1. INTRODUCTION
The working life of the recording engineer lies in a liminal space, what David Beer calls a ‘precarious double life’. ‘This precarious double life’, he writes, ‘might be understood to exist as a result of the tension that is created as the recording engineer balances their practice at the lines between artistic sensibility and logistical or technical know-how’ (Beer Reference Beer2014: 189). The history and historiography of the development of the logistical and technical tools, of course, is well-documented; but what of the other side of our engineer’s limin? While the linguistic and communicative aspects of other – laboratory – fields of engineering (as performed by so-called ‘lab technicians’) has been thoroughly explored in the last decades in the intellectual currents that group themselves under the names ‘Actor-Network Theory’ (Latour) and ‘Social Construction of Technology’ (Pinch and Bijsterveld Reference Pinch and Bijsterveld2004), Beer points out that little work, either theoretical or ethnographic, has been undertaken to study that particular role of the recording engineer, beyond the work of Kealy, Porcello and especially Horning, to whom we will return later. As we will see, one reason for this may be the ‘bilingual’ or liminal nature of the engineer’s utterances; in any theoretical methodology based on the study of speech acts – as each of the above is, in one degree or another – a bilingual, or at least bi-modal, speaker complicates matters.
Unlike the ‘lab tech’, whose communication partners are generally others with similar backgrounds – likely in the hard sciences – the audio engineer faces a communicative difficulty, inasmuch as their communication partners are likely not themselves schooled specifically in the engineering aspects of sound. The need to constantly balance the ‘engineering’ and ‘artistic’ parts of the labour itself, so aptly studied ethnographically by Beer, is complemented and complicated by the requirement to constantly be ‘translating’ engineering concepts into musical concepts and vice versa. Especially on a larger project, the audio engineer may find themselves speaking one moment with a record producer with little to no technical experience but a clear artistic vision, and the next moment relaying instructions in highly technical terms to an engineer’s assistant to realise that vision.Footnote 1 It is a communication problem that is, in some very important senses, a kind of conceptual bilingualism, or code-switching.
In 1954, in the journal Word, Hans Vogt described bilingualism this way:
bilingualism is one of the major factors in linguistic changes … Code-switching in itself is perhaps not a linguistic phenomenon, but rather a psychological one, and its causes are obviously extra-linguistic. (Vogt Reference Vogt1954: 368)
The theoretical concept of code-switching, that is changing the language (or dialect) one is speaking as a result of psychological rather than linguistic factors, can help us understand the lexical manoeuvres performed by audio engineers as they traverse the music/technology or music/engineering boundary. Understanding this ‘double life’ as a kind of bilingualism allows us to apply Vogt’s contention that bilingualism is a major factor in linguistic change to an examination of how a once very technical terminology has come to have intuitive and metaphorical meaning outside the audio engineer’s legerdemain. In other words, conceiving the ‘precarious double life’ as one mediated by a bilingualism, we can trace the effect of this ‘language contact’ on the two languages that successful audio engineers must speak.
Since its development in the field of linguistics in the mid-twentieth century – and especially since the 1990s – the term ‘code-switching’ has taken on a variety of meanings beyond its original scope. Especially in the field of post-colonial studies, ‘code-switching’ has come to be understood as a political, in addition to linguistic, act, and to apply to switching between dialects, as well as between languages. Lourdes Torres writes that in literature ‘code-switching is an artistic choice with political ramifications. Using Spanish in an English language text serves to legitimise the much-maligned practice of mixing codes in vernacular speech’ (Torres Reference Torres2007: 75). This sociopolitical understanding of code-switching has generated and continues to generate a vast and variegated literature. Within that literature, Evelyn Nien-Ming Ch’ien’s Weird English carefully details code-switching in literature understood in this way. Tellingly, she details that code-switching expresses ‘a desire to create a new English that represents [the author’s] perspective in a world that seeks to marginalise the voices of these writers’ communities’ (Ch’ien Reference Ch’ien2004: 8). While the political dimensions of code-switching are not active in my study, Ch’ien’s observation that code-switching in literature and speech seeks to create a new English that represents a perspective from between communities echoes Vogt’s observation that code-switching is a psychological rather than linguistically functional phenomenon, and Beer’s observation of the liminality of the engineer’s work-life. Within the field of formal linguistics, however, code-switching continues to be studied without its political content. Even very recent work such as that by Kracht and Klein (Reference Kracht and Klein2014) addresses itself to the formal linguistic properties of code-switching without addressing the political dimension. Though the political reading of code-switching is much more common in fields such as ours, it is the narrow meaning, shorn of its political dimension that I will deploy. In sum, ‘code-switching’ as used here lacks the sociopolitical dimension it carries in post-colonial studies, because this study is examining the interplay between labour groups, rather than polities.
Alongside code-switching, Vogt discusses the familiar linguistic concept of a ‘loanword’, a new word appearing in one language due to a high degree of bilingualism in a given population, and carried by some mechanism out of that bilingual community into the wider community of speakers of just one of those languages. Vogt – Norwegian himself – was doing much of his work on various Nordic and Scandinavian languages, those languages closest to him. He gives examples primarily from Norwegian, but we need not stretch far to think of examples in English. The Yiddish ‘kitsch’, entering English through the highly bilingual – rarely only two, in fact – and particularly prone to code-switching Yiddish communities of the Lower East Side of New York City, entered the wider public consciousness through the explosion of Yiddish Theatre actors and writers ‘crossing over’ into mainstream Broadway and film productions and bringing a loanword with them (for more on the particular case of Yiddish, see Szulmajster-Celnikier Reference Szulmajster-Celnikier2005).
We will explore here a brief and necessarily incomplete history of the terminology devised by recording engineers for the description of sound, beginning in the mechanical recording era, and the code-switching and vectors that imported these loanwords into common usage. We will then examine the ways in which some technical terminology – especially regarding frequency and spectrum – have subsequently acquired metaphorical meaning almost-entirely divorced from their technical birthplace, and thus have gained the power to refer to sounds and concepts that are conceptually pre-technological in nature.
2. ELECTROMECHANICAL RECORDING
In a series of articles published in both general-interest and scholarly journals in the late 1920s and early 1930s, engineer Joseph P. Maxfield (then of Bell Labs) set forth a description of the application of concepts from electrical engineering to acoustic recording. In the first of these, he opens with an already-developed understanding of the dual character of the audio engineer: ‘A successful solution to the problem of phonograph reproduction requires an alliance between science and art’ (Maxfield Reference Maxfield1926: 104). Writing as he was in Scientific American, a general-audience publication, much of the article in question deals covertly with translating between scientific and musical terminology, often with less-than-precise results. After all, in a 1973 interview Maxfield very clearly stated: ‘I am an engineer, I am not a musician’ (Polkinghorn Reference Polkinghorn1973).Footnote 2
Maxfield devotes a large part of his 1926 Scientific American article to a discussion of the reproduction of harmonics, and what he calls a ‘flat’ frequency response. The response he describes is wholly inadequate by modern standards, but the language he deploys around it is striking. He begins by explaining why ‘the weaker sounds made during the act of performing music’ are important:
Since many of these sounds, such as the sibilants in speech, the intake of breath of the singer, and the touch of the bow of a stringed instrument are carried mainly by the higher frequencies, this portion of the musical range becomes important in the connection and is very important. (Maxfield Reference Maxfield1926: 104; emph. mine)
A few things are evident from this short excerpt. First of all, despite assuming a readership at least lightly versed in the concept of sonic spectrum (at no point in the article does he define what he means by ‘the higher frequencies’) he appears to feel the need to justify the inclusion of higher partials in recorded sound. Second, as an audio engineer he had clearly encountered the concept of musical range at some point in his work. But, at least today, it would be rare to refer to the intake of breath or the sound of the bow as a portion of the ‘musical’ range of an instrument. We could attribute this to a mere misunderstanding of the concept of musical range, but I argue that there is more involved here.
There is no point in the article in question where Maxfield refers to the ‘range’ of harmonics or frequencies of the recording system itself. In his characterisation, the system does things – reproduction and storage (ibid.: 105) – but at no point appears to have any quantitative qualities outside of a ‘design’. The difference between recording systems are defined in terms not of differences between ‘the systems’, but of differences between ‘the designs’. While this may seem a minor difference, it is telling.
In modern audio engineering discourse, the ‘frequency range’ of a recording medium is of paramount importance, of course, and we treat the standards of our media – especially digital media – as fixed by the medium itself. Various standards have been developed and are hewn to closely in studios around the world. However, for Maxfield, the recording medium was not a fixed-enough system to be attributed a range that belonged to it, even as his articles are concerned with carefully describing such a system. We can see here the concept of ‘range’, at first a musical term for the ambitus of an instrumental or voice part, starting to become a loanword borrowed into audio engineering discourse, but here only partially. We will see at the end of this short study what effect that might have on the re-borrowing back into musical or metaphorical discourse.
While a complete discussion of the system described in Maxfield’s Reference Maxfield1926 article is well beyond the scope of this survey, there are a few other details of note in the Scientific American article. First, the technical ‘meat’ of the article is the application of an electronic metaphor to the construction of a purely mechanical reproduction system. Maxfield refers to ‘designing the acoustic transmission system as the analogy of the proper electrical system’ (ibid.). What shall we assume that ‘proper’ means in this statement? I would argue that we are seeing here the beginning of a discourse about fidelity that prizes new technologies over older ones. In fact, Maxfield even describes the system in terms of both a physical ‘cutaway’ figure and the electronic circuit analogue.
The final section of Maxfield’s Reference Maxfield1926 article is headed ‘Harmonics reproduced in Wide Range’. There are three improvements to the reproduction and recording of sound that the author suggests his system provides.
In the first he lists, ‘the reproduction of the bass and the higher harmonics’ (ibid.) (what we would now call increased frequency range). Interestingly, this is the first time in the article that the expanded bass frequencies are mentioned, and Maxfield turns immediately to using the electroacoustic terminology of frequency and spectrum to describe the value of the bass frequencies: ‘The reproduction of the bass, that is, the lower frequencies, adds the “body and weight” to the music’ (ibid.: 105). We will turn at the end of this article to the metaphorical application of ‘bass’ as refers to the body and to weight, but note now that this embodied discourse around what we would now call ‘low frequency sound’ was already active by the time frequency ranges were being discussed between the two world wars.
The second improvement listed, which took up a fair bit of the original article but I left out in my gloss, is issues around ‘atmosphere’ and reverberation time. This discussion is vital in a discussion around fidelity, but – as it is undertaken almost devoid of electroacoustic terminology, a surprise in itself – has little bearing on the argument here. ‘The third improvement relates to loudness’, Maxfield writes. Here again the author conflates, or rather lacks the musical terminology to disambiguate, ‘note’ and ‘frequency’. Drawing our attention to a figure representing the frequency response (almost certainly calculated rather than measured, though Maxfield is not clear on the subject) of ‘the old and new types of phonograph’, he notes that ‘the notes that were reproduced most loudly on the old machines are not as loud on the new machine as they were on the old’ (ibid.).Footnote 3
Again, we cannot be sure if Maxfield’s conflation of spectral content and ‘note’, like his conflation of spectral content with ‘range’, is a concession to a readership he assumed – possibly correctly – to be devoid of musical knowledge or whether it represents an as yet undeveloped stage in the audio engineer’s bridge between musical and scientific terminology. Regardless, what we see here is a nascent stage in the process of the scientific language of spectrum, of ‘frequency content’, moving its way as a loanword into musical – including of course non-electronic – discourse. The next great leap in this loanword process took place in the era of the hi-fi.
3. HI-FIS AND TONE CONTROLS
Even the earliest home listening devices with electronic, rather than mechanical, reproduction contained equalisation circuits. The RIAA curve, a standard spectral modification, was applied to every US recording by 1949. (The history of these standards is beyond the scope of this article, but is quite interesting and under-theorised.) This ‘treble pre-emphasis, bass roll-off’ curve was designed to compensate for the mechanical work of the cutting lathe in disk manufacturing, and a compensatory circuit was needed in the consumer’s playback devices to restore the ‘neutral’ spectral balance of the recording (Howard Reference Howard2009) (Figure 1).
Figure 1 The RIAA equalisation curve.
Importantly, this spectral modification circuit (‘filter’) was not alterable by the consumer. Until the late 1950s there was no consumer expectation of tonal control of the sound output. The engineer, producer and record company had control over the final output of the sound, and the consumer? The consumer consumed.
By the late 1950s (ibid.) manufacturers, first of high-end then of mid-range audio playback devices, begain to include special circuits that allowed the consumer a degree of control over the spectral content of the output of their reproduction system. While originally designed to compensate for less-than-ideal room acoustics and to be as simple as possible, we can easily see how these ‘tone’ knobs, limited at first to one or two, fundamentally altered the place of the audio engineer’s discourse in the wider discourse around recorded music, from the consumer’s perspective.
Contrary to Maxfield’s emphasis on ‘fidelity’ as seeming ‘present with the performer’ (Maxfield Reference Maxfield1926: 105), the hi-fi era ushered in an era of ‘consumer control’, despite being still wedded rhetorically to the notion of ‘fidelity’ (Hamilton Reference Hamilton2003: 350). And fundamental to this control was a democratisation – or at least popularisation – of electroacoustic terminology away from the recording studio. After all, if one gives the consumer a knob, one must label it. And the engineers building the reproduction system are suddenly put in the position of translator: not face-to face as of speech, but translating a text – what does the knob do – into another text – what shall we tell the consumer this knob does? Such level of control became such a feature of hi-fi systems by the 1970s that it became a selling point in its own right, leveraged to appeal to the usually male, usually affluent demographic expected to purchase expensive hi-fi equipment (Figure 2).
Figure 2 A hi-fi advertisement from the 1970s.
It fell to audio engineers not of the recording-type but of the consumer electronics-type to determine the function and labelling of these knobs. Function is essentially a technical matter, but also – in conjunction with product designers and others with less understanding of their function – these engineers had to name them. The labelling, unlike the circuit design, is fundamentally a speech act. Like Maxfield’s justification of his improved recording system, audio engineers needed to talk to product designers (the analogues in this structure of the record producer above) to translate from one domain of experience to another.
Horning describes an analogue to the visual thinking central to design engineering praxis, one that she calls ‘aural thinking’. As a kind of non-verbal cognition, ‘[a]ural thinking’, she writes, ‘evolved as the technology made it possible. It developed in concert with significant technological shifts’ (Horning Reference Horning2004: 714). If the aural thinking of the audio engineer is then non-verbal, or logically prior to verbalisation, being forced by the fact of consumer design to put a verbal label on a non-verbal idea forces the audio engineer into an act of code-switching: to try to find the right word in a verbal language for a non-verbal idea.
Various firms developed various linguistic solutions to describe the spectral modifications they allowed the consumers to make. Many chose the ‘bass’, ‘mid-range’ or ‘treble’ labels, which have clear musical analogues, but do not, in fact, bring out the bass or treble instruments but rather the bass or treble frequencies, and therefore differ fundamentally from the purely musical usages of the terms. Others chose more metaphorical terms to describe the same circuits; the JBL speakers above the workbench in my office contain two tone controls, marked ‘warmth’ and ‘presence’, for example. A quick perusal of the user’s manual will invariably reveal the technical changes that these knobs are making to the output spectrum of the playback device, but the choice to label them as such remains an interesting one.Footnote 4 The ‘double life’ of the engineer here is going both ways; where Maxfield was primarily interested in loaning scientific terms to musical discourse, these myriad engineers were loaning musical terms to scientific discourse. The proliferation of the ability of consumers to alter the tonal output of their systems, to take on the role of the audio engineer in practice if not in training, was a powerful vector for the introduction of a once-specialised vocabulary into common discourse. If nearly everyone with a cheap record or cassette player had ‘tone controls’, nearly everyone with a cheap record player would someday play with them. In so doing, the wider listening public of the 1960s, 1970s and 1980s became familiar with a terminology of spectrum that, in 1926, needed to be painstakingly explained to readers even of a scientific magazine. As we might expect, as these electroacoustic loanwords made their way via the electronic listening experience into the wider culture, they proceeded to accrete a range of metaphorical meanings.
4. METAPHORS OF TONE
What began as a ‘precarious double life’ for the audio engineer had become a ‘precarious double life’ for some electroacoustic terminology by the late 1980s. Increasing consumer/public familiarity with electroacoustic loanwords as applied to musical reproduction led consumers to further loan these terms into everyday discourse, where meaning could slip into metaphor.
Because sound is invisible, it lends itself to metaphorical description, and this process goes both ways (Hamilton Reference Hamilton2003: 355). In a deft analysis in her article on tacit knowledge, Horning describes an event in which audio engineer Clair Krepps, needing to mute or ‘baffle’ one side of a bi-directional microphone without isolating the singer, ran out to a drugstore and ‘bought a box of Kotex’ to tape to one side of the mic. She writes, ‘[o]ne might speculate about how Krepps came up with this solution. Had he made a word association with the idea of “leakage” or “signal bleed”?’ (Horning Reference Horning2004: 713). This sort of process, in which an experienced technical aspect of sound – signal leakage in this case – works backward into a metaphorical relationship, was once confined to the tacit or explicit languages of audio engineers. As we have seen, however, bestowing upon the consumer some of the explicit terminology of the engineer allowed this process to spread into wider discourse. One such example can be found in two seminal hip-hop tracks of 1989.
Biz Markie’s 1989 single ‘Just a Friend’, from the album The Biz Never Sleeps, which chronicles a flirtation gone wrong, contains the line (describing the beginnings of said flirtation) ‘I started throwin’ bass, she started throwin’ back mid-range’ to describe their conversation. Similarly, the Beastie Boys single of the same year, ‘Hey Ladies’, from the album Paul’s Boutique, contains the similar line ‘Some such nonsense is the bass that I’m throwing / Talking to a girl telling her I’m all knowing’ to describe a similar – though more successful – flirtation.
Given the temporal and geographical proximity of these two references – Biz Markie from Long Island, and the Beastie Boys from nearby in Brooklyn, New York – it seems logical to assume that this is more than a case of independent discovery. It seems reasonable to assume that ‘throwing bass’ as a synonym for flirting was in some manner of wide use in the mid-Atlantic region in the late 1980s. But what accounts for the application of this particular acoustical term ‘bass frequencies’ being associated with the particular speech act of flirting? All the way back in 1926, Maxfield gave us our clue.
We saw Maxfield say above that ‘the bass, that is, the lower frequencies, adds the “body and weight” to the music’. This association with the body may be the vector by which ‘bass’ came to acquire its metaphorical meaning. Whereas Maxfield was attempting to use the metaphor to describe to an audience untutored in music the importance of the bass frequencies, over 60 years later the tables have entirely turned. Biz and the Boys’ metaphor – whether in common usage within their audiences or not – assumed a familiarity with the aural effect of the bass frequencies; an intuitive understanding that the bass frequencies add ‘body and weight’. Markie even further develops the metaphor, when his intended paramour starts ‘throwing back mid-range’, a clever turn of phrase that at once tells us that his target is flirting as well, and points to the timbral differences between men’s and women’s voices (especially Biz himself, in possession of a quite deep speaking voice). We see in this metaphor, too, that the songwriter expects his audience to possess an understanding of the relative relationships between ‘bass’ and ‘mid-range’ frequencies, one that could not be readily expected of an audience in the 1930s or 1940s, before the hi-fi made manipulating these aspects of sound an activity available outside the audio engineer’s domain.
If the bass frequencies metaphorically reference the body, then, to use ‘throwing bass’ as a metaphor for flirtation implies a throwing of the body towards the intended target. Bass frequencies, once managed in studio by a select group of experts in electroacoustic engineering, have escaped the studio, gone native and started trying to make friends.
While the term ‘throwing bass’ as a generally understood term for flirting may have passed from common discourse – at age 37 I have never heard it used that way outside of late 1980s hip-hop – the association of the bass frequencies with the body and warmth continues unabated. Meagan Traynor’s 2014 debut single ‘All About that Bass’ uses the bass frequencies as a metaphor for the female body itself, in an attempt to promote an image of body-positivity and combat fat shaming.Footnote 5 The ‘bass’ here represents the ‘body’ and ‘weight’ of the female form itself, and the ‘treble’ represents an undesirable skinniness. One wonders, confronted with this slightly different usage of ‘bass’ metaphorically, whether the process of associative slippage that Horning applies to ‘sound leakage’ above can account for the continuing association of the bass frequencies with the actual human body – as distinct from the ‘body’ and ‘weight’ of the sound as Maxfield described it. The homophone between ‘base’ and ‘bass’ might account for the bass frequencies ‘stickiness’ with regards to this particular metaphorical usage: ‘base’ providing an impression of solidity and steadiness that ‘bass’ can then pick up, using ‘base’ as a bridge to attach the acoustic terminology of ‘bass frequencies’ to something physical and real: the body.
5. CONCLUSION
Frequency terminology is only one small domain of the audio engineer’s language that has provided loanwords as metaphor into broader discourse. A Google Ngram search, which charts relative frequency of the appearance of a word or phrase over time in a large corpus of books and magazines, reveals that the phrase ‘scratchy sound’ was nearly unknown in English before the rise of the phonograph – despite string performers being described as ‘scratching away’ with their bows as early as the eighteenth century. Similarly, the adjective ‘tinny’, as applied to sound, generally describes a frequency response lacking in low frequencies and with a pronounced peak in the high frequencies, as one might get from the tin horn of a phonograph. In fact, as an engineer, if I were asked to make something sound ‘tinny’, I would almost certainly apply the frequency curve from Figure 1.
The ‘precarious double life’ of the engineer, using their bilingualism or code-switching to mediate between the worlds of science and music, provides an ideal vector for the transmission of specifically electroacoustic – or even merely acoustical – terminology into general usage. Once safely ensconced in wide use, these terms are free to take on new, metaphorical meanings and apply once-specialised vocabulary to referents that are non-technological, pre-technological or even corporeal in nature.
