As we write this, we are just emerging from of a period of lockdown related to the COVID-19 pandemic. This period began in England in the spring, when birds were beginning to sing, and many people noticed that they were hearing and listening more to birdsong. We certainly were, and it has helped us to focus our mind on this writing. We were drawn to the sound of birdsong, but as much as we listened, it was very difficult to put into words, or summarise, that experience.
We present a device for translating birdsong into English using the arcane Solresol musical language. We describe the methodology of its construction and demonstrate a variety of applications, outputs, and modes of engagement. Our motivation was to create an artwork that would serve as a site for discourse on inter-species communications in the context of the current environmental crisis.
In this section, we contextualise the work and explore its significance in relation to specific theoretical positions, particularly those of Donna Haraway, Bruno Latour, and Tim Ingold. The specific framing employed is Haraway’s concept of a speculative fabulation, which is explained further below. We consider how the device works as an artistic research strategy and propose potential future lines of enquiry.
Several overviews of contemporary art practice examining human and non-human animal connections have been published (Broglio 2011, Baker 2013, Ramos 2016), and the recent Animalesque / Art Across Species and Beings show at BALTIC, Gateshead, UK (BALTIC 2019), provided a survey of this field. To be ‘animalesque’ means to be like an animal, and the exhibition invited visitors to consider the possibility of becoming an animal, and in doing so, reassess their relationship with the wider environment. Animal/human transmogrification is a common theme in folklore and mythology (for example, werewolf stories), but our approach here focuses on communicating with the other rather than imagining being the other. Communication is a two-way process, and one of the pre-requisites is the ability to understand or translate what the other is saying. The current project is about the first stage of translating from bird to human, but in this discussion, we will mention communication in general and briefly consider issues of speaking to birds as a future work area.
Our device has some resemblances to a gadget that might appear in a science fiction film; alien first contact and the associated communication problems have long been a trope of that genre (Washbourne 2015). In these fictions, a translation tool is frequently used, which may simply be a convenient narrative device to bypass the issue (for example, the Babel Fish in Hitchhiker’s Guide to the Galaxy; Adams 2005). Alternatively, communication issues may be a central concern, like a thought experiment that investigates underlying political, cultural, and philosophical ideas. The Solresol Translator project operates in a similar manner, though the focus here is on communication with non-human inhabitants of our own planet rather than inhabitants of extra-terrestrial worlds.
There is a long history of visual artists acting as shamans in order to communicate with animals (Wallis 2019) – either working with real animals, such as Joseph Beuys’ performance I Love America and America Loves Me (1974), in which he spends three days in a room with a coyote (Beuys 2015), or in other ways, such as Marcus Coates’ Journey to the Lower World (see Broglio 2011) in which he offers to communicate with animal spirits on behalf of the residents of a condemned building.
In The Last of Its Kind, Marcus Coates (2017) plays the last human, attempting to convince the landscape around him (on Fogo Island, Newfoundland) of human progress by repeatedly, and with increasing desperation, shouting out a list of our achievements to the iceberg-laden ocean: ‘We have the ball-point pen! … barbed wire! … pasteurisation! ... pyramids! … the fall of the Berlin Wall! … the end of Slavery! …’. At one poignant point, an albatross flies by nonchalantly, emphasising the gulf between humankind, its languages and achievements, and the natural world. When watching the video, one feels an agonising desire to reach out and talk to that bird. In this work, the artist creates a fictional account of a possible future, which could be interpreted as a warning and a call to reconsider our relationship with the world.
Tales of a Sea Cow (De France 2012), the example given in the introduction, is another fiction involving our connection to the environment. These works are included to demonstrate the approach we take in our work, which is to create a speculative fabulation (Haraway 2016). This is a process of inventing and telling stories or fables, but it does not imply being untruthful. It can be difficult to grasp the concept a ‘truth-like’ fiction. We aimed to accomplish this, and responses we have received lead us to believe that we have had some success. We have shown our work to a variety of audiences, including scientists (ethologists), artists, and members of the public. In many cases, there was engagement with the work and acceptance of the methods. The work is plausible, and audiences have been willing to ‘suspend disbelief’ in order to focus on the underlying issue of inter-species communication.
Our work, like Tales of the Sea Cow, entails attempting to make sense of animal sounds. This is part of a larger project to make sense of our world in light of the current environmental crisis. We now consider how this manifests itself in our work. We examine its relationship to two theoretical positions: first, Bruno Latour’s work on two contrasting types of meaning, and second, the separation of sound and sense in human and animal voices.
4.3 Two realities
What do we mean when we say we want to translate birdsong? Why, when it appears to be an impossible task, are we drawn to it time and again (consider the long history in mythology of communication with birds, such as Odin and his ravens, Huginn and Muninn; March 1898)? Perhaps it appears impossible because we approach the question in two irreconcilable ways. First, there are utilitarian evolutionary meanings, such as the role of birdsong in defending territory or finding a mate; ‘flirting and fighting’ (Collins 2004). The song’s physical acoustic properties can be examined and linked to functional benefits for the bird. For example, it is possible to say:
… nightingales singing more trills elicits stronger responses by rivals and predict mating success. (Naguib and Riebel 2014: 241).
On the other hand, we have meaning in terms of our given experience of birdsong, in all of its beauty and rich cultural resonances, such as those expressed in poetry. Hearing the song, for example, is a bittersweet experience for Keats in Ode to a Nightingale, in which he becomes ‘too happy in thine happiness’ as the bird ‘singest of summer in full-throated ease’ (O'Neill and Mahoney 2008: 443).
If we are to tackle the translation of birdsong, we must consider these two approaches to its meaning, but first we must consider how these two contradictory approaches may have arisen. Alfred North Whitehead (2006: 30; emphasis added) expressed dissatisfaction with what he saw in Western philosophy as :
… the bifurcation of nature into two systems of reality, which, in so far as they are real, are real in different senses. One reality would be the entities such as electrons which are the study of speculative physics. This would be the reality which is there for knowledge; although in this theory it is never known. For what is known is the other sort of reality, which is the byplay of the mind. Thus there would be two natures, one is the conjecture and the other is the dream.
As Latour (2014: 95) puts it:
[Philosophers] have forced upon common sense a rather stark choice between two types of meaninglessness: either the meaninglessness of senseless but real nature; or the meaninglessness of meaningful but unreal values.
Latour (2014: 105) accepts Whitehead’s analysis and wonders when this split occurred in history, asking:
When, in other words, does the nightingale ethologist who is recording the song as a wave, begin to claim that this wave allows her to deduct the song you hear from the total sum of experience?
This quote, more than any other in this exposition, gets to the heart of the issue of we are investigating through this artistic practice. The question is how the beauty and wonder of birdsong in human experience can be viewed as part of the total reality of what birdsong means. This is not to say that we should consider birdsong solely as a metaphor for anthropocentric concerns. However, we should not think of birdsong solely in terms of what it means for the birds. We are interested in considering the entire reality, which includes facts about the utility of bird song for birds, as well as the reality of human experience of it. We are using art practice research strategies of speculative fabulation and interspecies co-creation, which include birdsong and our attraction towards it.
We are inspired by Latour’s analysis that new ways of thinking are required to address contemporary issues such as mass extinction of species:
Reality is not defined by matters of fact. Matters of fact are not all that is given in experience. Matters of fact are only very partial and, I would argue, very polemical, very political renderings of matters of concern. (Latour 2004: 232).
What is required, according to Latour, is a new way of thinking that does not take one side or the other of the bifurcation, or even try to reconcile the two, but rejects the bifurcation entirely. This guides our thinking. We are attempting, in a small way, to contribute to ‘the immense building site … where every intellectual skill … is trying to reinvent an Art of Describing: matters of fact’ (Latour 2014: 121). This work is proposed in the spirit of trying to find new ways of thinking about our relationship to the world and its non-human inhabitants.
The current project is not concerned with what birdsong means for birds. Rather, in the spirit of multi-species becomings (Haraway 2016), we are interested in what birdsong means to humans and the entanglement of birds in human culture and technology. This is crucial to this project. This intention distinguishes our device from superficially similar ones seen in the previously mentioned alien-encounter science fiction trope. Despite the wide variety of approaches in that genre, they all have a functional approach in common. They are not interested in the aesthetic qualities of alien speech, but in understanding with a goal in mind, whether it is a practical goal or in a quest for knowledge. The current project’s central concern is the Great Divides between human and animal, organic and technical (Latour 1993, Haraway 2003), and the sensual and the cerebral. As an artwork, the device aims to capture the audience’s attention by emphasising this divide, creating tension and a space for contemplation. In the following section, we will consider the separation of sound and sense in human language and birdsong in greater detail.
4.4 Sound and sense in language and birdsong
According to Ingold (2007: 15), the separation of sound and sense in language is a modern construct. In modern Western culture, we consider the sound of a word to be a carrier of meaning rather than having any meaning in and of itself. To put it another way, we consider language to be mute, whereas we consider music to be devoid of verbal meaning. Language has lost its sound, so the meaning of words is the same whether spoken, written, or printed. Abram (1996: 139) and Ingold (2007: 27) argue that this was not always the case, and that this separation would be incomprehensible in purely oral cultures. Abram (1996: 144) suggests that in indigenous oral cultures, ‘animal cries and communicative calls are pondered, mimicked and replied to by human hunters, becoming as it were, part of their tribal vocabulary’.
It is ironic, then, that Sudre proposes a universal language based on music, but the relationship between the language’s sound and its meaning was purely arbitrary. This arbitrariness is reinforced by the fact that in Solresol, colours or hand signs can be used in place of sounds. We use this inherent irony in our work playfully to provoke contemplation about these differences. Agamben (2018) illuminates the relevance to inter-species communication by proposing that a splitting in the voice occurred at the moment we evolved into humans. This was a schism between what remained of animal language (such as a bird’s call or a lion’s roar) and the human language that was emerging in its place, external to individuals, as an organ of objective knowledge. According to Agamben, this was the source of another irreducible splitting that runs through human language. He compares it to the distinction between a musical and a semantic series:
These two series, which coincided in the animal voice, separate at each turn and oppose each other in discourse following a twofold and inverse tension, in such a way that their coincidence is impossible and, at the same time, irrevocable. (Agamben 2018: 17, emphasis added).
Let us look at a specific example from the current project: the Lite version of the translator displayed on the outputs page, where the Skylark song is being translated. On the left, the ‘song catcher’ element displays a waveform, and the ‘dissection module’ shows this being analysed into frequency and velocity (volume), before being passed to the ‘note colour engine’, which converts the sound into a stream of notes. These elements on the left side of the display are plausible, and the words being generated can be understood as English words, but how the sound and meaning relations are established remains a mystery in the ‘word melding and translation moiety.’ The explanation of the use of Solresol seems somehow not fully satisfactory.
To take another example, consider the Translator producing texts like the one shown in the ‘auto-poem’ Blackbird:
Watch moon, past complacent humidity gathers
Immortalize other, captivate water for religion’s fashion
Or the phrase ‘future earth scream now’. The viewer experiences dissonance as a result of reading the words and hearing the birdsong. They may believe that if the device’s goal is translation, the outputs should be objectively understandable, but they are not. However, there appears to be some poetic meaning in the words generated. It is similar to seeing faint stars in one’s peripheral vision that disappear when one tries to focus on them. There is a sense of a ‘pre-cognitive’ meaning that occurs in the overall experience of the words prior to bringing intellectual capabilities into play. This is more akin to how we experience birdsong than to how we normally think of meaning in language. The device’s purpose as an artwork is to play with these ideas and encourage their contemplation without providing an answer or a fully thought-out statement.
4.5 Other threads and future directions
This preliminary exposition provides an overview of this art practice research. Beyond the main focus discussed above, we briefly outline a variety of other directions that have been tentatively explored or are suggestions for future research in this section.
The section is lengthy because we propose that the translator is also another of Haraway’s SF’s, a string figure. These are games in which patterns are created in string and held between two hands, after which other players take the string and transform one figure into another; a popular example is the Cat’s Cradle. The playfulness, interaction between the players, and the bringing together of diverse strands within changeable semi-permanent forms are the relevant elements here. Another useful related concept is Ingold’s concept of the meshwork (Ingold 2011), in which lines become intertwined and entangled. In our case, there are many lines of becoming, including the line of human language and its development, the line of birdsong, human culture, technology, dreams of a universal language, art, science fact, and science fiction. As a result, the ability to stimulate a wide range of possible lines of thought and activity is inherent in the work.
Initially, the software was designed to create ‘passive’ artworks (videos, poetry, etc.), with more engaged and interactive applications added later in the project. The device’s use in performance merits further exploration. We read poetry generated by the system (see, example on outputs page). We were experimenting with voice tone and rhythm as a mode of expression separate from the meaning of the words. The device was used in a live jamming session with a musician, and a recording of this is available on the outputs page.
These more interactive aspects have promise for the future. For example, creating a large number of devices and distributing them to a group while they walk together, and then recording the conversations that ensue.
The Dada movement was interested in language and frequently manipulated it in ways that emphasised the acoustic and visual qualities of words over their semantic meaning (Schaffner 2005). Our work has spiritual and technical ties to Dadaist methods, particularly found poetry, redacted poetry, and sound poetry. The use of the Solresol dictionary outside of its original intent is also linked to ideas of the readymade and repurposing (Demos 2006).
We created poetry from the Translator's stream of words; unlike redacted poetry, we only deleted words sparingly. We read the poetry in two ways. First, we used a text-to-voice app (see methods and appendix) to add another layer of translation. This process removes control while increasing interest, especially when unusual pronunciations or rhythms occur. The other method was reading in performance, in which we experimented with tone and speed of voice, influenced by, but not directly mimicking, birdsong. We were influenced by sound poetry, specifically Kurt Schwitters’ Ursonata.
Sudre included colours within Solresol, in addition to musical notes, to help deaf people communicate. In our devices, translated words are displayed alongside the corresponding Solresol colours and musical notes. See the screenshot on the methods page, for example, where the translated word ‘lapis’ is also shown as the Solresol representation, indigo-red-red-violet; when live, the notes A C C B (equivalent to the Solresol word, ladodosi) play at the same time.
The hue of the colour is fixed in all cases as specified in the Solresol system, but as part of our experimentation, we created a version in which the saturation and lightness varied according to the volume of the sound. This is shown in the second example on the outputs page.
One could argue that further developing these colour elements would produce more aesthetically pleasing results. We concluded that they were, in fact, becoming a distraction from the core translation issues. As a result, subsequent versions of the software were simplified. This had the advantage of requiring less computing power, which was more suitable for the small Raspberry Pi versions.
4.5.4 Birdsong and music
There are several stages to mapping birdsong onto Solresol words. During development, we investigated the possibility of allowing the user to hear these intermediate stages in a variety of ways by changing the mode or key that is heard. There was also the option of adding a drone (fixed notes) as accompaniment and changing the virtual MIDI instrument. On the outputs page, a variety of examples of the music created in this manner can be heard. The music produced is influenced by both the input birdsong and our analysis thereof. The work usefully poses the question of whether this is a process of co-creation (or, more broadly, whether animals can make art; Yan, 2013).
As part of our broader contextualisation we considered musicians who were inspired by birdsong (for example, Vaughan Williams, The Lark Ascending) or created music based on a more direct transcription (see Messiaen; Chadwick and Hill, 2017). However, one could argue that birdsong is already music (Yan 2013). If so, then music and birdsong are on the same side of the previously discussed divide, and a term like transposition, rather than translation, might be more appropriate to describe that movement.
Whatever we think about the ‘meaning’ of music, one thing is certain: sound is critical. The central irony of Sudre’s conception of Solresol is that sign (sound) and meaning became separated, just as they do in conventional language. The selection of sounds for given words is completely arbitrary. The concept of music as a universal language is based on the idea of a universal aesthetic understanding, in which sound and meaning are inextricably linked. Aesthetic judgements about music and birdsong are in a similar register, whereas the current work focuses on the opposition of this type of aesthetic and objective meaning. Although we have discussed music creation, we have concluded that it tends to distract from our primary concern. Nevertheless, we may revisit this as a possibility to pursue more thoroughly in collaboration with musicians at a later date.
4.5.5 Speaking as a bird
Our long-term goal is to enable inter-species communication, but for the purposes of this presentation, we have focused on translating from birdsong into English. A significant advancement would be the creation of a ‘reverse’ device that would output birdsong in response to text or voice input. This then opens up a plethora of exciting possibilities for the two devices responding to each other and other potential interactions (with wild birds in the field or captive birds, for example, with due consideration for ethical concerns).
4.5.6 Other beings
The device parameters have been optimised for birdsong translation, but there is no reason why it cannot be used for any non-human animal, plant, or inanimate natural sounds. This raises several interesting questions. First, consider the influence of our culture and language on our listening experience. The concept of the beauty of birdsong is deeply ingrained in our culture; would we think about it in the same way if this were not the case? What about translating other calls, such as foxes? Could we avoid the overlay of cultural meaning if we used the Translator for them instead? The use of technology is central to the current work, but we should consider what philosophical issues arise if animal sounds are only accessible through special technology, such as frequency shifting required to hear the calls of bats, or underwater hydrophones required to hear whale songs.
4.5.7 Alternatives to Solresol
When the work is presented, one suggestion that frequently arises is the use of a different dictionary. Solresol could benefit from a modern update because it contains many archaic words that seem unnecessary in a limited vocabulary. Consider a few of the more unusual words or phrases in Solresol: Phantasmagoria, Zouave (a class of infantry regiment in the French army), Philotechnic, Choral Society, Cavatina, Arpeggiate, Daguerreotype, Baccalaureate, Fructify, Minister of Education and Religious Affairs, House of the Emperor, Apostolate, Seek to Convert, Judaize, Become Lukewarm, Sardonyx, Cadastre.
Examining what Sudre’s vocabulary choices reveal about him, the time and place he lived, and his worldview, could be enlightening. Solresol’s Eurocentric stance is undeniable, both in terms of its vocabulary and the use of the Western C Major scale, and this can be considered problematic. Sudre’s assumption that his set of words and musical system were a solid foundation for a universal language could be viewed as colonial, and these issues could be investigated further.
We could remove some of the more obscure words to make room new ones, such as Computer, Car, Mobile phone, Motorway, Space Shuttle, and Mars Rover. Alternatively, one could populate the look-up table with ‘bird relevant’ words that are currently missing, such as Preening Gland, Nest Box, Migration, Bird Table, and Fledgeling.
Instead, we could use a completely different vocabulary. We conducted a small experiment in this direction by employing Enochian, the language of angles as recorded in the journals of John Dee and Edward Kelly in late sixteenth-century England (Asprem 2012). As shown in the second example on the outputs page, the Solresol words were translated into this language.
The use of Sudre’s words adds a poetic touch to the work while also drawing attention to the topic at hand in a usefully provocative way. Exploring these aspects further was beyond the scope of the current work, but they do point to promising future research areas.
4.6 Concluding thoughts
The story of the search for the perfect language is the story of a dream and of a series of failures. Yet that is not to say that a story of failures must itself be a failure (Eco 1995: 19).
Sudre dreamed of the creation of a worldwide language, but ultimately failed. According to Eco (1995: 24), such failures were frequently caused by a conflation of the concepts of a perfect language and a universal language. A perfect language is one in which a sign and what it expresses are inextricably linked. A universal language entails a set of external signs on which all parties agree. There is a conflict here. A perfect language capable of exactly representing our deepest thoughts and every single variation of all objects will be impractical as a universal language since it would require infinite nuance to express every possible thought of every person.
The work presented here, ironically, alludes to dreams of a single universal language. In some ways, our work is a failure because it does not result in a translation in the traditional linguistic sense. However, it refers to the dream of a universal language as well as the dream of inter-species communication. We argue that, in addition to our given experience of birdsong and the ethological facts about its meaning, our dreams must be viewed as a component of reality as a whole.
It may be questioned as to how we can call this work research. It is carried out in the context of an academic institution and employs reliable cognitive processes (Ichikawa and Steup 2018). The methods are reliable because accepted techniques of sound analysis are used, the device is implemented using established software, the core of the translation uses a published system (i.e. Solresol), and the results are reproducible. It is true that a series of musical notes can approximate birdsong, and it is also true that Sudre invented a system for relating musical notes to words. We meet the novelty requirement of research as we are not aware of any similar devices. It is research in the sense that it entails a methodical, rigorous, and detailed investigation. As a result, many aspects of our methods similar to more traditional forms of research. But, if our hypothesis is that when a Blackbird sings, it means ‘watch moon, past complacent humidity gathers’ or ‘future earth scream now’, how can we say this is in any way ‘truth-like’? How can it be claimed to be a genuine serious-minded work (as we do)? Our position is that traditional approaches to research and knowledge are insufficient for solving the types of environmental problems we wish to address. We are looking for alternative methods, as advocated by Latour, and the work presented here is one such attempt in the spirit of Haraway’s speculative fabulations and multi-species worldings. Our contention is that the work provides a way of generating new thinking and novel insights. According to Eisner (2006: 17), ‘Good arts-based research ought to generate questions worth asking and ideas worth pursuing’. We can only briefly describe the practice and touch on some of its provocations in one exposition, but this is the spirit in which we claim this work to be artistic research that generates new knowledge.
Various threads in this project come together in the translation device, and this is where the new knowledge is found. This is not knowledge in the strict scientific sense, but knowledge in the way these elements are brought together, and held together, so that the viewer can contemplate them. Inter-species communication, sound versus sense, and the two realities of utility and beauty are central issues. There are many lines leading to and from this centre that can be followed by others or in our future work. In this spirit, we used readily available hardware and software and detailed the methods so that it could be recreated and modified by others and used as an artistic tool in a variety of ways.
In conclusion, the birdsong Translator presented here ironically draws attention to the difficulties in translating between the private inexpressible experience of birdsong and an objective universal understanding of its meaning. Our approach was to create a speculative fabulation or string figure as a type of thought experiment, allowing us to investigate ideas of (un)translatability of non-human animal voices. The device can be envisioned as a gathering place for many strands, inviting new ways of thinking about our relationships on this fragile planet.