Sound and vision blog

Amongst the literary treasures held in the basements of the British Library sits an extraordinary collection of sounds.  From recordings of extinct species, voices from the past, to music across all genres, the British Library’s sound archive is held on more than 1.5 million physical items, just waiting to be heard.

But all of these recordings, from those made on the earliest wax cylinders to contemporary CD-Rs, face a real and immediate threat.

Edison 'Concert' wax cylinders in the collections of the British Library

Within 15 years, the combination of physical degradation and the disappearance of the technologies that support physical media will make accessing the nation’s sound archive difficult, and in many cases impossible.  Without taking steps to preserve these recordings now, they will be lost.

These risks face all recorded sound collections, across the country; from boxes of forgotten cassette recordings to professional archives.

To understand the risks facing the UK’s sound collections, the British Library has initiated a project to collect information about our recorded heritage, to create a Directory of UK Sound Collections.

By telling us what you have, we can understand more about the breadth of the nation’s collections and the risks that they face, and this will help us plan for their preservation, for future generations.

Our aim is to be comprehensive; to search out sounds that exist in libraries, archives, museums, galleries, schools and colleges, charities, societies, businesses and in your homes.  And we’re not just interested in large collections: a single item might be just as important as a whole archive.   

So if you think you might have a rare or unique collection of sounds, or just a recording that should be preserved, let us know!

The census is live now and will run until the end of March 2015.  You can read more about the project, and send us information about your collections here: www.bl.uk/projects/uk-sound-directory

Responses have already started to come in, and we’ll be publishing updates on the project, and some of the things we’ve found on this blog, so enter your email address and click the Subscribe button at the top of this page to receive notifications by email.

The British Library’s Directory of UK Sound Collections is one of the first steps in our Save our Sounds programme; one of the key strands of Living Knowledge, the British Library’s new vision and purpose for its future.

You can follow the British Library Sound Archive on Twitter via @soundarchive and tag with #SaveOurSounds

Related articles

Save our Sounds: 15 years to save the UK's sound collections

The UK’s audio collections are under threat. Often created on media that over time has become unstable, they are at risk - both from physical degradation and from the obsolescence of technology to replay them.

Professional consensus is that we have approximately 15 years in which to save the UK’s sound collections before they become unplayable and are effectively lost. The solution is to digitally preserve them, but the scale of the task required is considerable and time is running out.

Digitisation of a 1960s magnetic wire recording

The British Library is home to the national sound archive, an extraordinary collection of 6.5 million recordings of speech, music, wildlife and the environment from the 1880s to the present day. We need both to ensure that the existing archive is properly preserved and that new systems are developed for the acquisition of future sound production in the UK.

The voice of Florence Nightingale recorded 30 July 1890 on wax cylinder

Save our Sounds is a new programme the British Library has created to answer this imperative need. It has the following aims:

•    to preserve as much as possible of the UK's rare and unique sound recordings, not just those in our collections but also key items from other collections across the UK
•    to establish a national radio archive that will collect, protect and share a substantial part of the UK’s vibrant radio output, working with the radio industry and other partners
•    to invest in new technology to enable the Library to receive music in digital formats, working with music labels and industry partners to ensure their long-term preservation.

You can listen to the kinds of sound recordings that we are preserving for the UK, by visiting our Sounds website at http://sounds.bl.uk. It has a selection of 60,000 sound recordings for all to enjoy, covering the entire range of recorded sounds: music, accents and dialects, drama and literature, oral history, wildlife and environmental sounds.

This playlist of a dozen tracks starts with an early recording of a Beethoven piano trio, one of the earliest recordings of chamber music, recorded in 1905. The violinist we hear performing was born just 30 years after the  composer's death in 1827.

We have launched the Save our Sounds programme with a major fundraising campaign to digitise and digitally preserve the most fragile and unique recordings. Save our Sounds is one of the key strands of Living Knowledge, the British Library’s new vision and purpose for its future, news of which was announced today by the Library’s Chief Executive, Roly Keating.

Sound recordings document some of our greatest creative endeavours, preserve key moments in our history, capture personal memories, give a sense of local and regional identity and they help us to understand the world around us. And they are extraordinarily powerful in bringing back to life past events: famous speeches, the voices of loved ones and those who have sadly left us, musical and other artistic performances, notable events in recent history and the familiar and exotic sounds of natural and urban environments.

We need to preserve sounds today - to listen to the past tomorrow. So it’s vital that we act now to ensure they are accessible for future generations.

Richard Ranft, Head of Sound and Vision

How to get involved:

• We are mapping the condition of sound archives around the UK to identify threatened collections – if you have a sound collection which you think could be at risk, get in touch and let us know. The census will run until the end of March 2015, and we’re keen to hear from those with private collections as much as the public and commercial archives out there. All you need to know is at www.bl.uk/projects/uk-sound-directory.
• We’ll be issuing updates on this blog, so enter your email address and click the Subscribe button at the top of this blog page to receive notifications by email.
• You can follow us on Twitter via @soundarchive – and do make use of the hashtag #SaveOurSounds
• Please consider your organisation making a major gift to support our programme: http://support.bl.uk/Page/Save-Our-Sounds

More information:

1. Save our Sounds website
2. UK Sound Directory
3. Living Knowledge: The British Library strategy 2015–2023
4. Launching Living Knowledge 12 January 2015
5. “Sounds of the past, ways of the future” (Financial Times, 9 January 2015).
6. British Library seeks £40m to 'save' sound archive (BBC News, 12 January 2015)
7. British Library recording of Florence Nightingale, originally made in 1890.
8. Rare Noël Coward recording rediscovered

Related articles

Seven unique voices saved by the British Library

British Library Fights To Preserve Historic Recordings Archive

 Jonnie Robinson, Lead Curator of Spoken English, writes:

Last weekend I went to Dulwich to watch my daughter play hockey, which gave me the opportunity of exploring Dulwich village for the first time. The highlight of a thoroughly pleasant stroll was stumbling across a rather intriguing use of the word chuffed [= ‘pleased’] – a term used by millions of speakers of British English on a regular basis. Chalked up on a blackboard inside a wonderful artisan baker’s was a sign thanking customers for nominating the proprietors for a local trade award (I hope they win: the bacon bap I had was delicious). The sign declared that the owners were very chuffed to be nominated; an expression that immediately struck me as slightly odd - do people actually say very chuffed? Isn’t very somehow just too mainstream to combine with a word like chuffed? Aren’t more colloquial intensifiers like really, pretty and so or vernacular forms such as dead chuffed, proper chuffed and well chuffed more natural?

A quick glance at several authoritative reference works seems to confirm my hunch. The entry for chuffed in the Oxford English Dictionary (online edition) includes six citations - the earliest from 1957 - and two examples each of chuffed, pretty chuffed and dead chuffed.  The New Partridge Dictionary of Slang and Unconventional English (2014) agrees with the OED's dating and also includes the following observation:

originally northern English dialect […], adopted by military, then wider society […] often qualified by intensifiers DEAD, REAL, WELL

Both dictionaries include the antonym dischuffed – presumably formed by analogy with pleased/displeased rather than happy/unhappy – and the Dictionary of Contemporary Slang (2014) provides further support for Partridge stating that chuffed:

probably originates in northern English dialect […] and is still most frequently heard in the North and Midlands [...] embellished forms are ‘dead chuffed’, ‘chuffed pink’ and ‘chuffed to arseholes’

In 2004/5 the BBC Voices survey investigated the words we use for 40 everyday concepts, including the notion PLEASED. Researchers in the British Library’s Voices of the UK project are currently compiling an inventory of the terms captured in the study and have thus far catalogued over 100 variants for PLEASED. Apart from pleased itself, chuffed was by far the most common response and certainly seems to have been taken up enthusiastically outside its northern and midland heartland, but as far as I’m aware we haven’t encountered many – if indeed any – examples of very chuffed. Plenty of contributors supplied dead chuffed, well chuffed, chuffed to bits, chuffed to naffy break (also in Partridge) and even chuffed to buggery, but not very chuffed. And yet, by extraordinary coincidence this week a contestant on the BBC quiz show Only Connect said he had been very chuffed with his team's performance in the previous round. I dunno - you wait for ages for a very chuffed and all of a sudden two come along at once.

If you'd like to hear any of the numerous  variants for PLEASED just listen to one of the 300 BBC Voices Recordings. From thrilled, delighted, tickled pink, cock-a-hoop and on cloud nine to made-up, thrimmed, over the moon, baktalo (Anglo-Romani for ‘happy/lucky’) and stoked each gives subtle clues to a speaker’s geographic background, age, ethnicity and/or social status.

My daughter’s team lost by the way, although in scoring her first goal of the season I suspect she was chuffed and dischuffed in equal measure, but – all things considered –  probably not very chuffed.

Jonnie Robinson, Lead Curator for Sociolinguistics writes:

This month we've uploaded linguistic descriptions of conversations about local speech in Coxhoe, Hartlepool and three discussions in Middlesbrough (one with a group of young friends, one with rival football fans and one with three generations from the same family). Together they constitute the set of BBC Voices Recordings made by BBC Radio Cleveland. The descriptions list the participants' responses to a set of prompt words and, in the case of Coxhoe and the friends in Middlesbrough, also include a detailed description of the phonology and grammar of the speakers.

Teesside was the focus of considerable linguistic debate last summer when BBC business correspondent, Sophie McGovern, revealed she received regular complaints from viewers about her Middlesbrough accent. This followed reports earlier the same year that a Middlesbrough primary school had solicited parental support in addressing teachers' concerns that local dialect features were compromising children's formal written English. Personally I've always enjoyed the accents and dialects of the region and local celebrities such as Bob Mortimer (from Acklam, Middlesbrough), Vic Reeves (from Darlington, County Durham) and Mark Benton (from Guisborough, North Yorkshire) are extremely popular figures whose speech patterns are, I sense, an integral part of their appeal.

The backgrounds of these three entertainers hints at the competing cultural and linguistic influences in the area as Teesside straddles the historic counties of Durham and North Yorkshire - Middlesbrough itself was historically in North Yorkshire. Successive changes in administrative nomenclature (North Riding, North Yorkshire, Teesside, Cleveland) have been hotly contested locally as confirmed by this Middlesbrough family and this repeated upheaval underpins the study of Middlesbrough dialect by academic linguist, Carmen Llamas, who explores how local linguistic features reflect speaker identities across generations.

Intriguingly the Middlesbrough accent is frequently confused with accents further up the east coast in Newcastle upon Tyne and, perhaps more surprisingly, on the west coast in Liverpool. Similarity with speakers in Tyneside is most apparent in a shared tendency to glottalise the sounds <p>, <t> and <k> between vowels in words like happy, better and lucky. This distinctive pronunciation is a recent innovation on Teesside thought to be influenced by increasing identification - especially among younger speakers - with the North East. The equally striking pronunciation of word final <t> in words like it and that (pronounced as if it were a <h> sound) and the vowel sound used in words like first, bird and turn is probably more widely associated with English on Merseyside and thought to reflect the earlier influence of large-scale migration from Ireland to Teesside - and Merseyside - in the nineteenth century. All three phenomena can be heard frequently in the recording with former Middlesbrough schoolfriends.

What is also clear from this set of recordings is that the area remains lexically distinctive with examples of 'pan-northern' vocabulary (e.g. lass [= 'girl']), words shared with the North East (e.g. netty [= 'toilet'] and cuddy-wifter [= 'left-handed']) and with Yorkshire (e.g. mafted [= 'hot']) and iconic Teesside items (e.g. nick off [= 'to play truant']). The emergence in Middlesbrough of a second person plural form yous is also shared with other northern cities (notably Liverpool, Newcastle and Glasgow), while the recording in Coxhoe is particularly rich in typical local discourse markers:

0:07:31 by you're all proper from Cassop man

0:17:34 saying that mind [...] they certainly have interest when it comes round to the presentation evenings

0:19:56 you used to get great crack to them how but

0:35:33 that's just as old-fashioned as the hills like

0:36:48 why let them start their own clubs

The British Library has recently acquired  a collection of field recordings made in South Sudan which document Dinka song culture. Songs of the Dinka of South Sudan - Diɛt ke Jiëëŋ ne Cuëny Thudän - can now be accessed via our catalogue, by searching for C1580, and listened to online.  Dr. Angela Impey, one of the researchers on the project, has written the following text which contextualizes the research project and gives some general information on Dinka culture:

The songs in this collection were recorded for a project entitled Metre and Melody in Dinka Speech and Song, which was conducted between 2009 and 2012 by researchers from the University of Edinburgh and the School of Oriental and African Studies, University of London, in collaboration with Dinka researchers in South Sudan. The academic aims of the project were, first, to understand the interplay between Dinka song structures and the Dinka language (which distinguishes words not just by different consonants and vowels but also by means of vowel duration, pitch and voice quality), and, second, to learn more about the song tradition and the ways it has responded to the intense disruptions caused by protracted civil war.  

Funding for the project was provided by the ‘Beyond Text’ programme of the UK Arts and Humanities Research Council. Project participants included Bob Ladd, Angela Impey, Bert Remijsen, Elizabeth Achol Ajuet Deng, Peter Malek, and Simon Yak Deng Yak.

Photo credit: Robin Denselow

ABOUT DINKA SONG

Almost everyone in Dinka society will accumulate a repertoire of personal songs during their lifetime, and most dialect groups follow a similar compositional process. Individuals who lack the ability to compose good songs will approach a talented composer in the community and commission a composition in exchange for a cow or an agreed sum of money. Occasionally a composer will be considered a talented lyricist only, in which case a second individual, who has an aptitude for good melody making, will be brought into the process. Upon completion, the song will either be taught directly to the ‘owner’, or if the owner is not a good memoriser, via a group of relatives or age-mates, who will gradually pass it on to the owner.

025A-C1580X0059XX-0001A0

[In this ox song (C1580/59), Deng Jok Ajuoong, praises his ox, Mading, which he compares with an elephant. He sings about how he acquired his ox through hard work.]

Most musical structures in sub-Saharan Africa are based on highly repetitive, multi-part vocal and rhythmic interactions, and melodies are typically based on the hexatonic (six tones per octave) or equi-heptatonic scales (seven tones equally distributed across the octave). In contrast, Dinka tuning systems follow a standard pentatonic scale (five tones) and songs are composed in an extended series of linear, interconnected song-segments that follow a simple, regular or semi-regular pulse. Certain song types are accompanied by clapping, clapping sticks or a small double-sided drum (loor), and are performed either solo, in unison or in simple call-response format. The only melodic instrument played by the Dinka (apart from more recently introduced western instruments) is a 5-stringed lyre referred to as rababa. Marked aesthetic variations do occur across the dialect groups, which are likely to be the result of different social, economic, environmental and political circumstances.

025A-C1580X0006XX-0001A0

[This song (C1580/6) is performed by a women's group during the war in South Sudan. It is an encouraging song about Dr. John Garang and Koryom, a battalion of the Sudan People's Liberation Army (SPLA). The women start the song by singing “The Arabs said we are afraid, how can we be afraid while John Garang is strong?”]

Apart from one book (available in British Library Reading Rooms) published on song lyrics by Francis Mading Deng in 1973, no formal research has been conducted on Dinka music. Yet songs play a fundamental role in the lives of all Dinka people, functioning as individual and social chronicles of relationships, experiences and historical events. In fact, the Dinka boast an usually complex taxonomy of songs – praise songs, war songs, songs of initiation, cathartic songs, religious songs, to name a few – each of which is defined by discernible melodic, rhythmic and performative features. Woven through all song types, however, is the poetic allusion to the interconnection between self, cattle and land or locality. As one musician explained: If you know our Dinka songs, you will know the Dinka people.

Jonnie Robinson, Lead Curator for Sociolinguistics writes:

This month we've uploaded linguistic descriptions of conversations about local speech in Cambridge (one with two young hip-hop artists and one with regulars at a pub), Ely, Peterborough and Wisbech. Together they constitute the set of BBC Voices Recordings made by BBC Radio Cambridge. The descriptions list the participants' responses to a set of prompt words and, in the case of Ely, also include a detailed description of the phonology and grammar of the speakers.

Unlike previous dialect surveys, the Voices project intentionally targeted speakers from a wide variety of backgrounds and these five recordings from Cambridgeshire represent a particularly diverse set of speaker groups. In Wisbech you can hear localised pronunciation features such as yod-dropping on commuters ['kuhmooters'] and stupid ['stoopid'] and dialect words such as docky [= packed lunch]. Two young Cambridge musicians provide examples of multi-ethnic urban youth slang such as spar [= friend], safe [= great, excellent] and crisp biscuit [= attractive], while elsewhere in Cambridge we hear evidence of polari expressions in bona capella [= nice hat] and I'm gonna get me riah [= hair] done. There's also insight in Peterborough into the code-switching that's typical of bilingual British Asians in forms like khabbu [= left-handed] and the list of pairs used to distinguish between paternal and maternal grandmother such as dadi and nani.

The Voices archive also includes a number of speakers of Received Pronunciation (RP) - the middle-class, geographically neutral accent of England and, to a lesser extent, other parts of the British Isles. Contrary to much recent popular and media opinion RP is neither out of date nor disappearing and has in fact always encompassed a range of speech types. The Voices data set includes examples of upper-class RP in recordings with an aristocratic family in Ingatestone and members of the Arnesby Hunt. RP as manifested in the Armed Forces is captured in Worston, while typical middle-class RP is represented by a group of female golfers in Sevenoaks and a recording with three generations of a long-established family of Devon landowners in Newton St Cyres. Or you can explore Public School RP by comparing former boarding school female friends in Clapham with young students here in Ely.

In common with all accents RP is changing. The recording in Ely, for instance, shows clear evidence of change on one feature - the pronunciation of the <t> sound between vowels.  As young speakers of British English it's not surprising to hear several examples of glottal stops, but pronunciations with a <d> sound occur pretty frequently too - the word little, for instance, occurs both with a glottal stop and with <d>. Although a pronunciation with <d> is  perhaps more readily associated with speech in the USA, it's actually a long-established feature of  RP and other British accents and the presence of both alternatives here shows how adolescent speakers fluctuate between a more 'conservative' <d> and  the innovative variant with a glottal stop.

Jonnie Robinson, Lead Curator for Sociolinguistics writes:

This month we've uploaded linguistic descriptions of conversations about local speech in Bristol, Knowle West, Dulverton, Ilminster and Wellington. Together they constitute the set of BBC Voices Recordings made by BBC Radio Bristol and BBC Somerset Sound. The descriptions list the participants' responses to a set of prompt words and, in the case of Knowle West and Dulverton, also include detailed descriptions of the phonology and grammar of the speakers.

A unique feature of Bristol dialect is 'parasitic L' (or 'Bristol L' as it's often called in popular descriptions). This refers to the process whereby a word that in most accents ends in a weak vowel - e.g. area, idea and cinema - is pronounced in Bristol dialect with a word final <l> - i.e. to sound like 'areal', 'ideal' and 'cinemal'. This feature is often caricatured in stereotypical portrayals of Bristolian speech, but it's likely very few people have heard an authentic example. You can hear several spontaneous examples in the conversation in Knowle West here, including:

0:41:11 I can remember the first time I had a bananal [= 'banana']

0:41:45 I know at one time he came home ... and our ma was stood at the living room windle [= 'window']

The second example above is particularly striking as it reveals a two-step phonological rule: firstly the final syllable of window is interpreted as containing an underlying weak vowel - i.e. 'winda'. This is a pretty widespread phenomenon in speech across the UK (and elsewhere) as confirmed by common pronunciations like 'fella' [= fellow], 'borra' [= borrow] and 'marra' [= marrow, friend]. There's a platform announcer at Kings Cross underground station in London, for instance, who on a daily basis alerts passengers to Metropolitan Line trains stopping all stations to 'arra' [= Harrow-on-the-Hill]. In Bristol this process creates an environment where parasitic L can occur - hence window > winda > windle. You might think, therefore, that pairs like idea and ideal or area and aerial are indistinguishable in Bristol dialect, but in fact speakers who use Bristol L invariably convert word final <l> to a weak vowel so that aerial is pronounced 'area' thus maintaining the distinction. Again you can hear evidence of this in the conversation in Knowle West such as: 

0:03:26 I lived at where the old origina Whitfield tabernaca was

Indeed the name Bristol itself (historically Bristow but re-interpreted locally and subsequently nationally as Bristol) is a tribute to this wonderfa loca feature that in my humba opinion makes Brista my ideal of an idea dialect areal.

Jonnie Robinson, Lead Curator for Sociolinguistics writes:

This month we've uploaded linguistic descriptions of conversations about local speech in Bradfield, Purley on Thames and three recordings in Reading (one with work colleagues, one with young British Asians and one with a Bajan family). Together they constitute the set of BBC Voices Recordings made by BBC Radio Berkshire. The descriptions list the participants' responses to  a set of prompt words and, in the case of the Bajan family and Purley on Thames also include detailed descriptions of the grammar and phonology of the speakers.

It's fascinating to see how these two apparently unrelated recordings illustrate the same linguistic process: dialect shift. Not surprisingly, the Caribbean heritage of speakers born in Barbados, but now living in Reading, is reflected in their accent. They all use several pronunciations here that are typical of Bajan English, such as the so-called GOAT vowel in words like stone, road and know. The younger generation - born and brought up in Reading - vary between this pronunciation and a more obviously southern British variant. Listen, for instance, to Kevin fluctuate between the two variants in the same utterance:

1:18:46 yeah, you call him a  'poser', "he's a poser, man, look at he posing with cheap gold and fake gold and designer clothes"

His initial pronunciation of the word poser contains a typical southern English vowel sound, but he subsequently uses a Bajan-like vowel on the second instance of poser, posing, gold and on the word clothes. This illustrates perfectly how dialect contact produces incremental change within a single family - one pronunciation is used consistently by older speakers, but competes in the next generation with a more dominant and/or socially prestigious variety in the community.

The recording in Purley allows us to observe the same process of dialect shift over three generations, albeit in the context of a well-established local family and on a  different linguistic variable: rhoticity - i.e. the presence or absence of a <r> sound after a vowel in words like start, letter and nurse. The oldest speaker (b.1928) invariably pronounces this <r> sound here, while his son (b.1947) varies between including and omitting <r> in this environment.  In contrast the two younger speakers (both born in the 1970s) consistently omit this <r> sound, showing that the loss of postvocalic <r> is completed over two generations in this family.

Although this process of change occurs on different pronunciation variables in each recording (and is presumably prompted by different reasons), it nevertheless shows how a linguistic feature gradually 'shifts' in prominence across successive generations of the same speech community. For speakers like the Bajan family who have relocated to a completely different country, the older speakers are outnumbered by speakers of the local dialect so the younger generation naturally accommodates towards their local peers. The loss of rhoticity in the Purley family, however, reflects a trend that has been noted across southern England over the last century, and can be attributed to continuous waves of migration of non-rhotic speakers, especially from London, out into the surrounding Home Counties like Berkshire.