Linguistic Landscapes: from old road signs to e-travelogues

“Linguistic landscapes” is a sociolinguist’s term for the study of public signs and writings you find along country roads, in urban jungles, and on walls, pillars and posts all around the world. Like all functionally specific objects, they tend to use special language rules, lots of proper names, abbreviations, and numerical data.

Yet instead of remaining curious relics of another age, these public language icons will almost certainly evolve into new kinds of digital information sources. Let’s see how.

Today’s road vehicles are growing smarter about their own processes. Soon this may extend to the roads and landscapes they drive through, as well as their own operations.

For example, an in-car AI development such as Wayve’s Lingo-1 can generate a running commentary in natural language on the vehicle’s functioning along the road taken. Extend this capacity to the world outside and around the road, and a smart vehicle might be able to inform passengers about everything else in the landscape they are driving through, not just the car’s performance.

Obviously, there is plenty to say technically about the specific communicative form and impact of language in road and street signs of all kinds around the world — whether they include more than one tongue, who produced them and how, who “uses” them, and whether they will be sufficiently informative for future uses, etc. But this is simply a linguist’s starting point.

Let’s accelerate forward into the tech future of languaged landscapes…

(Road)signs of progress

We can, for example, look at signs set in very varied offroad landscapes, featuring a few words of Latin or other moribund languages on tombstones in many European graveyards, or as living tongues inscribed on diner menus in different languages in a Pacific Rim urban eatery.

Today these sign events can all be photographed, analysed for their origins, history, language features, tensions, implicit local urban communication policies, and catalogued for posterity.

But most of all, we use street-sign content as a source of practical information about shopping, travel directions, and also legal issues (e.g. contesting the implications of content written on a sign you could not understand but which led to an accident). They are items of “public” language anchored to specific places and tasks.

If we can’t understand their language, we will increasingly use a phone/camera that can photograph, identify and decode the language items using an inbuilt translator app. The whole operation will take a few seconds. Eventually, available language options will be extended to cover all relevant tongues for the user base.

Ideally, this means being able to translate from languages such as ancient Egyptian, Greek, Latin, Hebrew and Arabic to aid the average tourist/traveller in the Middle East or visiting landscapes inside major museums around the world. After all, ancient inscriptions are also a form of public signage.

Above all, however, any worthwhile translating phone app should be able to handle all current, high-visibility, commercial and informational content in all current languages. This means that the apps must keep vocab and linguistic usage recognition constantly up to date.

In fact the automatic, instant updating of all publicly visible “official” data on signs of all kinds will probably become a legal requirement in the future, as well.

Let those old signs speak!

Speculating about new “language landscapes” some twenty years into the digital age, the obvious major improvement would be to digitize most road and other public signs at source, making them smarter and fully interactive. This way they can

• automatically update or change their content,
• direct information more effectively to phones and other apps,
• adapt to new language needs and publics based on up-to-date data collection,
• and generally evolve into a dynamic two-way source of rich information.

This would also enable more signs, and possibly more discreet signs to inform and enlighten us about the landscape we are travelling through.

There are naturally difficulties with this vision of interactive signage, apart from the cost and complexity of the roll-out of the basic infrastructure. We have already suggested that many countries have signs in their landscapes dating from both ancient and modern epochs, and these should be ideally preserved as part of our cultural landscape heritage. They will need, however, further explanatory content to help people understand their origin and purpose.

Digital landscape content (above all for roads, airports, and rail transport systems) that is designed to guide us locationally, inform us of available goods and services, or identify specific locations, sites, and historic buildings, etc. could all be permanently available via smart glasses. It could also be constantly updateable to address changing climatic, geographic, and landscaping conditions.

We already have a first generation of in-car information tools such as Waze, Sygic Parkopedia, Roadtrippers, AllTrails, Sygic Travel, Viator and many more, apparently available in up to 50 or so languages. But if we can render the signage itself interactive, far more interesting opportunities could be opened up.

Most pre-digital signs, composed of text and symbols on stone, wood, metal, and concrete, etc. could be read via a QR code or similar, and then decoded. As a result they will continue to transmit information to a phone or similar digital device for some time to come.

In this case, the user’s device is the obvious way to access an automated translation service for the relevant languages, together with other related content. But the special terminology used for such signage might need more careful inclusion.

Road transport signs, with their unique information presentation and purpose, can also be captured by an in-vehicle scanner, and explained in words or magnified visually so that drivers can integrate such information into the vehicle’s overall visual guidance applications in the case of e-vehicles.

However, the real interest would come from the very obvious concept of digitizing the road travel info infrastructure itself. What could this deliver?

Massive interactivity on the road

Current apps, then, will enable a driver or passenger to use natural speech to interrogate the knowledge base for further travel information — e.g. various aspects about either the road ahead and traffic/weather conditions, or interesting sites across the landscape.

Starting from this sort of standard in-car travel-info package, it should soon be possible to use generative tech to transform existing basic travel information into a rich interactive travelogue.

This would be spoken in the language of both passengers and driver (or in different languages for each passenger via their ear plugs), offering the history and eventually the ecology etc. of the route being taken. Such a service might be especially attractive for users of the coming generation of automatic-drive vehicles.

Such a service could inform (in various appropriate linguistic registers — adult, child, foreigner, etc.) about surrounding sites of interest, and other relevant details associated with the route. It could all be rapidly generated from a simple prompt and artificially articulated in real time in the appropriate language.

In other words, a friendly, eloquent co-pilot for your pleasure driving. To avoid annoying soundtrack noise, information could also be projected visually on an in-vehicle screen or passenger phone screens.

A final turn of the interactive screw would be to automatically answer questions or prompts from car or bus/coach/train passengers (and even drivers) using an AI app. This would ensure a fully interactive experience of guided travel, especially for tourists and other pleasure riders.

As well as receiving standardized information from the travelogue co-pilot commentary, participants would be able to question the device and probe further into different topics associated with the journey. Boring landscapes could come alive. This kind of personalized service could also be individuated locally, with each participant receiving their tailored content…

In other words, our language landscapes could gradually evolve into data resources that different sorts of media could then exploit at leisure, some highly practical, backed by the law, and others for pure enjoyment.

For your daily to and fro journeys, you would of course be able to turn the whole thing off!