In a 2017 video titled Hallelujah – How to Cover it Badly, designer, composer and music YouTuber Tantacrul (Martin Keary) explains how 40% of covers of Hallelujah are ‘bad covers’ and ‘[screw] up’. I greatly enjoy Tantacrul's content on the whole, but I couldn't help but feel that the points made in this video (far older than his recent rise to success, from a time his channel had less than 2000 subscribers) really fell short of the high bar set by his more recent content.

Specific criticisms

To make most of his criticisms, Tantacrul focuses on the ‘arc’ of the chorus, explaining that the music rises through the lines ‘It goes like this, the fourth, the fifth, the minor fall and the major lift’, peaking at the end of the line ‘The baffled King composing, Hallelujah’, and then softens and slows through the remaining ‘Hallellujah’s, leading in to the next verse.

Based on this analysis, Tantacrul proceeds to criticise his first victim (a cover by Noah Guthrie) on the basis that the ‘singer dipped down a semitone below the final note and then came up’ which ‘sounds a little bit lame’ – going to ‘remind [us] of how Jeff Buckley does it’. Needless to say, this argument is a total non sequitur. The fact that the cover may differ slightly from the original does not make it a ‘bad’ cover, and Tantacrul offers no explanation as to why this may be the case.

As to the unfounded assertion that the music ‘sounds a little bit lame’ – perhaps my ears are simply very unusual, but I find the melodic change quite inoffensive, as, indeed, has everyone else I've shown this cover to. Perhaps it does sound odd to Tantacrul's ears, but undoubtedly it sounds enjoyable to many listeners (clearly, including the singer and the 19,000 viewers who ‘liked’ his cover), and it is patently absurd to label a musical decision ‘bad’ or worth a ‘quibble’ on the mere basis of a difference in personal taste.

Tantacrul seems to try to justify this on more objective grounds, rhetorically asking ‘Why would the second last note be the lowest?’, as if this is also somehow completely conceptually bizarre – yet I'm sure he ought to be perfectly aware that nonchord tones like these serve well-established roles in music theory. Maybe the ‘overshooting’ of the melody creates additional dissonance, priming us for and increasing the sense of the resolution that follows. Or maybe it just sounds nice!

Turning to covers by Say Chance1 and Tiffany Day, Tantacrul criticises them for making ‘the most common and egregious offen[ce]’, transposing the last phrase of the chorus up one octave. Shock, horror! How egregious(!)

Tantacrul explains that this is bad because it's now in a ‘different register’ and therefore doesn't ‘sound right’. Again, to my ears, this sounds pretty acceptable – and I'm sure the 260,000 who ‘liked’ Tiffany Day's video would agree!

Tantacrul tries reinforcing this point by drawing the ‘arc’ from earlier, with a sharp jump in the final passage signifying the transposition, the implication being that because the music doesn't ‘fit’ the original ‘arc’, it must therefore be ‘bad’. Yet thinking back to exactly what the ‘arc’ is supposed to represent – a softening and mellowing towards the end of the chorus – one could argue the octave transposition accomplishes exactly that! Tantacrul even helpfully points out that the melody is now in a different register, and this jump into head voice a lighter-timbred register arguably precisely fits the arc. Or, again, perhaps it just sounds nice!

Music theory doesn't identify ‘bad’ music

Tantacrul's approach at various points in the video is strongly reminiscent of an attempt to use music theory to identify ‘bad’ music. While he doesn't pull out any of what we may call ‘hard’ music theory per se, the focus on ‘arcs’, ‘registers’, graph-drawing and the minutiae of melodic contour (god forbid the melody drop ‘a semitone below the final note’!) very much lies in similar territory, built on the assumption that with enough attention to detail and technical analysis, we can objectively measure how ‘bad’ a piece of music is.

This approach is fundamentally misguided. As 12tone puts it in his video Why You're Wrong About Music Theory:

Music theory is not a scoring system. It's not a way of determining which songs, albums, artists, genres or time periods are doing music ‘best’.

Ultimately, the only consideration as to whether music is ‘good’ or not is whether listeners enjoy it! Clearly, Tantacrul knows this – ‘sounds … lame’ and ‘does[n't] sound right’ are subjective measures that reflect this. But this doesn't stop him attempting to also appeal to the ‘arc’ of the music to bolster his argument, as if any deviation from this analytical tool somehow automatically renders a cover ‘bad’.

Covers aren't about faithful reproduction

Beneath all this discussion on subjective and objective elements of musical composition, though, lies an undercurrent that is far more concerning – the implication that deviation in a cover from the original music is somehow inherently bad.

Of course, Tantacrul does pay lip service to the importance of individuality and interpretation in a cover:

This isn't to discount the importance of individuality and alternative perspectives in music. There's more than enough room elsewhere in this song for that.

But what does this mean? Why is the chorus of Hallelujah singled-out as inappropriate for individuality as opposed to ‘elsewhere in this song’? More to the point, what part of Tantacrul's criticism couldn't be applied to any other part of the song? Every other part of the song has a particular ‘arc’, ‘basic structure’, melodic countour, ‘register’ and a particular amount of ‘warbling’ (or lack thereof). If a departure from the original in any of these respects during the chorus was worthy of criticism, then the same surely applies to any other part of the song.

The very nature of a cover is that the performer will bring their own personality and interpretation to the song. If you want something that sounds exactly like the original, then simply listen to the original.

That personality and interpretation may come through in many different ways. There may be changes to the melody, harmony, tempo and structure, and any one particular listener, like Tantacrul, may not like those changes. But so long as there are some listeners who do like those changes, and do like the performer's personal spin on the song (which, in the case of these covers, there clearly are), it seems far fetched to label the cover as somehow being ‘bad’, over what is essentially a difference in taste.

Reflections on classical and contemporary music interpretation

I'd like to cap off this discussion with a brief exploration of the nature of interpretation in classical and contemporary music.

Tantacrul is a clasically trained composer – my background, too, is in classical music, though his qualifications clearly far exceed mine! In the performance of classical music, the fundamental role of the performer is to give effect to the intention of the composer. The performer must have an acute awareness of the conventions of the period so as to contextualise what is written in the music. Every note should be played exactly, every direction must be adhered to. We even see performers going to such lengths as retuning instruments to A432 to be more faithful to conventions of the time, seeking antique instruments or restringing instruments with catgut, to be more ‘authentic’ to the period.

By contrast, many contemporary genres of music lean very much in the opposite direction. In the jazz tradition, a song might be reharmonised (I wonder what Tantacrul would think of Jacob Collier's recent interpretation of Hallelujah). A piece of pop music might contain significant embellishment which would not be reproduced in a written transcription, and a singer performing a cover might add or subtract their own embellishments,2 or change the music in ways which would be considered highly significant in classical music. Indeed, it would be nigh unthinkable for a cover to merely faithfully recreate the original!

Note that I am certainly not saying that either approach is more ‘correct’ or ‘better’ than the other. Each has its place, and there is a whole other discussion to be had about what this may say about cultural and technological aspects of music, and the role of intersectionality. But I think it is interesting to consider the parallels between these traditional approaches and those of Tantacrul's video.


  1. This video, at 06:12 in Tantacrul's video, is now unlisted, so I presume the singers would prefer that it not be linked. 

  2. Compare Tantacrul's much-maligned ‘warbles’. I'm certainly no fan either, but embellishment is very much part of the modern ‘America's Got Talent’-pop tradition, and adding those embellishments to Hallelujah is merely to play by the rules of this genre. As discussed, a cover is naturally transformative, and to judge the result solely by the conventions of the original genre seems unnecessarily closed-minded.