THE FLICKERING LEXICON
A list of incorrect things
Dannyno’s Concordance to the Song Lyrics of The Fall
‘This work has not yet reached cessation' - 2020 note: I will publishing a revised version shortly!
Credit: www.wordle.net, based on words in Fall songs. Common English words omitted
Thematic Analysis (i.e. lists of words by subject)
Welcome to The Flickering Lexicon, a valiant but flawed – doomed, even – attempt to construct a concordance to the song lyrics of The Fall.
Wikipedia defines a concordance as “an alphabetical list of the principal words used in a book or body of work, with their immediate contexts.” Concordances are a widely-used tool in linguistics, allowing researchers to compare uses of a word, study word frequencies, and so on. I claim no relevant academic credentials myself, but there is no reason why The Flickering Lexicon could not be pressed into scholarly service. However, I imagine my main audience is probably going to be fellow Fall obsessives.
Fall fans love lists (except for the Fall fans who really really hate lists), and concordances are great for list-makers. The Fall Online Forum (FOF) has a large number of threads devoted to listing things found in Fall songs, all of which would benefit from the use of a concordance (unless you’re of the camp which holds that making life easier and more accurate is contrary to the spirit of The Fall). Years ago, Graham Coleman wrote an article for the September 1996 (#7) issue of the fanzine The Biggest Library Yet (1994-2000), listing “Famous Names” which appear in Fall songs. And the heroically detailed Martin Peters wrote a series of articles for The Pseud Mag, analysing Fall lyrics from different angles: “Humour in The Fall” (#1 Dec-Jan 2004/5); “The Time of the Fall” (temporal language and dates, #2 February 2005); “The Insekt World of The Fall” (#3 April-May 2005); “The Medical” (health/body-related language, #4 Aug/Sept 2005); and “Words of Expectation: verbs used by Mark E Smith in Fall Songs” (#8, Feb/Mar 2006, p.38).
Given this tendency, it’s not surprising that the idea that someone should produce a concordance is not a new one. FOF member Glenselvy commented on 9 December 2003 that the Fall website was “the biggest library yet” – and that therefore “the Library needs Librarians”. Glenselvy then floated the idea of “a concordance to The Fall. Or summat.” [sic]. On 17 December 2009, FOF member delmore posted that he was “working on a Fall lyrics concordance.” But there was no progress on the concordance front until FOF member (and creator of the excellent “The Annotated Fall” site) bzfgt said on 16 August 2013, “What I really wish someone would do is a Fall concordance, but I don’t think I’m the man for the job.” I responded on 20 August with, “I’ve long thought of doing a concordance. I have the necessary mania but not, unfortuanately [sic], the necessary time…” But the idea wouldn’t go away, and so mania over-ruled lack-of-time and by 24 September I had produced a first attempt.
In October 2014, my attention was drawn by a fellow FOF member to the work of Dr Matt Davies of the University of Chester. It seems he has a research interest in the lyrics of Mark E Smith, and in July 2009 he presented at the International Conference of the Poetics and Linguistics Association (PALA) at Roosevelt Academy, Middleburg, Netherlands, on the subject “The Wonderful and Frightening World of WMatrix and the lyrics of Mark E Smith”. He spoke about his work at the 2014 Louder Than Words festival in Manchester, and was kind enough to mention The Flickering Lexicon.
There are already some song-lyric concordances out there. A Beatles concordance entitled “Things We Said Today” was published in 1980 (it was apparently the first time that pop/rock song lyrics had been given the concordance treatment), and a Bob Dylan concordance appeared in 1992. Web-based lyric concordances include one for The Grateful Dead. There aren’t many others. Having now done one for The Fall, I understand why.
Before the computer era, constructing a concordance was a horrendously laborious and time-consuming task, which is why the earliest concordances were produced for religious texts and the works of Shakespeare and classical writers: one of the most famous concordances is Cruden’s Concordance to the Bible, first published in 1737 after two years’ work. Access to computers has democratised concordance production, at least in principle, and there are several software products (freeware and paid-for) which can be downloaded and used with relative ease. Frankly, however, the process is still a complete pig; just not a pig that drags on for years and years.
Colin Campbell and Allan Murphy, the compilers of the invaluable Things We Said Today (usually referred to here as Campbell/Murphy), first had the idea of producing a Beatles concordance in 1976 – it took them four years from conception to publication, working thousands of miles apart in an era before email. They used punch-cards and a room-filling IBM system 370 model 155 mainframe with three megabytes of memory. Their book also contains definitive versions of all the Beatles’ songs (184 distinct songs, plus variants and minus exclusions, making a final total of 189), for each of which they had to obtain reprint permission – itself a laborious business. To put their achievement and my whining into perspective, it only took me a matter of months to produce an experimental first draft of my concordance, initially using the free Simple Concordance Program (v.4.09) on my laptop – which has 8 GB of RAM. So I have no right to complain.
It is of course foolhardy to set out to construct a concordance of the lyrics of a group who have not actually stopped writing songs (unless you’re picking a selection of songs representing a particular time period, perhaps). The attempt is particularly misguided when the group in question is The Fall, whose vocalist and main lyricist Mark E Smith is one of the most difficult in rock music. Although Smith’s lyrics can sometimes be clear and simple, they are often obscure or indecipherable. Furthermore, the usual pre-requisite for compiling a concordance is access to definitive texts. Unfortunately, there is no consistently credible (see “Acknowledgements” below) definitive edition of Fall lyrics, even those which are Smith-approved.
Smith himself has undermined the very notion of a “definitive text”, telling Mick Middles: “To me, a song’s never finished and it’s never good enough, that’s why I don’t write lyrics down. Once they’re on paper, you can’t change ‘em, and I like to change ‘em, even just before I’m going on stage. I am changing lyrics all the time. They are completely fluid. I can change and scribble and then change my mind, try it out at some gig in fucking Doncaster and, if I don’t like it, change it. Lyrics change shape and meaning all the time.” (Smith and Middles, pp.271-272).
So the sensible concordance-compiler would not touch this project with a barge-pole. And yet, there is perhaps something Fall-like about the ludicrously obsessive, worse-than-Sisyphean task I have set myself. Or perhaps I am just an idiot. It’s a thin line.
There will be those who would question the point of The Flickering Lexicon. Campbell and Murphy remark (p.xi) that “The value of such an analysis seemed self-evident to us”, a sentiment I endorse. This kind of thing interests me, and I don’t really care if it doesn’t interest anyone else. However, lyrical analysis, interpretation and archaeology has provoked fierce arguments on The Fall Online Forum. Some fans believe that The Fall resists such analysis, but that doesn’t seem to be an argument against analysis, just a further provocation to analysts. In fact, I find the resistance part of what is interesting about Fall lyrics in the first place.
Mick Middles, for example, has written: “People scrutinise Fall lyrics and spin deeper and deeper into literary theory. Which is to miss the point… Better just to accept The Fall for what they are. And enjoy.” (Smith and Middles, p.278). But who gets to decide what “the point” of The Fall is for any particular listener? Surely that’s something that cannot be imposed by a critic or even the artist themselves. And if we are to “accept The Fall for what they are”, we have first to have some understanding or notion of what The Fall are, which invites the scrutiny Middles (along with MES, sometimes) resists. In fact, Mark E Smith told Middles, “When I started buying records, the ones I liked were the ones I could only half-understand. What I don’t like about a lot of records today is that they’re too clear. There’s no fascination or mystery left.” (Smith and Middles, p.278). Smith is surely here expecting attention from his audience. He is giving his fans something they can engage with, not something they should just blankly consume without curiosity. Mystery inspires investigation, not indifference. What else can “fascination” with lyrics mean than to find those lyrics interesting and/or important and to become engrossed in them and want to understand them better?
Below I explain what I’ve done and how I’ve gone about doing it, and set out the reasoning behind my editorial decisions (where, that is, they haven’t been entirely arbitrary). But first, a comparative statistical overview.
This concordance to the lyrics of The Fall is based on a corpus of the texts of 493 songs (or significant variations of songs) released between 1976 and June 2014. There are 99,124 words (known in linguistics as ‘tokens’) in the corpus, of which 9609 are unique words (known as “types”, which I will refer to as the ‘vocabulary’ – note that I am including years, times and numbers as ‘vocabulary’). The type/token ratio (9609/99,124) is 0.0969. The size of the vocabulary is about 9.7% of the size of the corpus: or to put it another way, each word is used – taking the arithmetic mean as the average – about ten times. However, of the 9609 vocabulary words, about 47.4% (4557) are used only once and 86.9% are used less than the average. Just 127 vocabulary words (or 1.3%) account for about half the 99,124 words in the complete corpus.
Campbell and Murphy’s Beatles concordance, Things We Said Today (1980) was based on 189 songs recorded between 1962 and 1970. The corpus size was 35,390 words, and the vocabulary size was 2346 unique words (i.e. 6.6% of corpus size, or each word used on average about 15 times). The type/token ratio is 0.0663. 935 of the vocabulary words are used only once (39.85%).
According to Leech, Rayson and Wilson’s Word Frequencies in Written and Spoken English (2001), based on the British National Corpus, 52.44% of word forms occur only once in the BNC.
http://www.oxforddictionaries.com/words/the-oec-facts-about-the-language, looking at the Oxford English Corpus and referring to lemmas rather than unique words per se (a lemma is the base form of a word, i.e. falling, fell, fallen, fall and falls are all examples of the lemma fall) tells us that 100 lemmas account for over 50% of the corpus and just 10 account for 25%.
More statistical discussion and comparisons can be found here.
The most important decision I had to make was what to include in the corpus on which The Flickering Lexicon would be based, and what to exclude. This might sound like it would be a straightforward thing to decide, but it wasn’t. First of all, I spent some time thinking about whether The Flickering Lexicon would be a concordance to lyrics of songs by The Fall, or a concordance to the complete works of Mark E Smith. In plumping for the former, I put a lot of weight on the practical consideration that it would be much easier to restrict the concordance to Fall song lyrics: most of them have already been transcribed and published in various formats. Whereas on the other hand, Smith’s writing encompasses formats other than lyrics: memoirs, short stories, spoken-word/electronica solo projects, lyrics for collaborative musical projects, plays and more. Much of it is not easily obtainable in electronic format. It would not be practically possible to include everything which should in principle be included, which would make it a pointless undertaking. Furthermore, Fall songs are a distinct and distinctive body of work. Despite the many musicians and non-musicians who have passed through the ranks of the group, there is an aesthetic congruity to the songs and lyrics which legitimises the narrower focus. However, the logic of this position also leads to a wider remit than might perhaps be expected, and leaves some puzzles to resolve.
To summarise: the objective of the The Flickering Lexicon project is to construct and maintain a concordance to the lyrics of songs by The Fall. Is it obvious what “songs by The Fall” means? No, it’s not. My remit uncontroversially excludes Smith’s work with other artists (the Von Südenfed collaboration, for example, is out of scope), assorted poems (“Wigan Soul Poem” from The Fall Lyrics, “Village Bug” from Sinister Times, “Yeller Mania” (uncredited, but likely the work of MES) from the cover of the c.r.e.e.p. single), the contents of Smith’s two “solo” records (The Post Nearly Man, 1998; Pander! Panda! Panzer!, 2002), “Hark the Hoaly Lunatic” (printed in the NME 21-28 December 1985, p.15), “Musical Influence in Great Britain on Big-Head Here” (from the 1994 anthology “Idle Worship”, edited by Chris Roberts), and the short story “No Place Like It” (MES’s contribution to the 1999 Penguin anthology “The City Life Book of Manchester Short Stories”). But how should I treat lyrics that Mark E Smith did not write or does not sing? Campbell/Murphy included “all the songs written, recorded and released by The Beatles as a group between 1962 and 1970”, but excluded “songs which The Beatles recorded but did not write or songs which they wrote for other performers” as well as bootleg recordings and studio out-takes (p.xv).
In the end I decided to err on the side of inclusivity: cover versions and lyrics written or sung by group members other than Smith were included, the important thing being that the songs are under the rubric of The Fall. I think this follows logically from my decision not to produce a Mark E Smith concordance, but it’s a decision that has consequences which require some discussion. Cover versions have been an important part of The Fall’s repertoire, but there are no doubt those who, buying into the idea that The Fall basically reduces to the artistic vision of Mark E Smith, will claim it is perverse to include lyrics to songs performed by The Fall but not written by Mark E Smith, but to exclude lyrics to songs written or co-written by Mark E Smith but not performed by The Fall. But I think it is reasonable to say that The Fall as a unit is what will interest most users of the Concordance, and that the group should not be seen as merely one among many outlets for whatever it is that Mark E Smith thinks he is doing. It’s also worth noting that Fall cover versions do often differ from the originals, musically and lyrically. The lyrics to The Fall’s version of The Kinks’ Victoria, for example, has three verses to the original’s four, and with unexpected decorum renders the line “Sex was bad, called obscene” as “It was bad, called obscene”. The Fall’s cover of Ghost in My House alters a reference to a “coffee cup” to a “tea cup”, among other changes.
In chronological terms, coverage extends from the formation of the group in 1976 to the release of the track Fibre Book Troll on the multi-artist compilation Modeselektion Volume 3 in June 2014: a total of 493 individual songs and song versions. Other songs which have been played live or subsequently released on record during 2014 or afterwards will be candidates for inclusion in future editions of The Flickering Lexicon.
Having settled, at least to my own satisfaction, the issue of the scope of the concordance, the next stage was to collect the texts of the lyrics for processing. Before I could do that, however, I first had to create a canonical list of Fall song titles to use as a checklist. I decided that in principle the canonical list would include everything without distinction: every song performed, recorded, rehearsed, reported or reliably rumoured to have existed. All the official studio output would be there, of course, but I would also draw on live recordings (official, bootlegs and fan-traded), videos/DVDs, Peel sessions, compilation contributions, out-takes and remixes in order to pick up significant alternate versions (i.e. not just different performances, see “Assembling the Fall Lyrics Corpus” below) and otherwise unrecorded material. I didn’t stop there: I also scoured books, fanzines and the web for evidence of songs which remained unrecorded or otherwise unknown. If it was attributed to The Fall, it went in.
For the avoidance of doubt, Pearson’s Revenge, a poem attributed to C. Ritchie on the Fall’s Shift-Work and Holidays DVD (1991), did not make it onto the canonical list.
The finalised canonical list was then divided into included and excluded songs. To be included, a song obviously needed to have lyrics. Instrumentals were therefore quickly excluded, but there are a few borderline cases. Faced with a similar decision, Campbell/Murphy excluded Revolution Nine from their Beatles concordance, despite the fact that it does feature vocals. I have treated Symbol of Mordgan and Where’s the F***in Taxi? C*nt? as practically equivalent to instrumentals, as they consist entirely of pre-recorded dialogue. Also excluded were songs of which no official recorded version seems to exist (Classical Gas, for example): there is no way of knowing, in such cases and within the restrictions I have set myself, what lyrics were sung. Race Hatred is an exception because of the existence of a lyrics sheet, and You Don’t Turn Me On is included on the basis of Una Baines’ memory of what some of the lyrics were. Cab Driver has been excluded on the grounds that it is generally regarded as an early version of City Dweller and the lyrics are in any case very indistinct on the recorded version that appears on the B-side of the Behind the Counter single.
I have treated original studio recordings as the preferred canonical standard for lyrics, but have also allowed for selected remixes, significant alternate versions and lyrics sourced from official live recordings if nothing better seems to exist. The Concordance includes texts from two versions of the following songs: Mere Pseud Mag. Ed.; New Puritan; Touch Sensitive; Blindness; Clasp Hands; Recovery Kit; Cosmos 7; Devolute. The remixes included are: Last Nacht (remix of Bremen Nacht), and League Moon Monkey Mix (remix of The League of Bald-Headed Men).
Should albums be regarded as more definitive than singles? Should precedence be determined chronologically? Basically I have done whatever I felt like doing. For example, I have used the Middle Class Revolt album version of 15 Ways rather than the slightly different single version or Mark Goodier session version, despite the fact that the single preceded the album and the session version has some interesting lyrical variations. I just think the album version is better. The same consideration applies to 2 x 4.
A list of all the Fall songs included can be found above, as can a list of all the Fall songs that have been excluded – usually because they are instrumentals, but sometimes because no recorded version exists – or if it does, that the vocals are distorted to destruction and/or unlistenability (I did try to transcribe an early live version of Louie Louie, but gave up – I may be obsessive, but I’m not a martyr).
I have imposed my own line arrangement on the texts, freely disregarding any evidence from printed lyric sources. I took a few factors into account: a) what looks best when printed in the Concordance; b) what makes most sense according to the rhyme scheme, if there is one; c) what it sounds like the line arrangement might be. It is worth noting that MES is typically not writing poetry, and so line arrangement is perhaps not necessarily very important. Anyway, I’ve done my best to make things look neat and tidy, but remember that the source material is not remotely neat and tidy.
British English spelling conventions have been adopted throughout. However, following Campbell/Murphy I have used colloquial spellings whenever that sounded appropriate (for example, “gonna”), as well as abbreviated word forms. I have not typically tried to represent MES’s speech patterns and accent through spelling, nor have I adopted spellings which may have appeared in the lyrics books or elsewhere. Another challenge was (real or invented) foreign language vocabulary: I have partly relied on others’ transcriptions, and partly done my best according to what my ears were telling me. I’ve made my own decisions about which words should be hyphenated, and tried to be consistent in doing so.
Homographs are words with different meanings but the same spelling. So for example the word “cast” is used as a verb in Iceland (“Cast the runes against your own soul”) and as a noun in Weather Report 2 (“The cast deserved to die”). The current edition of The Flickering Lexicon makes no attempt to distinguish between such different meanings.
Fall songs are full of shrieks, screams, growls, “hup!”s, “huh-uh-huh”s, and, most notoriously, “-ah” or “-uh”s. The “-ah”/”-uh” suffix is really an artefact of MES’s accent, and I find the tendency of the media to laugh at it both patronising and tedious; users of The Flickering Lexicon therefore won’t find any examples here, with the lone exception of a couple of instances from the song Ol’ Gang where it seems to serve a different function. Leaving “-ah”/”-uh” aside, it is not possible in most cases to determine with any certainty whether exclamations, shouts and vocalised noises are technically supposed to be words or not. I have excluded wordless screaming, gargling and so forth, but following Campbell/Murphy I have decided to include some vocalisations as if they were words whenever I felt they seemed significant, despite the problems inherent in doing so. I have tried to apply standardised (often onomatopoeic) spelling (for example, I used a consistent spelling of “aarrgghh” throughout the corpus), but frankly it’s all a bit dubious.
Some Fall songs have straightforward backing vocal parts. Sometimes it isn’t clear which vocals are taking the lead, if any. Sometimes there are several colliding vocal parts, and it is very difficult to work out what is going on. When it is tolerably clear what the backing vocal is, I have put the lines in single brackets so they are distinguishable in the Concordance. Otherwise I have just muddled along as best I can.
Most cardinal numbers (one, two, three etc.) and ordinal numbers (first, second, third etc.) have been transcribed as words and are therefore sorted alphabetically for concordance purposes. For convenience, I will create a separate list of them. However, times (i.e. “9:30”) and years (i.e. “2001”) have been transcribed numerically and have their own section in the concordance. There are a handful of other items which I have transcribed as numbers: “18-30” as in Club 18-30, “45” and “45s” as in 45 RPM records, “555” as in the cigarette brand, and “59” as in Blob 59. I decided to transcribe what might otherwise appear as Roman numerals as words. So for example, it is “I am Roman Totale seventeen” (Second Dark Age) rather than “I am Roman Totale XVII”, on the grounds that that is how it is actually pronounced.
A corpus is more than just the text. In order to perform any analysis of the text, it needs marking up, to indicate the important features. This could include codes for parts of speech, or other linguistic features. For the purposes of The Flickering Lexicon, I have kept this simple and restricted the coding to song level. The corpus includes mark-up for the following information about each song:
· Year (usually first year of official release of the version I’ve used)
· Date (of official release of the version used)
· Code (an abbreviated form of the release title)
· Release Title
· Vocalist (the main vocalist or vocalists – future versions might try to mark up song texts to identify backing vocalists, or the precise lines sung by each main vocalist, and so on)
· Whether the song is by The Fall or is a cover version
· Format (album, single/EP, Peel session, Other radio session, Live, Other version)
In addition to possibilities recorded above, I will give consideration to including track timings and song part (i.e. chorus/verse/dialogue) in future editions of The Flickering Lexicon.
It is not uncommon to exclude certain very common words from concordances (“a”, “and” etc.). I have included everything.
The nature of song lyrics is that certain vocal lines are repeated – often dozens of times. Because I wanted to be true to the structure of the source text, I have decided to include every occurrence of every line. Obviously this approach means increased frequency for some words, for example those used as part of a repetitive chorus. Some users of The Flickering Lexicon may think that this will lead to distortion of statistical analysis, but I disagree.
In their released versions, several Fall songs include pre-recorded dialogue, announcements, radio broadcasts and other spoken word material which would not necessarily be performed live. I have usually decided to include the dialogue in the song texts I’m using. Here’s a list of the affected songs:
· God Box – introductory exchange between MES and Brix Smith.
· Mother-Sister! – introductory exchange: “What’s this song about?” “Er, nothing.”
· Lucifer Over Lancashire – introductory off-air recording of a radio phone-in, apparently involving Craig Scanlon.
· Australians in Europe – opening dialogue is a taped conversation between MES and Trevor Stuart.
· Rememberance R - includes a spoken word rant by Ding about bands reforming.
· Paintwork – includes an snippet of dialogue from an Open University broadcast, attributed to MES accidentally triggering the record button on his dictaphone/cassette recorder in his hotel room.
· Ivanhoe’s Two Pence – begins with some lines of dialogue taken from an Italian film entitled La spade normanna, dubbed into English (a version often titled, Ivanhoe, the Norman Swordsman).
· Jam Song – starts with a recording of one of the producers expressing the view that the song needed a chorus – he was sent home.
· I Wake Up In The City – includes an extract from an off-air recording of Kate Hoey (former UK Minister of Sport) talking about a requirement for schools to provide a minimum of two hours physical education per week.
· Assume – starts with some dialogue recorded from an unknown television programme.
But I haven’t been consistent. Here are some songs where the spoken word content has not been included in this edition of The Flickering Lexicon:
· Fortress – “And today, here on the vitamin B glandular show” omitted.
There are also some tracks which, as noted above, I have excluded from the concordance, because they consist wholly or mainly of pre-recorded material:
· Symbol of Mordgan – apparently a recording of Craig Scanlon being interviewed by John Peel about football.
· Where’s the F***in Taxi? C*nt? – MES, Ed Blaney and others talking, drunkenly.
As I have already indicated, the general lack of “definitive” MES-supplied lyric sheets means that Fall fans have usually had to rely on the evidence of their own ears when trying to decipher what is being sung. Inevitably, this leads to disputes and controversies, and in some cases the ensuing arguments have been long-running and intractable. For concordance purposes I have had to make decisions about how these controversies should be resolved, but obviously those decisions cannot and will not put an end to the arguments. I discuss what I regard as the most significant of these disputes below.
Probably the most notorious Fall lyrics dispute, with no resolution in sight. The Fall Online Forum thread on the subject currently runs to 16 pages. The dispute concerns the chanted refrain which the Lyrics Parade [see here] rendered as “Baghdad/Space Cog/Analyst”. The Lyrics Parade used to include a note of explanation, but this is omitted from the current site. The note read:
Russell advised on 14 February 1996: "I wrote to Cog Sinister with a pile
of questions, one of them being what the hell Brix was saying [in "Guest
Informant"]. Lucy [Rimmer] wrote back
and said she asked Brix, who said the phrase was "Baghdad/Space Cog/Analyst" [Source: Annotated Fall]
However, few listeners seem to agree with that.
This concordance is based in the first instance on the lyrics as they appear at the excellent The Annotated Fall lyrics website, which in turn is based on the texts at the wonderful and venerable The Fall Lyrics Parade, originally compiled by Jonathan Kandell and Jeff Curtis, and now edited by Conway Paton. I used Lyrics Parade texts when songs were missing from The Annotated Fall site (cover versions, principally but not exclusively). Those two sources were supplemented by Fall Tracks A-Z, part of the Reformation! Fall-fan site. There are two books of Fall lyrics, both published with Mark E Smith’s blessing. Typically, however, they are not entirely reliable. I have referred to them sometimes, but have not treated them as canonical, as compared to the evidence of my own ears (and the collective ears of Fall fans as reflected in the fan-published lyrics sources).
In constructing the Concordance, I have felt the need to make my own amendments to the texts I have used. Responsibility for what appears here therefore rests with me, as much as I might try to deny it later.
The final version of The Flickering Lexicon has been produced using Rob Watt’s Concordance Software. Watt is an Honorary Research Fellow in English at the University of Dundee.
Finally, I would like to pay tribute to the pioneering work of Campbell/Murphy: their book has been a useful source of guidance for my own project, but it is not responsible for my errors.
Campbell, Colin and Murphy, Allan (1980). Things we said today: the complete lyrics and a concordance to The Beatles’ songs, 1962-1970. Ann Arbor, MI: Pierian Press.
Griffiths, Dai (2003). From lyric to anti-lyric: analyzing the words in pop song. In Moore, Allan F. (ed.) (2003). Analyzing popular music. Cambridge: Cambridge University Press. Chapter 3, pp.39-59.
Mátyás, Csikós (2014). The change in the lexical complexity of Van Halen songs. Blog posting available at: http://wondefulworldofappling.blogspot.co.uk/2014/05/the-change-in-lexical-complexity-of-van.html (9 May 2014).
Fell, Michael and Sporleder, Caroline (2014). Lyrics-based analysis and classification of music. In Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: Technical Papers, pp.620–631, Dublin, Ireland, August 23-29 2014. http://www.aclweb.org/anthology/C14-1059
Khalifa, Jean-Charles (2007). A semantic and syntactic journey through the Dylan corpus. In Oral Tradition, Vol. 22 (1) pp.162-174. http://journal.oraltradition.org/files/articles/22i/Khalifa.pdf
Leech, Geoffrey; Rayson, Paul and Wilson, Andre (2001). Word frequencies in written and spoken English: based on the British National Corpus. Harlow: Pearson Education.
Schmidtke, Daniel (2013). Time out of mind: a corpus linguistic analysis of 50 years of Bob Dylan lyrics. Blog posting available at: http://danschmidtke.wordpress.com/2013/05/24/time-out-of-mind-a-corpus-linguistic-analysis-of-50-years-worth-of-bob-dylan-lyrics/ (24 May 2013).
Smith, Mark E and Middles, Mick (2008). The Fall. Updated edition. London: Omnibus Press.
Werner, Valentin (2012). Love is all around: a corpus-based study of pop lyrics. In Corpora, Vol. 7 (1), pp.19-50.
Beta version published September 2013.
First edition published October 2014
Latest Update: 6 June 2015