Some Thoughts on Gripping & Backtile

Gripping and Backtile are languages (or language sketches) by Sai, with collaborators, in the tactile modality--i.e., communicated through the sense of touch.

Gripping, developed in partnership with Alex Fink, is by far the more fully developed of the two. Sai documented the development of the phonology in two YouTube videos, Phonology of a gripping language, part 1 and Phonology of a gripping language, part 2, and Sai & Alex gave an introductory talk on the language at LCC3 (slides available here to follow along). There is also a reference grammar.

Backtile, developed in partnership with Nai Damato, is a language sketch, for which the linked Google Doc is the entirety of available documentation.

My initial approach to relating to either of these languages is to compare them with how natural tactile languages operate. There are not many of those, but between nat-tact-langs, Gripping, and Backtile, we can define a fairly large field of potential design space for tactile languages, which is pretty exciting from the point of view of providing inspiration for other conlangers. Unfortunately... I don't actually know a whole lot about natural tactile languages! In fairness, not many people do--the literature is sparse--but Sai and their collaborator for Backtile, Nai, at least have considerably more personal experience in that area (where "considerably more" means "any at all", and Nai actually has a relevant university degree). Fortunately, the Backtile document does explain its naturalistic inspirations, and Sai, Nai, and Alex have all been available to provide additional context.

I want to be annoyed at the limited documentation for Backtile, but it is admittedly just a sketch, and I can't fault the creators for not feeling like developing it further; if that's all there is to document, then that's all there is to document. As it is, Backtile intersects with some of the coding strategies used in actual Tactile ASL / Protactile. There is, however, also a smidgen of artificiality in the use of braille, with finger and palm pressure used to communicate the dot pattern of a braille cell for spelling. In fairness, this is something that is actually done in Deaf-Blind communities, but importing braille feels to me like importing oral spelling into English--it's a distinctly technological process, tied to the idea of writing, and as such goes against my own intuitions about what is "natural to the medium". Alex has expressed similar feelings, though neither of us can precisely identify which features make it feel artificial, aside from its history as a designed system. But hey, if real life Deaf-Blind people actually do that already, then clearly it works, so you might as well use it! The other aspect in which Backtile feels somewhat artificial is in the extremely restricted articulatory space--it's articulated exclusively on the upper back/shoulder blade (hence the name). However, this feeds into some desirable properties of the system--being less tiring than hand-following, less socially awkward than more high-contact tactile signing, easily accessible, and allowing full duplex communication between two people with hands resting on each other's shoulders.

Gripping is in some ways much more artificial, although it has some of the aesthetic of hand-following tactile signing. It is meant to be a covert communication system for couples in situations where it would not be weird for them to be holding hands in public. Though neither Sai nor Alex were aware of this when constructing Gripping, it has some similarities to Somali Tactile Livestock Negotiation Code, in that involves one hand from each person, communication by pressure on particular points of the receiver's hand, and is meant to be covert--although in the case of Somali Tactile Code, only the message is covert, not the fact that communication is occurring in the first place, and Gripping is considerably more elaborated and flexible! (Note that the linked article refers to the Somali system as a "language", but it's really not--it is a highly context-restricted code with limited expressivity.)

I mentioned above the concept of "natural[ness] to the medium", a useful concept which I was introduced to by Sai. I'll just quote, because they describe it at least as well as I could:

I have a conception of "natural to the medium", which has only a coincidental relationship with "is similar to natlangs". All natlangs are natural to their media (barring some bad conlanging by authorities), but not all things that could be natural to the media are found in an extant natlang, because the latter is by no means a covering set of the possibility space. There are things they just don't happen to have evolved to, but which could occur — and among those, there are both ones that could occur by derivation from extant/historical natlangs, and ones that could not occur from them because the founder effects plus diachronic changes wouldn't reach some part of possibility space.

This is a useful concept to have around for critiquing all sorts of conlangs, but especially for critiquing conlangs in media for which there just aren't a lot or (or any) natural points of comparison. And while Gripping does seem very artificial with things like Tactile ASL and Tactile Auslan as one's basis for comparison, I don't think it can be said that it isn't natural to the medium--and given that it has radically different design goals, I don't think we should expect it to look "naturalistic" with regard to natural tactile sign languages. Indeed, Sai's phonology videos go to great lengths to explain how it is natural to the medium, by detailing the extensive experimental work that Sai and Alex did to find sets of contrastive features that were easy to articulate and easy to perceive, which is really a rather unique opportunity available in this particular medium, because it still uses the human body, just in a new way. That kind of interactive experimentation with a new communication medium is considerably harder to pull off for, e.g., alien languages using senses and articulators that humans don't have, like my own Fysh A language.

One particularly interesting feature of Gripping is that the phonology is asymmetrical--the person whose thumb is on top in the grip produces signs differently at the articulatory-phonetic level than the person whose thumb is on bottom. This is to get around the fact that switching positions is awkward and slow. However, if one were to eliminate the secrecy requirement, a different solution inspired by four-handed Tactile ASL signing presents itself--you could use a double grip for a full-duplex communication, where each participant has a dedicated receiving hand and a dedicated sending hand! There's an idea if any other conlanger wants to pick it up. (Also note that there is a lot more phonetic space available to explore if you drop the requirements for secrecy, so if you want to experiment with your own one-pair-of-hands tactile language, don't assume Gripping has already covered the entirety of available phonological options!) Sai and Alex did not pursue this route for several reasons, in addition to the obvious loss of secrecy:

1. You can't move around easily while clasping both hands--this would make it a converse-in-place system. However, that's not too different from what actually happens with natural Deaf-Blind tactile communication, so it's not a deal-breaker if someone else wants to run with the idea.
2. Clasping both pairs of hands is considerably less comfortable in a relaxed position, and more effortful in an un-relaxed position.
3. Full duplex communication is absurdly mentally challenging, so while having the option is neat, lacking it really isn't a problem. Full duplex really just allows you the option to interrupt and talk over your interlocutor. (But perhaps this could be used as the basis for some linguistic sci-fi--imagine an alien species which does have the mental capacity for full duplex communication, and a medium that supports it!)

As-is, with the single-grip setup, Gripping is somewhere between half- and full-duplex on the phonological level, similar to audio languages in which it is possible to talk over people. Unlike audio languages, however, certain combinations of subordinate and dominant articulations are actually physically impossible to pull off! The half-duplex system currently documented is already an improvement over what is possible with, e.g., Tactile ASL, which does require repositioning to exchange the roles of sending and following hands. On the other hand, Protactile communication does allow for limited back-channeling (e.g., "ACK", "uh-huh", "yeah", etc.), and Sai has speculated that a similar level of discourse-management simultaneous communication could be added to Gripping using the subset of phonology that can be simultaneously articulated and perceived, though they have not yet tried it.

Additionally, just as oral languages can be partially lip-read and ALS signs can be felt, Gripping articulations are useful outside of their primary modality. Sai has given the example using Gripping as a very restricted sign language with no hand motion, for covert hand signals--which is very similar to the "small motion" Atreides sign language that David Peterson developed for the latest film adaptation of Dune.

Some Thoughts... Index