to a VRML (since renamed Web3D) site, and the plugin would display the 3D animation.

I followed Sean’s tip and discovered that there was a standard for specifying and animating humanoids in VRML, and a way to control the plugin through Javascript.  I created a basic animation of a sign using Javascript, but after numerous frustrations I concluded that it was best to assemble the VRML through a custom Perl templating system controlled by CGI forms.

Getting the word out

Once my system was working to the point that it could create intelligible signs, I showed it to the faculty and students who were studying sign languages.  The strongest interest came from Jill Morford, my psycholinguistics professor, who recognized that I had created a sign-language analog of the early speech synthesizers that were used at Haskins Labs to demonstrate the categorical nature of speech perception and its acquisition in language learners.

Jill realized that my system, which I had dubbed SignSynth, could be used to test whether perception of sign languages is similarly categorical, and how it is acquired.  The following year she obtained a grant from the National Institutes of Health to study the question, and included some money to support me part time and get an office that I shared with a couple other students.

Around that time I had a meeting with my Committee on Studies.  They liked my work, but told me that simply creating a sign synthesis system wasn’t theoretical enough for a Linguistics dissertation.  If I used it, I would have to use it to demonstrate an answer to some theoretical question.

Doubts

That year I also took a course in literacy education called Teaching Reading to the ESL Student.  The professor, Leila Flores-Dueñas, was happy to support me applying these lessons to Deaf students, but throughout the course she stressed the importance of grounding our teaching in the priorities of our students, and orienting our research to the goals of the populations we were studying.  She pointed out further that we should be working to help lift up people who were oppressed or disadvantaged, so when we are working with people in those situations we have a particular obligation to fit our work to their priorities.

I realized that the same principle applied to developing applications: an app that touches disadvantaged people should help them to fulfill their goals.  This brought back to mind the fact that I had never seen a Deaf person ask for a sign synthesis or natural language processing application.

I actually hadn’t talked much with Deaf people about language technology.  This was in part because there is a general suspicion of hearing people in Deaf communities, particularly of hearing people who come bearing language technology.  The suspicion is well earned; check out what Deaf people have to say about Alexander Graham Bell.

I had difficulty overcoming this suspicion because I had only been studying American Sign Language for a few years.  I could communicate with Deaf people, but only if they were patient, and many of them had no reason to be patient with some random hearing person.  I realized that I didn’t necessarily have any technology that could help them accomplish their goals, and if i did, they weren’t necessarily going to try it.

Challenges

That year I ran into two major challenges in sign synthesis, both relating to the placement of hands in space.  The higher-level challenge relates to signs that are called depicting or iconic verbs, or classifier predicates.  In these signs, the relative location, orientation and movement of the signer’s hands are used to refer to a similar spatial relationship or movement.  In a classic example, an ASL signer can depict the movement of a car by making the “3” vehicle classifier shape with one hand and moving that hand in a scaled-down version of the car’s movements.

Iconic verbs are very difficult to specify in a user interface, because the level of detail of the location and movement is only constrained by the signer’s control of their hands, and the audience’s visual perception abilities.  Specifying, representing and transmitting such signs is, at a minimum, a vastly different task than that for lexical signs, where there is a relatively small set of locations and movements.

The lower-level challenge involved figuring out how to bend the shoulder, elbow and wrist on a humanoid figure so that the figure’s hand winds up in a particular place, facing in a particular direction.  This is a well-known problem, called inverse kinematics, that we humans solve intuitively every time we make a gesture or pick up an object.

There are computer libraries for inverse kinematics, and I tried to connect my animation code to those libraries, but they were written in C++, not Perl, and after several months I still hadn’t figured it out.

I described these challenges of inverse kinematics and representing classifier predicates to my committee, but they did not consider them to be theoretical enough for a dissertation.  I was able to create the videos for Jill Morford’s categorical perception experiments, and she suggested that I could study categorical perception for my dissertation.  In retrospect I should maybe have taken her up on it, but instead I decided to study a cosmopolitan, largely privileged community where the speakers were all long dead: the history of Parisian French through theater.

Legacy

My papers on SignSynth are routinely cited as one of the pioneering works in text-to-sign, mostly by developers who didn’t have a Professor Flores-Dueñas to teach them the importance of serving the language community, and who don’t follow my lead in separating synthesis from translation.

The paper that Jill Morford wrote with me and my fellow students about categorical perception has also had lasting influence; we failed to find a strong effect of categorical perception to match those found in speech, and the implications of that are still being discussed.

I never wanted to commercialize SignSynth, knowing that average incomes for Deaf people tend to be significantly lower than the general population, so I made it publicly available as a web application.  The original application relied on third-party browser plugins to display the VRML, but over time, web browser makers dropped support for those plugins.

In 2022 I rewrote SignSynth from scratch in Javascript, using a relatively new 3D graphics library, Three.js, and made the code available on GitHub.  I updated the interface to follow “character pickers” like Richard Ishida’s IPA Character Picker, replacing the old interface that was heavy on drop-down menus.  I also created a standalone character picker for the Stokoe notation that it used for the original text.

What SignSynth meant to me

Thinking back to 1997, what got me so excited, what inspired me to move across the country and start a doctoral program with no promise of funding, living off of part time jobs and student loans, and spend so many unpaid hours every day, was the idea that I was creating something new, something that could be useful to people.

When I was young, I admired inventors, both real-life inventors like Thomas Edison and fictional ones like Professor Bullfinch from the Danny Dunn books.  My father, my stepfather and my mother’s boyfriends in between were all tinkerers, and they made useful things.  I didn’t have money to buy a lot of tools and equipment, or space to keep it in, but I did have access to computers, and the skills to program them.

Looking back, there was some ego in this.  Why was it important for me to make these things, and not someone else?  Why not be satisfied fixing the Novell servers for Chase Manhattan, or even sending faxes for John Hancock?

To be fair to myself, I have never been a competitive inventor.  While I was working on SignSynth I met several people who were working on similar systems, and I always tried to be positive, supportive and cooperative.  I let the quality of my work speak for itself, and trusted people to judge its value for themselves.

I’ve never felt like I was the best inventor, programmer or researcher in the world, but even in 1997 I had a fair amount of experience and education in research and technology.  I felt like those skills were being wasted when I was taking telephone messages at John Hancock, and even when I was reinstalling network drivers.  Of course, there are a lot of highly skilled people out there, maybe more than there’s a need for.

After putting SignSynth on the shelf and focusing on French negation for my dissertation, I didn’t feel the same level of excitement; the theoretical advancements that my committee demanded felt small by comparison, although I’m still proud of them.

I have felt a bit more excitement for my recent work for the New School.  It’s a more simple project, just retrieving class listings or final grades from the student information system and presenting it in a table, but it helps save time and effort for students and faculty.  Less exciting, but still satisfying.

Cropped cover from the Who's album Who's Next

Everybody fronts

Leave a comment

I could write – and have written – the past twenty years as a string of successes for my linguistics career: I have presented my work at many conferences on linguistics, digital humanities and literature, in several countries. My pioneering work in computational sign linguistics continues to be cited on a regular basis. In 2008 I was hired to teach linguistics at Saint John’s University, where I developed an introductory linguistics curriculum. In 2009 I received my doctorate in linguistics. In 2012 I was hired to work on natural language processing at New York University. In 2016 I released the first segment of the Digital Parisian Stage. In 2019 I published my first book, Building a Representative Theater Corpus: A Broader View of Nineteenth-Century French. In 2021 I released LanguageLab, an app for audio mimicry exercises.

But of course, on the internet people put up a front. I could just as easily write a history of failure; in fact I hear it in my head on a regular basis: In 2008 my dissertation advisor refused to write letters of recommendation for me; my entire committee has showed very little interest in my work and have barely cited me. In 2013 the grant I was hired to work on at NYU ended, and with a brief exception I haven’t been hired to do any further language modeling research. I’ve applied to dozens of full-time teaching jobs, but have not even been invited to an interview. I couldn’t make a living as an adjunct instructor in New York, so in 2015 I stopped teaching and in 2016 took a job as a computer programmer, with no linguistics focus. As far as I know, hardly anyone has used my sign language software, my French theater corpus, my book or LanguageLab.

The truth is somewhere in between glorious success and abject failure. A lot of my difficulty has been due to circumstances beyond my control. The professor who I thought would be my dissertation chair was not actively publishing by the time I started my studies, retired right before I proposed my thesis, and sustained severe brain damage in a bicycle crash shortly after I defended my dissertation. I graduated into an incredibly difficult job market, due to cuts to liberal arts education and general research funding and a glut of underemployed people with humanities PhDs.

I have also made principled choices that have affected my career. The professor who I cited extensively in my dissertation has never publicly acknowledged my work, I suspect because I declined to participate in her patronage system. I intentionally chose not to work on projects that I suspected were extractive, exploitative or overhyped. In order to take care of my young child and prioritize my wife’s career, I moved thousands of miles away from my university, suspended and delayed my doctorate, and limited my job search to opportunities near her job. In the past ten years I’ve similarly limited my job options in order to care for my mother, who is severely disabled with Parkinson’s disease.

For decades I’ve been a prolific poster on the social media of the times – email lists and USENET in the nineties, bulletin boards and blogs in the early 2000s, Twitter and Tumblr in the teens. I’ve mostly projected the first image of myself – the authoritative expert – while fearing that people would find out about my failures.

That confident persona often came off as patronizing or condescending, but I felt it was necessary to assert my authority and stake out my territory in order to be taken seriously, to advance my career. I tried to do it with as much compassion and generosity as I could.

I’m still proud of my knowledge and my accomplishments, and I want to be recognized for them. But now that my academic career appears to be stalled at best, there doesn’t seem to be much point in being brash the way I was.

The thing is, we’re all putting up a front. Academic competition seems to demand it. It’s very hard to do it without being cruel or indifferent to others.

Recognizing that everyone fronts means that it’s important to have compassion for the ways that other people front. Judging others for being cruel or indifferent is reasonable. Judging them for fronting at all is not.

Flu. What is Thisby? A wandring knight? Quin. It is the Lady, that Pyramus must loue. Fl. Nay faith: let not me play a wom?: I haue a beard c?-(ming. Quin. Thats all one: you shall play it in a Maske: and you may speake as small as you will. Bott. And I may hide my face, let me play Thisby to: Ile speake in a monstrous little voice; Thisne, Thisne, ah Py-, ramus my louer deare, thy Thysby deare, & Lady deare. Qu. No, no: you must play Pyramus: & Flute, you Thysby.

The History of English through SparkNotes

1 Comment

Language change has been the focus of my research for over twenty years now, so when I taught second semester linguistics at Saint John’s University, I was very much looking forward to teaching a unit focused on change.  I had been working to replace constructed examples with real data, so I took a tip from my natural language processing colleague Dr. Wei Xu and turned to SparkNotes.

I first encountered SparkNotes when I was teaching French Language and Culture, and I assigned all of my students to write a book report on a work of French literature, or a book about French language or culture.  I don’t remember the details, but at times I had reason to suspect that one or another of my students was copying summary or commentary information about their chosen book from SparkNotes rather than writing their own.

When I was in high school, my classmates would make use of similar information for their book reports.  The rule was that you could consult the Cliffs Notes for help understanding the text, but you weren’t allowed to simply copy the Cliffs Notes.

Modern Text FLUTE Who?s Thisbe? A knight on a quest? QUINCE Thisbe is the lady Pyramus is in love with. FLUTE No, come on, don?t make me play a woman. I?m growing a beard. QUINCE That doesn?t matter. You?ll wear a mask, and you can make your voice as high as you want to. BOTTOM In that case, if I can wear a mask, let me play Thisbe too! I?ll be Pyramus first: ?Thisne, Thisne!??And then in falsetto: ?Ah, Pyramus, my dear lover! I?m your dear Thisbe, your dear lady!? QUINCE No, no. Bottom, you?re Pyramus.?And Flute, you?re Thisbe.

When I discovered SparkNotes I noticed that for some older authors – Shakespeare, of course, but even Dickens – they not only offered summaries and commentary, but translations of the text into contemporary English.  It was this feature I drew on for the unit on language change.

While I was developing and teaching this second semester intro linguistics course at Saint John’s, I was also working as a linguistic annotator for an information extraction project in the NYU Computer Science Department.  I met a doctoral student, Wei Xu, who was studying a number of interesting corpora, including Twitter, hip-hop and SparkNotes. Wei graduated in 2014, and is now Assistant Professor of Computer Science and Engineering at Ohio State.

Wei had realized that the modern translations on SparkNotes and eNotes, combined with the original Shakespearean text, formed a parallel corpus, a collection of texts in one language variety that are paired with translations in another language variety.  Parallel corpora, like the Canadian Hansard Corpus of French and English parliamentary debates, are used in translation studies, including for training machine translation software. Wei used the SparkNotes/eNotes parallel Shakespeare corpus to generate Shakespearean-style paraphrases of contemporary movie lines, among other things.

When it came time to teach the unit on language change at Saint John’s, I found a few small exercises that asked students to compare older literary excerpts with modern translations.  Given the constraints of this being one unit in a survey course, it made sense to focus on the language of instruction, English. The Language Files had one such exercise featuring a short Chaucer passage.  In general, when working with corpora I prefer to look at larger segments, ideally an entire text but at minimum a full page.

I realized that I could cover all the major areas of language change – phonological, morphological, syntactic, semantic and pragmatic – with these texts.  Linguists have been able to identify phonological changes from changes in spelling, for example that Chaucer’s spelling of “when” as “whan” indicates that we typically put our tongues in a higher place in our mouths when pronouncing the vowel of that word than people did in the fourteenth century.

When teaching Shakespeare to college students it is common to use texts with standardized spelling, but we now have access to scans of Shakespeare’s work as it was first published in his lifetime or shortly after his death, with the spellings chosen by those printers.  This spelling modernization is even practiced with some nineteenth century authors, and similarly we have access to the first editions of most words through digitization projects like Google Books.

With this in mind, I created exercises to explore language change.  For a second semester intro course the students learned a lot from a simple scavenger hunt: compare a passage from the SparkNotes translation of Shakespeare with the Quarto, find five differences, and specify whether they are phonological, morphological, syntactic, semantic or pragmatic.  In more advanced courses stufents could compare differences more systematically.

This comparison is the kind of thing that we always do when we read an old text: compare older spellings and wordings with the forms we would expect from a more modern text.  Wei Xu showed us that the translations and spelling changes in SparkNotes and eNotes can be used for a more explicit comparison, because they are written down based on the translators’ and editors’ understanding of what modern students will find difficult to read.

As I have detailed in my forthcoming book, Building a Representative Theater Corpus, we must be careful not to generalize universal statements, including statements about prevalence, to the language as a whole.  This is especially problematic when we are looking at authors who appealed to elite audiences, but it applies to Shakespeare and Dickens as well.  Existential observations, such as that Shakespeare used bare not (“let me not”) in one instance where SparkNotes used do-support (“don’t let me”) are much safer.

My students seemed to learn a lot from this technique.  I hope some of you find it useful in your classrooms!

Remembering Alan Hudson

1 Comment

On Saturday I found out that Alan Hudson died. Alan was my doctoral advisor at the University of New Mexico until his retirement in 2005, and a source of support after that.

I first met Alan when I visited the UNM Linguistics Department in 1997. Alan welcomed me into his office with a broad smile, and asked, “So Angus, have you made up your mind about whether you want to come here?”

“Well…” I said. I had been accepted into the PhD program, but had just come from a very discouraging encounter with another professor, and was ready to give up and go home. Before I could continue, Alan said, “Is there anything I can say to convince you?” I replied, “Well, I guess you just did.”

Alan was not a big name in linguistics; he never published a book. I regularly had to tell people that my advisor was not Dick Hudson. But Alan had a profound insight about the sociology of language that changed my career trajectory and my thinking about language and social justice.

In a seminar on Societal Bilingualism the next year, Alan led us through the case studies laid out by Joshua Fishman, his own advisor, in his book Reversing Language Shift. Fishman’s book is of interest to anyone concerned with language “death” (a problematic metaphor unless the language users themselves are being killed). As a Dubliner who had become fluent in Irish through compulsory government schooling, Alan cared deeply about his national language, but he did not have high hopes for it recovering its status as the primary language of Ireland.

Fishman argues that we can prevent large numbers of people abandoning a language by establishing “diglossia’ – arrangements where language H is used for some functions and language L is used for others. Charles Ferguson had shown in 1949 that diglossic arrangements tend to be stable over time. Fishman believed that if language users can establish similar functional separations, they can stop language shift.

Drawing in part on his own research in Ireland and Switzerland, Alan observed that the cases Fishman categorized as diglossia did not fit with Ferguson?s examples. The key factor in Ferguson’s cases was that there were no children in the speech community who are native speakers of H: no child speakers of High German in Switzerland, no child speakers of Metropolitan French in Haiti, etc. In Ireland, by contrast, there are millions of English-speaking children, and in the Netherlands Frisian-speaking children go to school with Dutch-speaking peers.

The result of this contact is that most of these children eventually shift to the higher-prestige, better-paying language, and will not pass their native languages on to their children. There are only two ways to stop it: reverse the power dynamic (as happened in Finland when Russia conquered it from Sweden, I discovered in a term paper that semester) or isolate the children (as Kamal Sridhar observed in her Thanjavur Marathi community).

This was an important insight, with major implications for linguistics. None of us in the course were interested in segregating language groups from each other, and as linguists we were not positioned to shift the socioeconomic power differentials between groups. If the prescription for reversing language shift can be captured in a single sentence, that leaves no ongoing role for linguists.

Since then I have not been terribly surprised that Alan’s insight has not been enthusiastically embraced by other linguists. As Upton Sinclair said, “It is difficult to get a man to understand something, when his salary depends upon his not understanding it!” Alan published two articles describing his definition of diglossia, but framed it in theoretical terms, downplaying the implications for efforts at language maintenance and revitalization.

Alan Hudson supervised my studies and my comprehensive exams, but retired before I was ready to begin my dissertation. He continued to provide valuable advice, and attended my dissertation defense. He will be remembered as an insightful linguist and a supportive teacher.

Data science and data technology

Leave a comment

The big buzz over the past few years has been Data Science. Corporations are opening Data Science departments and staffing them with PhDs, and universities have started Data Science programs to sell credentials for these jobs. As a linguist I?m particularly interested in this new field, because it includes research practices that I?ve been using for years, like corpus linguistics and natural language processing.

As a scientist I?m a bit skeptical of this field, because frankly I don?t see much science. Sure, the practitioners have labs and cool gadgets. But I rarely see anyone asking hard questions, doing careful observations, creating theories, formulating hypotheses, testing the hypotheses and examining the results.

The lack of careful observation and skeptical questioning is what really bothers me, because that?s what?s at the core of science. Don?t get me wrong: there are plenty of people in Data Science doing both. But these practices should permeate a field with this name, and they don?t.

If there?s so little science, why do we call it ?science?? A glance through some of the uses of the term in the Google Books archive suggests that it was first used in the late twentieth century it did include hypothesis testing. In the early 2000s people began to use it as a synonym for ?big data,? and I can understand why. ?Big data? was a well-known buzzword associated with Silicon Valley tech hype.

I totally get why people replaced ?big data? with ?data science.? I?ve spent years doing science (with observations, theories, hypothesis testing, etc.). Occasionally I?ve been paid for doing science or teaching it, but only part time. Even after getting a PhD I had to conclude that science jobs that pay a living wage are scarce and in high demand, and I was probably not going to get one.

It was kind of exciting when I got a job with Scientist in the title. It helped to impress people at parties. At first it felt like a validation of all the time I spent learning how to do science. So I completely understand why people prefer to say they?re doing ?data science? instead of ?big data.?

The problem with being called a Scientist in that job was that I wasn?t working on experiments. I was just helping people optimize their tools. Those tools could possibly be used for science, but that was not why we were being paid to develop them. We have a word for a practice involving labs and gadgets, without requiring any observation or skepticism. That word is not science, it?s technology.

Technology is perfectly respectable; it?s what I do all day. For many years I?ve been well paid to maintain and expand the technology that sustains banks, lawyers, real estate agents, bakeries and universities. I?m currently building tools that help instructors at Columbia University with things like memorizing the names of their students and sending them emails. It?s okay to do technology. People love it.

If you really want to do science and you?re not one of the lucky ones, you can do what I do: I found a technology job that doesn?t demand all my time. Once in a while they need me to stay late or work on a weekend, but the vast majority of my time outside of 9-5 is mine. I spend a lot of that time taking care of my family and myself, and relaxing with friends. But I have time to do science.

Online learning and intellectual honesty

Leave a comment

In January I wrote that I believe online learning is possible, but I have doubts about whether online courses are an adequate substitute for in-person college classes, let alone an improvement. One of those doubts concerns trust and intellectual honesty.

Any course is an exchange. The students pay money to the college, the instructor gets a cut, and the students get something of value in return. What that something is can be disputed. In theory, the teacher gives the students knowledge: information and skills.

In practice, some of the students actually expect to receive knowledge in exchange for their tuition. Some of them want knowledge but have gotten discouraged. Some wouldn?t mind a little knowledge, but that?s not what they?re there for. Others just have no time for actual learning.

If they?re not there for knowledge, why are they there? For credentials. They want a degree, and the things that go with a degree and make it more valuable for getting a good job: a major, a course list, good grades, letters of recommendation, connections.

If learning is not important, or if the credentials are urgent enough, it is tempting to skip the learning, just going through the motions. That means pretending to learn, or pretending that you learned more than you did. Most teachers have encountered this attitude at some point.

I have seen various manifestations of the impulse to cheat in every class I?ve taught over the years. Some people might be tempted to treat it like any other transaction. It is hard to make a living while being completely ethical. I fought it for several reasons.

First, I genuinely enjoy learning and I love studying languages, and I want to share that enjoyment and passion with my students. Second, many of my students have been speech pathology majors. I have experienced speech pathology that was not informed by linguistics, and I know that a person who doesn?t take linguistics seriously is not fit to be a speech pathologist.

If that wasn?t enough, I was simply not getting paid enough to tolerate cheating. At the wages of an adjunct professor, I wasn?t in it for the money. I was doing it to pass on my knowledge and gain experience, and looking the other way while students cheated was not the kind of experience I signed up for.

I?ve seen varying degrees of dishonesty in my years of teaching. In one French class, a student tried to hand in an essay in Spanish; in his haste he had chosen the wrong option on the machine translation app. I developed strategies for deterring cheating, such as multiple drafts and a focus on proper citation. But I was not prepared for how much cheating I would find when I taught an online course.

The most effective deterrent was simply to get multiple examples of a student?s work: in class discussions, in small group work, in homeworks and on exams. That allowed me to spot inconsistent quality that might turn out to be plagiarism.

In these introductory linguistics courses, the homeworks themselves were minor exercises, mainly for the students to get feedback on whether they had understood the reading. If a student skipped a reading and plagiarized the homework assignment, it would usually be obvious to both of us when we went over the material in class. That would give the student feedback so that they could change their habits before the first exam.

The first term that I taught this course online, I noticed that some students were getting all the answers right on the homeworks. I was suspicious, but I gave the students the benefit of the doubt. Maybe they had taken linguistics in high school, or read some good books.

Then I noticed that the answers were all the same, and I began to notice quirks of language that didn?t fit my students. One day I saw that the answers were all in an unusual font. I googled one of the quirky phrases and immediately found a file of answers to the questions for that chapter.

I started searching around and found answers to every homework in the textbook. These students were simply googling the questions, copying the answers, and pasting them into Blackboard. They weren?t reading and they weren?t discussing the material. And it showed in their test results. But because this was a summer course, they didn?t have time to recover, and they all got bad grades.

I understood where they were coming from. They needed to knock out this requirement for their degree. They didn?t care about linguistics, or if they did, they didn?t have time for it. They wanted to get the work out of the way for this class and then go to their job or their internship or their other classes. Maybe they wanted to go drinking, but I knew these Speech Pathology students well enough to know that they weren?t typically party animals.

I?ve had jobs where I saw shady practices and just went along with it, but in this case I couldn?t do that, for the reasons I gave above. My compensation for this work wasn?t the meager adjunct pay that was deposited in my checking account every two weeks. It was the knowledge that I had passed on some ideas about language to these students. It was also the ability to say that I had taught linguistics, and even online.

The only solution I had to the problem was to write my own homework questions, ones that could be answered online, but where the appropriate answers couldn?t be found with a simple Google search.

The next term I taught the course online I had to deal with students sharing answers – not collaborating in the groups I had carefully constructed so that the student finishing her degree in another state could learn through peer discussion, but where one student simply copied the homework her friend had done. They did it on exams too, where they were supposed to be answering the questions alone. This meant that I also had to come up with questions where the answers were individual and couldn?t be copied.

I worked hard at it. My student evaluations for the online courses were pretty bad for that first summer, and for the next term, and the one after that. But the term after that they were almost as good as the ones for my in-person courses.

Unfortunately, that?s when I had to tell my coordinator that I couldn?t teach any more online courses. Because to teach them right required a lot of time – especially if every assignment has to be protected against students googling the answers or shouting them to each other across the room.

The good news is that in this whole process I learned a ton of interesting things about language and linguistics, and how to teach them. I?ve found that many of the strategies I developed for online teaching are helpful for in-person classes. I?m planning to post about some of them in the near future.

The Photo Roster, a web app for Columbia University faculty

Leave a comment

Since July 2016 I have been working as Associate Application Systems in the Teaching and Learning Applications group at Columbia University. I have developed several apps, including this Photo Roster, an LTI plugin to the Canvas Learning Management System.

The back end of the Photo Roster is written in Python and Flask. The front end uses Javascript with jQuery to filter the student listings and photos, and to create a flash card app to help instructors learn their students’ names.

This is the third generation of the Photo Roster tool at Columbia. The first generation, for the Prometheus LMS, was famously scraped by Mark Zuckerberg when he extended Facebook to Columbia. To prevent future release of private student information, this version uses SAML and OAuth2 to authenticate users and securely retrieve student information from the Canvas API, and Oracle SQL to store and retrieve the photo authorizations.

It would be a release of private student information if I showed you the Roster live, so I created a demo class with famous Columbia alumni, and used a screen recorder to make this demo video. Enjoy!

Nobody’s Boy

Leave a comment

I got a paper rejected from a generativist conference a few years ago. A generativist friend of mine said, “Why did you bother submitting your paper to that conference? You knew they were going to reject it.” I said, “Well, the conference was in town, so I figured I’d send something in anyway.”

My friend proceeded to tell me a story from her early grad school days about reviewing papers for her school’s signature conference. She sat down one evening with Professor Big Deal, who glanced through the stack of anonymous submissions and sorted them one by one into piles. “This is from one of Professor X’s students, and this is from one of Professor Y’s students. Here’s another from Professor X’s group. This must be Professor Z.” She continued like this until all the papers were sorted, and then as I recall she had some formula for allocating time to each professor and their students.

I think about this a lot, because I’m not a Student Of anyone in particular. On paper I may look like a student of Professor Bigshot, and that’s probably how my paper got accepted to a conference where Professor Bigshot was a keynote speaker. But I’m not really a Student Of Professor Bigshot. I didn’t ask her to be on my committee. And I know she doesn’t think of me as a Student Of hers, because she was sitting in front of me later in that conference, and walked out of the room right before it was my turn to present my paper.

My relationship with my actual advisor is Complicated, but suffice it to say that we don’t work in the same subfield of linguistics, and I’m tied to the New York area, where she doesn’t have the pull to get me a job anyway. My relationships with my other committee members are problematic in various ways. I’m on good terms with plenty of other linguists, but since I’m not their Student their loyalty to me is always secondary.

Even if my friend’s story about Professor Big Deal is an egregious outlier, it is still a regular occurrence to see professors co-authoring and co-presenting papers with their students, making introductions and writing letters. If you know me professionally, I can pretty much guarantee that we were not introduced by Professor Bigshot, or by any member of my committee. If you’ve seen me present my research, or read it anywhere, or hired me, it’s entirely through my own hard work. I have not had any of the advantages that come with being a Student Of anyone.

You could say that it’s my fault for not choosing the right advisors, or for the problems in my relationships with my advisors. In my defense I would argue that most of the problems in these relationships had to do with my supporting my wife’s progress on the tenure track and my kid’s not being in daycare ten hours a day over my own progress on the PhD. But even if you disagree, does that mean that I deserve to be a second-class citizen in the field?

I know I’m not the only academic orphan out there. Maybe we should get together and found a Home for Orphaned Linguists, where we can hope to someday be adopted by professors with generous allocations of reassigned time, who will co-author with us and introduce us and attend our talks. Some day…

On being a public linguist

1 Comment

People say you should stand up for what you believe in. They say you should look out for those less fortunate, and speak up for those who don?t get heard. They say that those of us who come from marginalized backgrounds, like TBLG backgrounds for example, but have enough privilege to be out in relative safety should speak up for those who don?t have that privilege. They say that those of us who have undertaken in-depth study in the interest of society have a particular responsibility to share what we know with the world as ?public intellectuals.? They say that we linguists need to do a better job of applying our knowledge to real-world problems and communicating solutions to the public at large.

They?re right of course, but there?s a reason more people don?t do these things. They?re hard to do, and even harder to do right. Lots of people are strongly invested in the status quo and in thinking of themselves as good people, and they don?t like to be told that what they?re doing at best ineffective and at worst harmful. Lots of people think that because they?re trans they know everything there is to know about trans issues, or that because they use language they know everything there is to know about language.

Case in point: after watching with increasing frustration for years as the word ?cisgender? was invented and abused, back in December I wrote a series of blog posts about it. I know this is a controversial topic, and I was a bit apprehensive since I was on the job market, but my posts was not idle rants: as a linguist, a trans person, and someone who has observed trans politics for years, I had been trained to do this kind of analysis, and pursued these topics beyond my training.

I anticipated a number of potential objections to my argument and addressed them in the first three posts. As I published each one I was worried it would get a huge backlash, but there was barely a peep (more on that in another post). So for the title of the last one I went big: ?The word ?cisgender? is anti-trans.? Not much reaction.

A few weeks ago I came across a Facebook post by a gender therapist asking for opinions about ?cisgender,? so I left a link to my blog post, identifying it as ?my professional opinion as a linguist.? The therapist then shared my post without identifying me as either trans or a linguist.

Then there was a backlash. Several people immediately called my post ?garbage? and ?horse shit.? There were a handful of substantive disagreements, all of which I had anticipated in my post and previous ones that I had linked to. There was some support, but the vast majority of comments were negative. There were several similar comments made on my blog post itself, most of which I left unpublished since they were repetitive and unhelpful.

I know that plenty of people face far worse reactions to things they post. I didn?t receive any comments on my looks, rape threats or death threats. But it was still very upsetting, particularly as it was posted the same day I began my first full-time job since receiving my Ph.D. – an event that was positive on a number of levels, but upsetting on other levels.

The gender therapist, who presumably helps people with their mental states, showed no interest whatsoever in mine. They made no effort to moderate, did not intervene in the comments, and sent me no personal messages. The idea that a trans person might be losing sleep over these attacks on their page may not have even occurred to them.

The response my post has gotten from other public linguists has been minimal. A columnist who?s written about the issue and encouraged me to write gave my post a few tweets. A radical feminist whose writings about language and politics inspired me for years completely ignored it. It has not been picked up by any of the popular linguistics blogs, or by anyone talking about language, gender and sexuality.

It?s quite possible that these linguists disagree with me. There are some very specific linguistic questions at stake. But linguists love to argue, and I would welcome respectful, constructive engagement with these questions. So far there has been none.

I have also gotten very little support from other linguists. When I was first formulating these arguments a few years ago on Twitter, there were at least two linguists who explicitly denied that I had any standing to contest the arguments for ?cis? that they were retweeting. They were satisfied with the flimsiest of pseudolinguistic rationales in pursuit of their political and social goals, and for whatever reasons I did not qualify as an authentic voice of the trans community in their eyes. I stopped following them on Twitter, and as far as I could tell they had no reaction whatsoever to my posts.

I know that a lot of people don?t want to get involved in flamewars on Twitter or Facebook. It?s really hard to know who?s right and who?s wrong. At first glance I look like just another white guy, and I project an image of success and confidence on social media because that?s what everyone tells me I need to do. Some people may disagree with my stance on a political basis.

I mostly came out of the Facebook flareup okay, although it’s hard to tell how much of my insomnia and touchiness relates to that as opposed to other stresses. Re-reading some of those comments just now was pretty upsetting. I made a decision to focus on the new job, and avoided reading comments, posts or links for a week or two. Now it?s blown over – but there?s no telling when it?ll get shared by someone else.

My main point is that being a public linguist isn?t easy. Speaking out isn?t easy. Fighting on your own behalf instead of some Little People somewhere isn?t easy – even if you?ve got a certain amount of privilege. If you?re wondering why people don?t fight for themselves more often, why they don?t speak up, why linguists don?t write more public posts about issues that matter – there?s your answer. It?s much easier to bury your nose in a book and write about grammaticization vs. reanalysis in Old Church Slavonic.

If we really want people to take a stand on these things, we need to support them. We need to stick up for linguists who speak out in public. We need principles that go beyond identity and political and social affiliation. And we need people who are willing to support linguists who speak out based on those principles. We need people who will make themselves available to back up other linguists on the Internet. Without real support, it?s all empty rhetoric.

What Professor Bigshot said

Leave a comment

I was feeling very nervous, sitting there in Professor Bigshot’s office. I had just been accepted into the PhD program, and was visiting the department to get to know everyone and see if it was the right fit. I hadn?t applied to any other PhD program. If I didn’t go here, I probably wouldn’t get a PhD.

You can figure out pretty easily who Professor Bigshot is, if you care. I guess you could say I’m giving her a pseudonym for SEO reasons.

The student who was showing me around the department had asked, “Oh, have you met Professor Bigshot yet?” I had not. I had heard of her, but I had absolutely no idea what her work was: what she studied, what she had written, what her theories were. I was nervous, sitting there in her office, because I was afraid she would find out that I hadn’t read anything she’d written. I was right to be nervous, but for a completely different reason.

“So Angus,” Professor Bigshot asked me, “You know that the job market in linguistics is very tight? You understand that we cannot guarantee you a job when you graduate?”

I relaxed a bit. I knew this one. I had thought long and hard about it. I said, brightly, “Oh yes. But that’s okay. I have computer skills, and I can always get another IT job if this doesn?t work out.”

“Well, at this university,” Professor Bigshot?s face abruptly twisted into a snarl. “We are not in the business of granting recreational PhDs.”

That was the last thing I was expecting to hear. I did the only thing I could think of: I thanked Professor Bigshot politely, got up and walked out of her office.

I still had a day and a half before I left town. I had planned to visit classes and see the rest of the university.

I didn’t quite know how to tell my student guide what Professor Bigshot had said, so in a few minutes I was sitting down in Professor Littleshot’s office. I didn’t know what he had done in linguistics either, but at this point it hardly seemed to matter.

“So Angus,” said Professor Littleshot. “Have you made up your mind whether you?re going to attend our program?”

I opened my mouth. “Well…”

“Is there anything I can say to convince you?”

I shut my mouth and thought for a minute. ‘Well, I guess you just did.”

That was slightly over nineteen years ago. Professor Littleshot retired before I could propose a dissertation topic. I wrote a dissertation in Professor Bigshot’s theoretical framework, received my PhD in 2009, taught linguistics as an adjunct for seven years, sent out applications for tenure-track jobs and was invited to exactly zero interviews. Last week I started working as a Python developer in the IT Department at Columbia University.

Recreational PhD? Well, there have been times that I’ve enjoyed quite a lot. And yes, I suppose you can get a back injury, chronic insomnia and thousands of dollars of debt from plenty of other recreational activities. Maybe I would have enjoyed it more if I hadn’t tried so hard to prove Professor Bigshot wrong.