A Bad Procedural Argument (For Reducing Online Hate-Speech).

~60 min sketch. Best in webkit browsers. Click here to interact. 

This sketch allows a user to ‘reduce’ hate speech online by ‘limiting bandwidth’ with the obvious catch that the occurrences can’t be reduced in proportion to other kinds of speech, only speech overall can be limited. It presents the premise that hate speech is always in proportion to other kinds of speech and from this a kind of naïve argument that  limiting one kind of speech is just to limit all kinds of speech. (This description is admittedly very imperfect though  as there is nothing to indicate, procedurally or otherwise, that the purpose of the manipulable features is to affect the occurrences of “HATE”.)

I’d hope the clarity of the argument’s insufficiencies could lend it educational utility; that a procedural argument like this would immediately raise, in the mind of the user, objections or questions about why, for example, the occurrences of “hate” have no effect on the overall exchange, either at the time they appear or afterwords; or how online speech and dialog might be represented in ways other than atomic packets traveling from one end to another (for example as ripples in a pond). Ideally, bad procedural arguments like this would make good instructional material for students learning about procedural arguments and serve as the basis for a student response in kind, e.g. in procedural form.

However, there is a significant hurtle—as with any programming, it is very difficult for novices to distinguish between ideas that are possible for them to execute and ideas that are very much beyond their reach. And so student response would need very technical guidance, distinct from the meat of the argument they might wish to present.

As such, a ‘bad procedural argument’-as-instructional-material would do well to carry with it, implicitly, a variety of procedural criticisms that could be expressed relatively easily. Better yet, it would be great if the underlying learning objectives could include a focus on arguments about certain kind of general phenomena that could be represented in different but related ways, algorithmically, such that students could build their understanding of representational considerations in parallel with technical skill building.

But what genre of phenomena would be a good fit for an objective like this? Would a survey of ‘networks’ allow breadth (touching on human interactions on social media media, technologies, business phenomena, and mycelium) without being too general?

(On a side note, ‘bad arguments’ was an early idea for the focus of sketches on this site.)

Leave a Reply

Your email address will not be published. Required fields are marked *