GPT based text generators like ChatGPT or Microsoft Copilot have rapidly become a “cultural sensation”. Here I provide scientific background and guidance on how to think critically and mindfully about these tools in academic writing and research.
I saw a thing fly by on PsyArxiv and could not ignore it so I’m doing a drive-by peer-review. It claims that English-based AI-generated norms are “of particular value for under-resourced languages”. Is that pesky linguistic diversity bothering you? Here, try on these rosy English-tinted glasses and everything will look all prim and proper, promise. Warning: snark detected.
It takes two to tell a story: narrator and audience. Response tokens or continuers like ‘mhmm’ play a key role in making stories work. Two new papers extend the study of continuers across languages and modalities. Work by Lutzenberger et al. reveals the importance of minimal tokens that don’t occupy the main articulators in British Sign Language and Spoken British English. And a study by Börstell showcases a neat methodological replication and extension of the sequential search method, applied to Swedish Sign Language corpus data, with promises of broader applicability.
Academics often feature a few selected papers on their home page. Typically these represent big projects or work published in prominent venues. What I’d like to see more of is “niche papers”: work to be proud of even if it has managed to remain a bit obscure. What are your niche papers?
For years now, I have responded to review requests from Elsevier journals with a friendly explanation of why I cannot in good conscience devote my free labour to their for-profit venture. I always include an out: make some work in the same journal available in open access. Somehow they always find this isn’t possible.
Lab rotations are a regular feature of work in my research groups. Students join the lab and figure out a…
You hear a word like tugɯn-dugɯn and two possible meanings ‘heartbeat’ or ‘gentle movement’. Which one do you pick? People have intuitions about the fit between forms and meanings, even for words they have never encountered. But can we explain those intuitions? And can we use that explanation to predict what people do in experimental tasks? That is the question we seek to answer in The Anatomy of Iconicity.
Knots are fascinating: they tie together topology, embodied experience, and material culture. Here I discuss a paper about knots and intuitive reasoning in Open Mind. The paper ties itself into knots about intuitive physics, but pulling at some of the threads, it turns out it’s actually more like a Trojan horse for 4E cognition.
Remarkable promises “integration” with popular cloud storage services but offers only the most clunky implementation possible, where you need to copy files manually one by one. This is not integration, it is the mud run approach to user interface design. A surprising miss given the company’s stated mission to use a human-centric approach to technology.
Reading Latour can feel like sorting through ideas the way you deal with laundry fresh out of the tumble dryer, sorting things out, reuniting pairs of socks, finding the inevitable singletons, creating some semblance of order and accepting loose ends. It all comes out in the wash.
Writing is thinking. The writing process is the most neglected part of our job. We spend millions on fancy equipment and uncountable hours on training for using this or that toolkit. Yet we assume the BA-level academic writing course we once followed is sufficient; the rest we’ll just learn on the job and hopefully soon we’ll automate away with LLMs. It is all formulaic anyway. To think this way is to hollow out the very foundations of scholarly work. Can’t think original thoughts if you don’t find your own voice.
Following on from my post about setting up a Raspberry Pi 400 as my kids’ first personal computer, here I share how I’ve made the system easy to manage remotely and imposed some light parental controls, all while keeping the system open to tinkering. It will be a matter of time until they discover how to break out of it — indeed that is pretty much part of the journey of discovery that I hope they’ll take on.
Our kids (6 and 9) wanted to learn to type, and I think it’s useful for them to become computer literate sooner rather than later, so I spent some time figuring out options. I settled for a Raspberry Pi 400 and they love it.
Readers of this blog will know that I’m an avid user of Zotero and of the reMarkable paper tablet. Zotero is a supreme reference manager and reMarkable is simply the best when it comes to reading and note-taking. In this post I share some tips and tricks for making them work better together.
I am extremely happy to announce that NWO will be funding the project Futures of Language over the next five years. We will start in September 2024; stay tuned for news about positions for postdocs, PhDs, and research software engineers.
We study artisanal and artificial ways of languaging to better understand language + technology, and to reimagine our linguistic futures.
I have been blogging at The Ideophone since 2007, and not all of it has been as ephemeral as my PhD promotor once feared. Over the past week I have worked with Rogue Scholar to archive selected content from The Ideophone and make it more durably accessible. This posts documents the process and some of the choices made.
A common trope in recreational mathematics is the grazing goat problem: for a goat tethered to some piece of rope, what is the area it can graze given the length of the rope and various other variables like the shape of the field? In my recent Annual Reviews article I argue that linguistics has something of an inverse grazing goat problem.
Randomly clicking around in the ASIO4ALL 2.15 settings in Ableton Live, as one does, I changed something and promptly got the error “live does not support 8000 as a sample rate”. Skip this post unless you’re one of the lost souls looking for a solution.
Out now in Annual Review of Linguistics: Interjections at the Heart of Language. This review critically considers received views of interjections as involuntary grunts and provides a number of alternative ways of thinking about interjections. I would be very happy if you read it.
Language makes us human. But there is an interesting asymmetry in our willingness to ascribe linguistic capacities to non-humans: animals are seen as having none, whereas computers may well master it according to many. What curious conception of language makes this asymmetry possible? And what do Descartes and Turing have to do with it? Notes from a new essay about language between animals and computers.