The Golden Age Visualized

2009 October 16
by Simba


When was the Golden Age of tango, exactly? It is fairly common to say 1935-1955 mas o menos, but while I was playing around with the data at, I thought it would be fun to make a plot to see when the Golden Age was. And here it is.

These are the number of unique recordings in the database, so if a recording is issued on several cds, it is only counted once. I only used data with a positive genre identification and positive year, not necessarily month or exact date, so there may still be some small errors, but it should be fine, I think. This gave about 12k data points, which if the number of 100k tango records in total is correct, gives us a sample of about 12%.

I was a bit surprised that the Guardia Vieja had such a large number of recordings, as I suspected the data, if skewed, would over represent the golden age. It is quite telling, that the interest for tango seems to fade away at the end of the Old Guard, then suddenly comes back in the early 1930s with a new spike in the 1940s. The interest is certainly vaning again by the end of the 1950s, so I’d say the usual definition is a good rule of thumb.

It is also interesting to see that the number of recordings of valses and milongas is so small compared with tango. And that the valses start coming earlier than the milongas, which had a comeback in the tango world in the early 1930s.

The data quality is steadily getting better at, so I should revisit this in a couple of years, or if someone knows about a better source, I’d be happy to know. If you  have time to investigate discographies for an album where some information (e.g. dates) is missing, I know Tobias is very happy to receive contributions.

Coming up are plots by orchestra, they become very messy when mixing too many, so three or four at the time seems to be the sweet spot. Any wishes for artist combinations you’d like to see?


Here is one example, Canaro, D’Arienzo and Di Sarli. I knew Canaro was very productive, but wow, over 70 recordings in one year!

The raw data are available at I thought it would be a nice subtle tango reference in my work environment to have a coffee mug with this graph. Now you can have one, too 😉

4 Responses leave one →
  1. 2009 October 17

    haha…Simba, that’s an interesting coincident. I have been listening and buying quite a few 20s music these couple of days. I was reading a Harlequin’s cd pamphlet yesterday. And it claims the golden age is between 20s-50s. I guess that the reason why 20s isn’t mentioned or heard or played as often is because most of the pieces were lost in years. The remaining tracks are mostly in poor conditions. But I must say that there are very interesting and beautiful pieces in 20s.

  2. 2009 October 18

    Interesting. That would cover almost all recorded tango. I still think it is more usual to use the ‘golden age’ for the period after D’Arienzo revitalized the tango (dance) scene with the return to 2×4 in 1935 (for which we can at least partly thank Biagi) and that it ended somewhere in the fifties. The Baroque period is often said to have ended with the death of J.S. Bach in 1750, so maybe the death of Carlos Di Sarli in Jan 1960 marks the end of the golden age of tango?

    No doubt there is a lot of good stuff from earlier times, but I always assumed it was the other way around with the recordings of the 20s, that they were unavailable because the really great stuff was mostly from the 40s, give or take a little, and thus in higher demand.

  3. 2009 October 19

    I was on the phone a couple hours ago with Carlos Alberto Anzuate (79) with whom I chat daily so he can take a break from attending his wife who has Alzheimers. I asked him, “when was the golden era of tango?” to which he replied, “1920 to 1940.” Your first chart confirms it. Thanks for the informative post.

    I look forward to meeting you and your family next month in Buenos Aires.

  4. 2009 October 20

    Thank you Janis,

    maybe it does, I am not quite sure. My immediate interpretation was that the three peaks correspond to the classification Stephen Brown uses, starting with ‘Old guard’ and ‘Early Golden Age’ before ‘Golden Age’. But giving it some more thought, that might be a very dance-centric way of seeing it. If you want to count in Carlos Gardel, for instance, he died in 1935, and wouldn’t make it. It seems that the term is somewhat ambiguous, but including the 20s makes a lot of sense.

    Only two short weeks now 🙂

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS