The Complete Recordings

hello and welcome to early music sources.com, my name is elam rotem and in this special episode we won't discuss renaissance or baroque music, we will talk about historical recordings from the beginning of the 20th century. as early music lovers we are very much interested in old performance practices, the way music was played at the time of its composition. it is clear to us that music from 400, or even 200 years ago was performed very differently than it is performed by today’s leading performers. but what about more recent times?

thanks to historical recordings we have a window to the performance practices of roughly 100 years ago. a time of many great composers, and many great musicians who studied with the great masters of the 19th century. although these recordings are indeed close to the 19th century and the romantic era, it is still quite a leap to say that these recordings are a faithful representation of the romantic performance practices. however, the differences between what we hear in historical recordings and what is considered today as the best performances of the classical music world

are so substantial that they cannot be ignored. they raise many questions about the basic concepts of modern performance, be it the “standard” classical music scene or the “early music” one. in this episode we will not attempt to give any concrete answers, but will merely present, for those who never heard, some examples of historical recordings, and try to distill in what ways its characteristics are different from modern practices. we’ll start with an excerpt from tchaikovsky’s swan lake, as played by the london philharmonic orchestra and conducted by john barbirolli, on july 20th 1933.

we should probably stop here. on our special page you could find all the links to these recordings, so you can always continue the listening later if you wish. now, without saying much, we’ll listen to a recording from nowadays. ok, this should be enough for now. it is clear that there are big differences between the two recordings, but can we point out what the differences are exactly? let’s try. first, the historical recording had a much faster tempo than the new one. then, it’s evident that the oboe soloist

is very much more independent from its accompaniment, almost as if he were often arriving too early. not only does he rarely meet the orchestra on the beat, but also his inner rhythm is rather special, he adds a certain inegalite to his eighth notes. listening to the new recording we hear that the notes are pretty equal in their length, and also, that the solo is perfectly synchronized with its accompaniment. back to barbirolli, we see that when the brass enters,

the tempo is suddenly slower than the first tempo. in the modern recording however, the tempo stays the same throughout. so, these were quite remarkable differences for only one minute of music. but is this a special case? or can we find similar characteristics in other recordings? let’s see. in this example, we’ll listen to frederic lamond playing a piece of his teacher, franz liszt: liebestraum no.3. now, we’ll listen to the same excerpt, but played by a very famous pianist, only one generation later than lamond.

so, first of all, we notice that the older recording has a rather flexible tempo, it goes from slow to quick and quick to slow within short periods of time. in both performances the melody is clearly stronger in its dynamics than the accompaniment, but in the older recording the dynamic range is wider and more nuanced, both between the melody and the accompaniment, and within the melody itself. furthermore, dynamics are not enough for lamond, just like we heard in barbirolli’s swan lake, he separates the melody from the accompaniment by having it rhythmically independent. practically, it means that the melody is almost never together with the beat of the accompaniment.

this was nice, check the rest of it on our website. after hearing these examples one may wonder if the performers simply allowed themselves to take greater freedoms in their interpretations than in later times. if only we could listen to the actual composers play their own music… oh but we can! there are several historical recordings of composers performing themselves. let’s listen to grieg playing grieg. here is a recording from 1903 of grieg playing his piece “to the spring” from op. 43.

like in many historical recordings, we have to make an effort to hear beyond the problematic recording quality. and here is a performance by a prominent pianist from nowadays. going back to the performance of the composer, we hear that the tempo is significantly faster, the accompanying chords in the right hand are much lighter, and thus, the melody is clearer, both in its dynamics and in its shape. having recordings of composers performing their own music is something extremely precious, something that fans of earlier musical styles can only dream of.

check our special page for some more such recordings. now, let’s listen to some vocal recordings. here is a performance of the famous habanera from the opera carmen, sung by conchita supervã¬a in 1930: and now, a more recent performance by famous and acknowledged performers: going back to 1930, we hear that the performance is faster, lighter, and the ostinato in the bass has a certain irregularity to it. the singing has a more declamatory character in comparison to the newer version, the note lengths are dictated more by the text than by the written note values

furthermore, there are more tempo changes, some of them quite extreme. portamento is a feature found very often in historical recordings. in this case also the modern singer did it, but one should note that in this aria it is explicitly written in the score. in this example, with adelina patti singing mozart’s famous "voi che sapete", we can hear how she used the portamento very often. notice, patti was born in 1843, 15 years before puccini(!). another thing we can very clearly hear in this example, in addition to the typical tempo fluctuations, is the almost non-existent vibrato,

a feature that we would have more difficulty finding nowadays. visit our special page to listen to the rest of this example as well as other recordings of famous singers from the beginning of the 20th century. so we heard some singing. singing was always a source of inspiration for instrumentalists, and especially for violin players, who could imitate their glisandi and portamenti. here we can hear a recording from 1925 of renã©e chemet playing saint-saã«ns. here we can enjoy the tempo fluctuations we by now expect, the dis-synchronization of the melody and the accompaniment,

the very light and sometimes non-existent vibrato, and the extensive use of portamenti. when the early music movement started, it was common to say that the way of playing of the “modern classical scene” is “romantic” and therefore not appropriate for music composed before the romantic era. however, after listening to recordings from the beginning of the 20th century this statement seems rather doubtful. the countless recordings of students of the great romantic masters demonstrate very different characteristics

than those of the performance practices nowadays. here is a little summary of these characteristics. the research into historical recordings is a very broad field, it is therefore clear that the points brought up here are rough generalizations. tempo: the tempo tends to be more flexible and often changes throughout the piece. also, in many cases, the basic tempo is simply faster. melody: the rhythms of the melodies are often slightly altered. series of notes can be played with a little inegalite for example.

in modern practices there is a more literal interpretation of note values. melody vs accompaniment: the melody and the accompaniment are often not well synchronized. the melody is arriving very often after or before the beat of the accompaniment. portamenti & glissandi: in recording of singers and string players one can hear a great deal of portamenti & glissandi. vibrato: the vibrato of singers and instrumentalists is more varied than in modern practices, and there is also the possibility of not using vibrato at all.

ornaments and passaggi: in many recordings one can hear ornaments and passaggi that are not written in the music. check our special page for some nice examples of that. now you might think that all these things are related only to “classical music”, but in a lot of popular music from the first half of the 20th century one can find similar features. after we’ll say goodbye we’ll play for you “somewhere over the rainbow” performed by judy garland, only 17 years old at the time. when you listen, notice how almost no note of hers

is sung together with the accompaniment, notice her beautiful glissandi, the subtle and not so subtle tempo changes, and her very light and nuanced vibrato. there is still so much to say about historical recordings, this was only a little introduction and we hope you enjoyed it. special thanks to gilad katzenelson for all his insights. don’t forget to check the special page on our website with all the footnotes and extra information.

feel free to comment share and like, see you next time at early music sources.com.

Share this :