Tango Time Signatures and The Beat

(There are many music and audio clips on this page. It might take a moment to completely load on your platform. And if that platform is iOS then some won’t load at all. My apologies but that’s an unfortunate problem with some of tech used here.)

Anyone not familiar with note value (duration) or s should first read Notes and Rests: Their Duration.

This post discusses time signatures, early tango’s use of the as the pulse, and the change to 4 s per in tango music. Specifically tango music, not milonga or vals.

To review, the habanera rhythm looks and sounds like this:

There are several music/audio examples prepared to show time signatures do not necessarily determine our perceptions of the beat. And I include some recordings of El Choclo, an early tango that has been arranged and recorded by many of the great s. I’ll show that time signatures are malleable. Composers, arrangers, musicians can make us hear and feel “2” or “4” in any of the time signatures used in tango: 2/4, 4/8, or 4/4.

Given what I just wrote, it is not surprising there is a great deal of confusion about tango time signatures, the beat, or “count”. Because milonga is always in 2/4 and vals in 3/4, and neither have undergone as many changes as tango, most writers talk about their time signatures and the beat correctly. When it comes to tango, there is much mis-information, or misinterpretation regarding time signatures. There needn’t be.

A couple examples I’ve recently come across, paraphrased: 4/4 is twice as long as 2/4, therefore it is much easier to make longer or more complex phrases; Dos por cuatro is 2/4 played with a lilting rhythm that is (edit) the habanera. These statements are incorrect. It should become apparent why as we go through the post. (I’m not being critical – the examples come from people who have made a worthwhile contribution to the tango literature. They just aren’t musicians.)

Horacio Salgán writes about the habanera in early tango, the beat, and time signatures:

“The time signature of the Habanera is 2/4. This time signature of 2/4 was still being used even after the rhythm of the accompaniment switched to the “four” pattern.

   “Later, considering the fact that the tango was being played using four 8th notes to the measure but counted in 2/4, the time signature was changed to 4/8. Nowadays the tango uses the time signature of 4/4…The most important aspect in this case is the sound, that is to say how it soundsIs there any difference in the way a tango is played, whether its time signature is 2/4, 4/8, or 4/4? Absolutely not…In practice, the use of a certain time signature in the tango has neither altered nor influenced its performance nor its spirit”.

Horacio Salgán. Tango Course, 2nd ed. 2001. page 22.

I can see how non-musicians may get confused. Salgán is a musician and tango legend, still with us at the age of 97. But a couple points he makes needs some clarification for non-musicians. Saying, in translation, “the time signature of the Habanera is 2/4″ is easily misinterpreted. The habanera is a rhythmic pattern. Rhythmic patterns exist in a time signature – everything does – but they can sound the same in different ones. What is meant here is the habanera rhythm takes two full beats to execute completely, and if the goal is to have the syncopation on the off-beat of the strong beat (beat 1) leading into the second beat, then the time signature must be 2/4. And because the habanera is the bass line and the accompaniment, repeated over and over, it becomes the foundation, the driving pulse of the music. (Pulse and beat are often one in the same and simple definitions of the terms will equate them, but in this case the pulse is not the beat.)

One other thing. Saying “the rhythm of the accompaniment switched to the ‘four’ pattern” can be confusing too. (Not to musicians who understand the nuances in what Salgán is saying in the above quotes). Marking the underlying beat, whether “2”, “3”, or “4” is not strictly speaking a rhythm. It is the beat. Rhythm happens when notes of (usually) different duration move around the beat(s).

Salgán’s second paragraph is very important and the rest of the post explains what he means. (I’ve had portions of this post in “draft” mode for several weeks and planned to show how 2/4, 4/8, and 4/4 can be one in the same as far as our ears and feet are concerned. I received Salgán’s book a few days ago and got inspired to finish my post. The “Course” is mostly about arranging tango standards and keeping faithful to the music. Not quite what I was expecting, but there is worthwhile information in there.)

The “Four” and Time Signatures
Before the second decade of the 1900s tango moved away from the habanera as the underlying driving pulse, to the “four”. Maybe that was done to rhythmically distinguish tango from milonga – just speculating. (Roberto Firpo and Francisco Canaro were two leading proponents of the new “four” style as early as the late 1910s.)

What is meant by the “four”? Simply that tango came to have 4 distinct beats per bar, rather than 2, and that they are almost continuously “marked” in the accompaniment. How was it done? Very simply. Take away the syncopation on beat 1 of the habanera (dotted 8th – 16th) and play two 8th notes instead. Now there are four evenly spaced 8th notes per bar in the accompaniment. Repeatedly play four 8th notes per bar, bar after bar, and the pulse becomes the “four”.

(Milonga remained in “2” and continued to use the habanera and its variations as the driving pulse. The habanera and its variations are prominent in tango, but they do not consistently define the pulse. And soon the second one, called specifically “syncopa” by tango musicians, came to dominate. They are used as rhythmic accents or highlights around the beat and sometimes as the rhythmic pulse. See The Habanera Rhythm and Habanera Variations for more detailed discussion with music/audio and some recordings.)

The usual time signature in tango is 2/4. Even though tango music may very clearly have 4 distinct beats in the bar, most of it is actually written in 2/4. A browse through the tango sheet music on Todo Tango (laborious!) or the scores in three pdf files available from Mandrágora Tango shows 2/4 to be by far the most common tango time signature. There are far fewer tangos in 4/8. Hardly any in 4/4, and those are from the 1950s onwards.

Now, the sheet music available to us are generally older tangos and are piano arrangements, not actual scores, so perhaps 4/8 was used more than is represented in the collections. For example Di Sarli recorded Bahia Blanca in 1957 and it is in 4/8. Tango music immediately after the habanera pulse era has a continuous underlying “four” marked by the accompaniment. According to the sheet music, and Salgán, these early tangos were written in 2/4.

How does 2/4, which means 2 quarter note beats per bar, become the “four”? It is a simple thing to make 2/4 sound implicitly in “4”. As said above, by subdividing the two quarter note beats in half, to four 8th notes and when the four 8th notes are played consecutively and often enough in the accompaniment the music sounds in “four” not “two’. (Breaks the definition of but many definitions and rules have exceptions). Consider this version “subdivided 2/4″.

So the beat is now marked by four 8th notes. That is one reason why, when composers or arrangers chose to explicitly recognize the “four”, they used 4/8. The other reason to use 4/8 rather that 4/4 is because the music looks the same in 4/8. Although both 4/8 and 4/4 have 4 beats per bar, because the beat is marked by the quarter note in 4/4 and the 8th note in 4/8, the music looks different in 4/4. Music written in 4/8 looks identical to music in (subdivided) 2/4, because in both the 8th note is the beat. In 4/8 the commonly used rhythmic patterns and syncopations look identical. Now of course musicians have no difficulty reading and playing music in any time signature, but I think the convenience of having the same look and feel in the printed music is another reason why 4/8 was used.

I suspect once tango music began to exist independently of the dance in the 1950s and new “concert” tangos were being written (late Pugliese, Piazzolla, some Salgán, Stampone, et.al.) there was no need to keep the tradition of 4/8, which is quite rare in most other musical genres. 4/4 is the most frequently used time signature. So common it is also called “Common Time” and the 4/4 time signature is replaced with the letter “C”.

Whether a tango is in (subdivided) 2/4 or 4/8 or 4/4 makes no difference to listeners or dancers, only to the musicians playing it. All the phrases take the same number of bars to complete. In all cases there are four distinct beats. If the musicians want it that way. They are still free to emphasize only two beats in a bar of four, the first and the third ones – a “2” count feel in 4/8, 4/4. And as we saw, it is very simple to create the “four” in 2/4. The music will always guide us. Simply listen and feel the beat.

Well, that was a lot of words. Time for some music and audio examples.

Time Signature Examples
If all the above has been somewhat confusing, hopefully a few examples will clarify.

El Choclo (premiered in 1903) is in 2/4 and the habanera is almost constantly in the bass line accompaniment. The following examples use the first 4 bar phrase and I’ve added a (rather annoying, but functional) drum beat so we can hear the music and the habanera in “2” and “4”, in 2/4, subdivided 2/4, 4/8, and 4/4. Let’s see if there are differences and if so identify them.

As written in 2/4. Notice the tempo is a quarter note = 85 beats per minute and there are 2 beats per bar. The habanera rhythm lines up perfectly in those two beats and provides the underlying pulse below the quick flowing melody.

In 2/4 with the snare drum beats on every 8th note, subdividing and marking every beat. The habanera loses its crisp syncopation on the dotted 8th-16th notes. It is still there in “four” but being spread over two beats, not one, lessens the effect.


Re-written in 4/8. Looks identical. And sounds identical too. But again the 4 snare drum beats get in the way of the crisp clarity of the habanera. The rhythm doesn’t fit so well in “4”, at least when every beat of the “4” is equally accented. Notice the tempo is an 8th note = 170 beats per minute. Double the 2/4 bpm because an 8th note is half as fast as a quarter note. There are 4 beats per bar and the beat is marked by an 8th note. More beats; same amount of time to play the section.


In 2/4 with the habanera syncopation replaced by 8th notes and no snare drum. It is now the “four”, but a very uninspiring way to do it.

In 4/8 with the habanera syncopation replaced by 8th notes and without the snare drum beats. Looks and sounds identical to the above example in 2/4.

Re-written in 4/4 with the habanera. Looks very different now. Notice the difference in time values of the notes. They are double. The tempo is a quarter note equals 170 beats per minute, same bpm as the 4/8 examples, just a different reference note duration. Still 4 beats per bar and still takes exactly the same amount of time to play the example. And still sounds the same.

In 4/4 without the habanera. Now marking the “four”. Sounds the same as the habanera-less 2/4 and 4/8 examples but looks different.

If I removed the snare drum beats every example with the habanera would sound identical. And we heard the examples without the habanera (in 2/4, 4/8, and 4/4) all sound the same too. Despite what you may have heard about 2/4, 4/8, 4/4 being different or allowing different types of music to be written, in terms of what we hear and feel, there are no necessarily inherent differences between them. Composers, arrangers, musicians can accomplish their musical goals equally well in any one of them.

The accompaniment matters and of course tango doesn’t have a traditional rhythm section marking every beat. However, the composers immediately following the habanera pulse era (notably Firpo and Canaro) maintained a mostly constant marking of the “four” count in the accompaniment. Eventually more sophisticated accompaniments were added, either over top of the “four” or replacing it now and then with habanera based syncopations.

And of course many tangos are clearly in “two” not “four” and even those in “four” may accent beats 1 and 3, sounding like in “two”. I wrote all that to make a point: the time signature in tango music is not the deciding factor in how we hear or feel the beat or the pulse of the music.

El Choclo Recordings
Let’s listen to how some of the masters have interpreted El Choclo. Regarding the habanera, Juan Maglio “Pacho” (1910s) plays it extensively, but not constantly. And Roberto Firpo plays it too. The others (Carlos Di Sarli, Juan D’Arienzo) do not play it at all.

Juan Maglio “Pacho”, mid 1910s. Uses the habanera as the rhythmic pulse and the beat is in “two”.

Much later tango music may still be in “two”, as will be apparent in Firpo and Di Sarli’s El Choclo. Only two beats are marked in the accompaniment. Firpo plays the habanera and is somewhat bound to the “two”. Di Sarli doesn’t play it. He chose to play in “two”, perhaps to honour the 2/4 origins of El Choclo, if not the old habanera accompaniment.

My comments refer specifically to Section A, bars 1-16, but they are essentially true in the other sections too, more or less. Click any bar to stop the playback.

Firpo 1940. The habanera is there. Not in every bar but it is obvious when it is and isn’t. Sometimes Firpo plays four chords on the 8th notes in the accompaniment – marking “4”. But nonetheless, the music has a distinct “2” rather than “4” count, because Firpo is honouring the habanera pulse. (Not as true in other sections of the music…)

Di Sarli, 1954. It is impossible to hear this performance in anything other than “2”. Almost every beat is aggressively accented and there are two per bar. The habanera isn’t played at all. I thoroughly enjoy this performance and I think comparing it to D’Arienzo’s, to follow, gets to the essence of their styles. They are so different, worlds apart, yet they are both equally brilliant musicians.

Compare to D’Arienzo. D’Arienzo was called “El Rey del Compas”, “King of the Beat”, and it is said he restored tango music to the dancers. Meaning he emphasized rhythm and the beat over melody and the singer. His beat is usually in “four”, with highly rhythmic patterns and accents. Often the melodic rhythm is adapted to match the accompaniment rhythms; there is a blurring, a blending, of melody and accompaniment.

D’Arienzo’s El Choclo is distinctly in “four”. Same music; same time signature; same melody. And again, no habanera. There is a very prominent marking of 4 beats per bar in the accompaniment. That is especially noticeable when it stops towards the ends of the 4 bar phrases, letting the piano fills between the phrases take over, then resuming as the next phrase starts. Each “beat” is an 8th note, implicitly sounding in 4/8. In this case beats 1 and 3 correspond to beats 1 and 2 in the 2/4 score. Subdivided 2/4 or 4/8 or 4/4? Doesn’t matter!It’s in “four”.

D’Arienzo 1954. To get the full effect of the “four” I recommend listening to the entire performance.

Aside from the faster tempo, D’Arienzo sounds like there is more action and movement. There is. Because the beat has doubled and is being aggressively marked in the accompaniment. It’s a highly creative and brilliant, multi-layered interpretation of a very simple piece of music.

The two numbers, one above the other, written near the beginning of the first staff line (or whenever it changes), specifying the number of beats in a measure and the type of note that gets one beat.

Milonga came before tango and its characteristic rhythm was, and is, the Cuban habanera. Early tango's rhythm was based on the habanera also. Soon other syncopated patterns dominated. The 1st variation, the "syncopa" is the most commonly heard syncopated rhythm in tango music. All three versions are used in milonga.

Habanera Variations describes the syncopation and relationship between these patterns.

Beat is the underlying and regularly spaced pulse of the music, measured in beats per minute. There are a fixed number of beats in a bar, indicated by the time signature. Tango (2/4, 4/8, 4/4) has 2 or 4 beats per bar, vals (3/4) has 3 and milonga (2/4) has 2.

(There may be a sense of 4 beats even though the time signature is 2/4. Tango very often subdivides the 2/4 beat, doubling the count from 2 to 4, effectively using a 4/8 time signature. Some tango music is explicitly written in 4/8, most are in 2/4. See Tango Time Signatures and the Beat).

A bar or measure is a small segment of music containing all the number of beats as specified by the time signature.

An orquesta típica is an ensemble of musicians who play tango music. Typically,  there is a string section, a bandoeon section, a piano, and sometimes a singer or two. There is no specific rhythm section – no drums or other percussion instruments. An orquesta típica is an expanded version of a sexteto tipico, which includes 2 bandoneons, 2 violins, double bass, and piano.

I call any band that plays tango, no matter what the instrumentation, an orquesta. Not entirely accurate but it simplifies things.

The two numbers, one above the other, written near the beginning of the first staff line (or whenever it changes), specifying the number of beats in a measure and the type of note that gets one beat.

13 Responses to Tango Time Signatures and The Beat

  1. Jantango says:

    Eureka! Gold at last! This is the treasure I’ve been waiting to discover.

    You confirmed my instincts based on my musical training and what I feel in the music when dancing tango. I have scores in 4/8 — Dejame vivir mi vida by Roberto Rufino, Pastora by Alfredo De Angelis, Que me importa tu pasado by Retama — but I didn’t know the reason why 4/8 and not 2/4.

    Most tango dance teachers cannot talk about the music, because they lack musical training. This post is perfect for them to study or at least recommend to their students.

    Gracias! Your music lecture is priceless.

  2. Henri says:

    This is a great post, thank you!! Could you do something similar for milonga and vals? Beautiful work and truly well done. Instructive for dancers, musicians, and anyone who knows some music theory and wants to increase their appreciation of tango!

  3. Lenore says:

    An Argentine Tango novice dancer without a musically trained background your fabulous post gives me hope I can improve my tango musicality. Appreciate the sharing : )

  4. Theresa Faus says:

    I love this post! I’m always irritated when Argentineans speak of the fundamental difference between 2/4 and 4/4.
    But I’m a bit confused with the scores you show in the examples of di Sarli and d’Arienzo. The scores do show the Habanera rhythm which is absent in the recordings.
    Theresa from Munich

    • tangomonkey says:

      Thanks. There is a difference between 2/4 and 4/4, but the tango musicians subdivided the 2 beats in 2/4 to 4. In effect creating a four count in a 2/4 time signature. Eventually the more accurate time signature, 4/8. was used. In tango 4/4 comes later – Piazzolla and some others. But subdivided 2/4, 4/8, 4/4 all have a four count.

      The piano sheet music is what was originally printed and not meant to be a score as such. The habanera rhythm was used in early tango but soon another couple of versions were used instead. The original habanera remained one of the primary rhythms in milonga. I used the scores mainly to show the sections and phrases and the basic melody and harmonies.

      There are a couple of posts on the habanera and it’s versions, here, and here.

  5. contrabajo says:

    I find it remarkable that in tango history we talk about the transition from 2/4 to 4/8 at all, let alone make a big deal about it. It is such a technical detail! Statistically, the fact that you can’t find that many examples of tangos in 4/8 is suspect to say the least. I think everything points to a big confusion., and you acknowledge that in many parts of your essay.
    But I insist, why do musicians and historians mention this arcane technical aspect as a significant development in tango music? Was there a historical moment when something happened to the music THAT COULD ACTUALLY BE HEARD? Because as you demonstrate, simply changing from 2/4 to 4/8 does not produce any difference… What are people talking about, really?
    I am very happy to have read this article too, because it has confirmed for me that we are looking at the explanation of this phenomenon in the wrong place. I believe there is a historical moment when this occurs, and it is timed somewhere in the early 1920s around De Caro.
    But is it metrical, rhythmic.. or something else?
    I think I have the answer, but I won’t write about it here, sorry. Among other things I have to continue my research, and “prove it” to you. But I could do it using the examples in this page. It’s right in front of you so to speak, but you need to gather more elements to understand what happened.
    Pablo

    • tangomonkey says:

      Hi Pablo

      Nice to hear from you.

      Using 4/8 (and later 4/4) instead of 2/4 is not a significant development in tango music. (Don’t believe I said or implied it was). Some (non-musicians) make statements that there is an important difference between 2/4 and 4/8 tangos – that four beats makes, or allows, the music to be different

      It is the recognition that music with a 4 count should have a “4” in the top of the time signature. To have 4 beats in 2/4 (marked almost continuously for three minutes or so) is a kludge, a breaking of the rules and purpose of time signatures. Using 4/8 corrects that. A technical point of course.

      It is the rhythmic change from marking 2 beats per bar, using the habanera rhythm, to 4 beats per bar (without, or seldom using the habanera) that matters. And that happened while still using 2/4, during the 1910s with Canaro and Firpo. For quite some time the four count was march-like, almost always present. Later works rely less on a consistently marked four count, marking beats 1 and 3, using a two count and other rhythms (habanera relatives for example). And it is common to have more than one of these rhythmic techniques in the same piece of music; another style change.

  6. contrabajo says:

    Well, yes, except tango historians literally claim that:

    En una carta al diario “La Nueva Provincia” de Bahía Blanca (20-5-98), el Señor Vicente Ramón Ciappina hace mención a varios documentos sobre este tema:

    En el diario “Crítica” del 22 de septiembre de 1913, con la firma de “Viejo Tanguero”, se dice que “fue en este período que se produjo la mayor modificación de la historia del tango: el pasaje del 2/4 al 4/8”.

    H. Bates, en su “Historia del Tango”, dice: “La cumparsita fue el primer tango que fue escrito originalmente en 4/8. Este cambio fue tan importante que cambió para siempre el ritmo de nuestro género musical”.

    excerpted from
    http://www.taringa.net/post/info/2029616/Tango—Por-que-no-hay-que-llamar-2×4-al-tango.html

    • tangomonkey says:

      I have yet to read anything which causes me to believe the change from 2/4 to 4/8 was anything more than a formal recognition that the music had adapted (while still in 2/4) to music with 4 beats. As I said in the article, I believe the reason 4/8 was used instead of the vastly more common 4/4 was simply because the music looks the same in 4/8 and different in 4/4. As a musician I prefer 4/8 (or 4/4). If there is a four count I’d rather count 1-2-3-4 than 1&2&, but of course that doesn’t matter to listeners/dancers, or to how the music sounds.

      The link you included talks about 2 by 4 (dos por quatro). I’ve seen/heard the term 2 by 4 refer to the old habanera based tangos and the marcado in 4, accenting beats 1 and 3: 2 by/in 4. Personally I don’t use the phrase because there is – at least in English speaking countries – too much confusion.

      I believe you agree the change from 2/4 to 4/8 is largely technical. As musicians we understand that. Tango music changed drastically in the first couple or so decades of the 1900s, but the time signature is an irrelevant factor. I wish non-musicians would stop making it seem like a watershed event – especially since composers/arrangers continued to use 2/4 as well as 4/8 (and 4/4 in Piazzolla and his contemporaries, and today).

  7. contrabajo says:

    the two citations above are important tango history documents: the Viejo Tanguero article is often cited (Gobello, for example), and the Bates Historia is the first book on tango history (1936).
    However, tango history is mostly “folkloric” non-academic, and the Bates book in particular has received a lot of criticism for perpetuating myths and untruths about the origins of tango in particular (see Lamas, Binda)
    What strikes me is that there is a discussion in this amateur world, of a “largely technical” aspect of music notation. Why?
    This makes me think there is something else going on, that, as I said in my first post, something DID happen with meter, … or something else, and these “historians” are not able to describe it, but it is talked about among musicians.
    The tango did not just go from 2/4 to 4/8, which, as you point out, can only be seen on the page, and not heard, and it is largely a concern for musicians.
    Tango slightly before 1920 not only starts to be felt in 4, it slows down, significantly. Compare Vicente Greco’s 1911 recordings with early De Caro (1924-26). There is about a 30% drop in tempo. That’s not a trivial detail, to me this is at the core of “what happened”.
    Like I said, I am still researching, metronome in hand, but I have no doubt that the crux of this historical moment (see above “la mayor modificación de la historia del tango” says one, “Este cambio fue tan importante que cambió para siempre el ritmo de nuestro género musical” says the other source!) is to be found in tempo, not meter.
    Like you said: “Tango music changed drastically in the first couple or so decades of the 1900s, but the time signature is an irrelevant factor.” I completely agree. So ask yourself, what happened? How do you describe the change in musical terms?
    Tempo.

    • tangomonkey says:

      Tempo is certainly an important factor. Early tango music was very similar to milonga in tempo and rhythm. To stand on its own, tango music had to differentiate itself from milonga. Tempo slowed and milonga’s repetitive habanera rhythm was dropped, and its cousins, the syncopa and 3-3-2, given prominence. And the underlying pulse became a simple marking of the beat. I’m not sure exactly when a four count happened – not many recordings available – but by the time of early Firpo and Canaro (and others) in the 1910s it’s there, as you note. The tempo is still quite fast, slower than milonga but faster than what was to become usual. Slower tempo allows for lyricism, and that was something new and very influential (De Caro).

      This all very general and not the entire picture. But I think it’s on track

  8. Chris Scott says:

    I am a complete novice to the tango, or any Latin music, but thank you for a clear explanation. (I often find it confusing to have music counted contrary to the clear feel of the beat and this article explains some of that happens.)
    I wonder, on the point of tempo, whether it is worth looking at the habanera in use in the classical tradition. Thinking of composers such as Bizet (Carrmen) and Debussy (Puerto del Vino, for example) the tempo that they use is certainly quite slow. It would make sense that the French “love affair” with things Spanish in the 19th C meant that the modes of music were probably transcribed accurately – Debussy had perfect pitch and travelled to Spain, so is probably a good source.(Interestingly, Bizet doesn’t dot the initial quaver, but instead opts for quaver-semiquaverREST- semiquaver. I guess this was to accentuate the articulation of the strings, or maybe anticipating the reverberation of a large venue.) Incidentally, both Bizet and Debussy chose to write their habaneras in 2/4.
    Anyway, I suppose my timid point is that the tempo may well have increased and then been restored – a simpler reason for tempo change is that it was in response to changes in taste on dance floors?

    • tangomonkey says:

      Hi Chris

      Yes, the French had a fascination with the “exotisism” of Spain and the use of the habanera rhythm in classical works reflects that.

      It isn’t clear whether the habanera rhythm originated in Spain (in some form or other) and was brought to the Americas, whether it is an African rhythm, or that it is a combination of both. There are historical cases to be made for each view.

      The Contradanza wiki page seems quite accurate, from what I’ve read in more definitive and trustworthy sources. It provides some interesting history.

Leave a Reply to Lenore Cancel reply