If you would like to participate in discussions, please sign in or register.

Sign In with Facebook Sign In with Twitter

In this Discussion

Most Popular This Week

Hooks and Glue in Chord Progressions

Dear members of the SMT Discuss community,

I would like to know if anybody could recommend any music theory articles that relate to the idea of “hook” vs. “glue” in chord progressions as discussed by Coker, Knapp, and Vincent in the following passage from their introduction to their 1997 jazz text-book “Hearin' the changes: Dealing with unknown tunes by ear.” I am particularly interested in finding articles (or books) that discuss this topic in the context of music other than jazz but any jazz references are welcomed as well.

“There are two kinds of commonality extant in the average [jazz] tune's chord progression, and they could be referred to as "Glue" and "Hooks." "Glue" is represented by those aspects of a chord progression which are so common as to occupy approximately 60-90% of its entire length. This would include chord roots which move in the cycle (or circle) of fifths (especially the IIm7-V7-I progression and its multifarious variations and extensions) and chord roots which descend chromatically. In itself, "Glue" does little to attract interest or create excitement. It is so commonplace that it sounds virtually uneventful, mostly serving to cement chords and keys together in a logical fashion. However, the glue portions of a progression are easy to recognize by ear and easy to memorize. They are a good place to begin our study of chord progressions, and at the very least are a signal to the improviser that the progression is momentarily "idling" somewhere within the key and probably not requiring much in the way of harmonic interpretation with regard to scalar/note adjustments. The "Hook," on the other hand, is represented by those aspects of a progression that are highly significant, perhaps even the sound that distinguishes one tune from another, or enables an improviser to recognize a specific tune (perhaps the only tune) which utilizes that sound. "Hooks" offer contrast to the more 'vanilla' "Glue," and exist in much smaller quantity, perhaps only 1-3 "Hooks" in the entire progression. They are generally unexpected surprises of a dramatic, inspiring sort. Some of the possibilities for a "Hook" are unusual root motion, unusual chord-types, an unexpected chord resolution, or a sudden modulation to a remote key. A progression made entirely of "Glue" would probably be dull, placing a heavy burden on the quality of the given or improvised melody. A progression having nothing but "Hooks" would risk sounding illogical, fragmented, weird, harsh, aimless and, believe it or not, dull because of the sameness of its unpredictability. So the most successful/popular progressions will be mostly "Glue" (60-90%), but with at least one well chosen "Hook." The wise improviser· will use the Hooks as emotional, dramatic peaks in his/her solo, well worth expounding upon, being more relaxed and conversational during the "Glue." It is comforting to discover that even specific "Hooks," though less commonplace than "Glue," generally exist in a substantial number of tunes. For example, a modulation to a new key that is a major third above the first key (as in the key of C to the key of E) is a dramatic-sounding modulation (a "Hook"). Yet songwriters, recognizing that dramatic effect, have continued to write many songs which utilize that particular modulation.”




Ivan Jimenez, PhD

Music Theory and Composition

Visiting Researcher

Sibelius Academy, University of the Arts

Helsinki, Finland





Sign In or Register to comment.


  • 1 Comment sorted by Votes Date Added
  • Ivan, your question did not elicit many answers... The reason probably is that your (or Coker, Knapp, and Vincent's) definition of "Glue" and "Hooks" really is too vague.

    It would be extremely difficult to distinguish the "multifarious variations and extensions" of II7–V7–I that would still qualify as "Glue" from those that must be considered "Hooks", without an explicit theory of this difference. Hearin' the changes claims that "Glue" include "chord roots which move in the cycle of fifths": in either direction? Also, a progression II–V–I must start from somewhere, probably from I; but is I–II also a "Glue" progression, even although it neither follows the cycle of fifths, nor descends chromatically? What does it mean that "Glue" progressions "occupy approximately 60-90% of [a tune's] entire length"? Is it really a matter of duration? Does it make a difference if the "Glue" progressions are, say, in whole notes, while the "Hooks" are in eighth notes? Are phenomena such as "unusual root motion, unusual chord-types, an unexpected chord resolution, or a sudden modulation to a remote key" really of the same nature?

    I once proposed a classification of root progressions in common practice tonal music, of which an early version was published in MTO. It met with limited success (only until now, I hope ;–)), but some of my students and others did use it as a starting point for their research. You'll find a bibliography about the theory on my website (mostly in French, unfortunately). See in particular the publications by Philippe Cathé, who studied among others the progressions in the Beatles, with quite interesting results. See also the PhD thesis of Paul Scott Carter, which concerns progressions in Pop-Rock music.

    Philippe Cathé made statistics on harmonic progressions in the same chorale (Vater unser im Himmelreich) in 10 different harmonizations covering roughly the early 17th to the late 19th century. Click on the image below to see his results. The percentages are (in number of root progressions, not in duration) what I call "dominant progressions" ("dominant vectors") as opposed to "subdominant progressions". You'll see that from Pachelbel to J. S. Bach, the percentage is above 85%, with the notable exception of Zachow who is known to have written in a highly conservative style. One might be tempted to conclude that this asymmetry between dominant and subdominant progressions somehow defines tonality itself, but I don't consider that a reasonable conclusion: tonality is much more than that.


    Cathé has shown that the percentage of dominant progressions could be less than 50% in some of the songs by the Beatles, which is utterly different from anything known historically (and which indicates, if necessary, that the harmony of the Beatles is not without interest).

    I hope that this can help.