Osvaldo Pugliese
Osvaldo Pugliese & Nicolás Lefcovich:

8 controversial recordings

Nicolas Lefcovich
1. The Lefcovich statement
2. Recording details
3. Do they exist ?
4. Searching through modern CD editions.
N.N. - update
5. Conclusion

(enable JavaScript to play the sound samples)


1. The Lefcovich statement

Nicolas Lefcovich is a well known tango investigator and discographer, unfortunately deceased a few years ago (Sept., 2004). Many of his more extensive works, like the discographies of Francisco Canaro and Roberto Firpo (each several thousands of titles) are unique in its kind. Also, his Pugliese discography is still a classic reference.

In the chronological overview of the recordings by Osvaldo Pugliese, a strange remark can be found when arriving at the end of the year 1952. Eight recordings are involved: the legendary "La Yumba" and also "N.N.", "La Tupungatina", "La Cachila", "Pasional", "Mala Junta", "Barro" and "La Noche Que Me Esperes". All these titles were already mentioned earlier (between 1943,  the year when Pugliese started recording with his own orchestra for the Odeon label, and that same year 1952) but about these particular recordings Lefcovich writes the following:

"(*) These 8 recordings may be laboratory products.

On various occasions I have asked the maestro Osvaldo Pugliese if he had recorded two times his tango "La Yumba" and always he said that he didn't. The same happened with Osvaldo Ruggiero [1], in the presence of Oscar Del Priore [2], he told me that he didn't remember to have recorded these themes more than once.

I have decided not to accept as certain that the orchestra recorded these themes two times, although the recordings do sound different. Based on the information that I collected, I concluded that these recordings are the result of laboratory manipulations and consequently I didn't number them [3]. It remains difficult to believe that in a laboratory, even the voice of Alberto Morán could be manipulated. But the truth is that these versions exist and, strangely enough, these recordings were published on long play records, and not the original ones."

(from the 1995 edition, but very likely written much earlier)

[1] Osvaldo Ruggiero: bandoneonist and arranger of Pugliese's orchestra (from the 1st orchestra in 1939 until the formation of the Sexteto Tango in 1968), also composer.
[2] Oscar Del Priore: renowned tango historian and publicist
[3] In his discography, Lefcovich labeled every recording with a unique number.

When read from a large distance in both time and space, the rapid jump from an LP edition to a laboratory product feels rather discomforting, especially with the archaically sounding reference to a "laboratory".
Nowadays, we live in the age were the "remastered / remasterizado" label is a selling argument on CD editions of tango music. This is not the first time in history though : already in the late fifties and sixties, many of the old 78 rpm records (discos de pasta) were reedited on vinyl LPs (larga duración, vinilo). As can be seen in the following citation, taken from such a reedition (in this case of Carlos Di Sarli with Roberto Rufino, and for another label, RCA Víctor) sound improvement and error corrections were already used. Interestingly, the same term "laboratory" also occurs. Now that we have our PCs loaded with all kinds of sound editing software, it may seem unusual to us, it certainly wasn't in those days.



Because of the constant demand for these jewels of popular and classical music, RCA VICTOR ARGENTINA, as a result of patient work in the laboratory, has composed this Long Play using the original masters from its historical archive [1]. These masters were recorded some years ago [2] with equipment, inferior to the one that is now available, and moreover, suffered from the action of time. They were reprocessed in our modern studios to eliminate deficiencies and original shortcomings. Nevertheless you will still encounter certain inevitable imperfections which every admirer will be able to ignore."

[1] This text was very likely published in 1960, maybe one year earlier. Just a few years later a considerable part of this "historical archive" will be deliberately destroyed.
[2] This LP ("El maestro y yo") contains Di Sarli recordings with Roberto Rufino made between 1940 and 1943.

Now that the Lefcovich statement becomes a little bit more acceptable, the question arises : does it still hold ?
But first the details of the recordings...

2. Recording details

In the following table, the data of the "first recordings" are juxtaposed to those of the alledged "second recordings", all as given by Lefcovich. In the right-most column, I have also added 2 titles ("El Embrollo" and "Manos Adoradas") which were recorded in the same period (not questioned by Lefcovich), they are not "second recordings" but chronologically they belong there. The acoples, the accompanying titles on the same record, are inferred from the record numbers as given.
Title, authors and,
optionally, vocalist
First recording
(Date - Record number -
Master number - Acople)
See also: Short note on 
tango record identification
Second recording ?
(Date - Record number -
Master number - Acople)
See also: Short note on 
tango record identification
El Embrollo
(Esteban Enrique Gilardi)
(Manos Adoradas - with A. Morán)
La Yumba
(Osvaldo Pugliese)
(Mal de Amores - instr)
(N.N. - instr)
(Osvaldo Ruggiero)
(Desilusión - with A. Morán)
(La Yumba - instr)
La Tupungatina
(Christino Tapia)
(Ahora No Me Conoces - with A. Morán)
La Cachila
(Eduardo Arolas)
(Mentira - with A. Morán)
(Jorge Caldara - Mario Soto)
with Alberto Morán
(Bien Milonga - instr)
(La Cachila)
Mala Junta
(Julio De Caro - Pedro Laurenz -
J.M. Velich)
(Muchachos Comienzo
La Ronda - with R. Chanel)
(Barro - with A. Morán)
(Osvaldo Pugliese - Horacio Basterra)
with Alberto Morán
(De Vuelta Al Bulín - with A. Morán)
(Mala Junta - instr)
La Noche Que Me Esperes
(Juan Canaro)
with Alberto Morán
(Cualquier Cosa - with A. Morán)
Manos Adoradas
(Roberto Rufino - Horacio Sanguinetti)
with Alberto Morán
(El Embrollo - instr)
Obviously, Lefcovich felt somehow lost with these recordings. The master numbers were just copied from the original recording with "/1" appended, the conventional indication for a "take-2" recording. The differences (15758 vs. 18758/1, 18137 vs. 17137/1 and 18339 vs. 18939/1) have the appearance of transcription errors. This is not a regular feature in the work of Lefcovich but could have been caused by the decoupling of master number and recording date, making it more difficult to spot errors.
On the other hand, Lefcovich must have had access to documents which provided recording dates and record numbers. These numbers sometimes come pair-wise, indicating regular 78 rpm records, contrasting with the "long playings" mentioned in his comment. And yet, Lefcovich concludes that these recordings may not exist.

These observations should not be understood as a depreciation of the work of Lefcovich, on the contrary. They show that Lefcovich, in spite of a few printing errors that slipped through, was very keen on checking the factual basis of his data, preferring a possible omission over the creation of a chimerical recording.
After all, a good discography is one to which titles have to be added. A bad discography is one from which titles have to be removed.

3. Do they exist ?

Lefcovich acknowledges the fact that some editions exist that do "sound different". Could they really be rerecordings ?

A first observation is that Pugliese, unlike others (Di Sarli, Sassone, Canaro, D'Arienzo...), did not record very often the same title twice. Not counting the live recordings (like the Teatro Colón concert in 1985) and the titles discussed here, there are only some ten cases. Not much in a repertory of more than 400 recordings, but they show that the fact cannot be excluded in principle.

"Los Grandes del Tango" devoted an issue to Pugliese (N°7, December, 1990). However charming these booklets are (more than 100 appeared between 1990 and 1993, each about 20 to 40 pages), their discographies are not among the most reliable and complete. In the Pugliese discography they mention only single recordings for "La Noche Que Me Esperes" and "La Tupungatina" in 1952 but they do list "La Yumba" and "N.N." in that year. "Barro" and "Pasional" are mentioned in 1951, but not in 1952, neither are "La Cachila" and "Mala Junta".

The Buenos Aires Tango Club published the second part of the Pugliese discography (from May, 1952 on) in the Revista del Buenos Aires Tango Club N°7 (July, 1999). None of the Lefcovich duplicados is mentioned.

On Todo Tango, recognizing Lefcovich as a source, the 8 recordings are listed, with the same details as given above but they ommit the Lefcovich comment. Because it didn't fit into their tabular format ? On the other hand, among their sound samples ("La música") both "La Yumba" and "N.N." are identified (by date and numbers) as the second recordings.

Interestingly, Roberto G. Miglio, in his discography of Alberto Morán (El Tango y Sus Intérpretes - Vol.2), mentions for each of "Pasional", "Barro" and "La Noche Que Me Esperes" the two recordings. In the case of "Pasional", the double recording not only occurs in the listing but is also explicitly mentioned in the short biography that preceedes it.

But most fascinating is a communication by Oscar Del Priore (already called as a witness by Lefcovich himself !) which was transmitted through the Cueva de Tango at the end of 2003. He assured that these titles were indeed recorded twice and added that Pugliese took the opportunity to work with the new recording technologies that had become available (magnetic tapes - until that time, music was recorded directly onto the master disk, in the early fifties magnetic tape became available as the first recording medium, which was then transferred onto the master disk in a second step).

Different opinions...  I guess that only Odeon documents, if they exist, could settle the case. Apparently, some of the editions "which sound different" have made it on to CD and from what I hear I feel tempted to consider some of them indeed as different recordings. Would it be logical to think that some technician tries to improve Pugliese, turning knobs and dials to insert fragments or modify parts, instead of being a musician himself? Moreover, I think that Lefcovich overestimated the capabilities of sound engineers at that time.


 Osvaldo Pugliese &
Nicolas Lefcovich
(updated 08-2007)
Part 2


Short note on tango recording identification
Old and "golden era" tango recordings are often identified by 2 numbers
- a master number (numero de matriz) , this identifies the physical carrier of the recording (on relatively soft material, but also the nickel alloy negative copy used for producing the records)
- a record number (numero de disco or numero de etiqueta), which identifies the edition (on 78 rpm shellac disks).
When, during a single session, several "takes" were recorded (it did happen but it was certainly not a standard procedure) it was common practice to reuse the same master number with a suffix: e.g. in 1927 Pedro Maffia and Alfredo de Franco recorded two takes of the tango "Diablito", the corresponding master numbers are 1048 and 1048/1). Modern editions of "previously unreleased material" often refer to these earlier takes (which, by the way, were not considered good enough for editing at that time). Generally, master numbers are a good indication of the chronological order in which recordings were made. Nevertheless, sometimes the companies changed the numbering system of their masters which leads to some discontinuities.
Record numbers may have A and B indications, implying some hierarchy between the two sides, but more neutrally the "other side" is often referred to as the acople. Again not common practice, but it did happen occasionaly, that the same recording (that is : the same master number) was reedited on another record, with another acople, breaking the strict one-to-one relation between master and record numbers).
In the early tango days, recording was a relatively straightforward process (at least when seen with modern eyes). Very often the two sides of a record were recorded on the same day. For the mayor companies (like Odeon and RCA Victor) recording dates can be identified. For others, like Brunswick and Electra (which did not survive into modern times) but also Music Hall, exact recording dates are not always known, even sometimes the exact recording year is open for debate (especially in the 1929-1931 period).
For modern recordings, with separate recording of each instrument and post-production becoming more important, concepts like "master number" and "recording date" don't make sense anymore... The necessary technology became available, slowly, from the fifties on, (four-track tape recorders became standard in the 60s) where we see for the first time separate recordings of orchestra and vocalist. For instance, that is how Philips could couple the voice of Edmundo Rivero to instrumental tracks of Carlos Di Sarli (although they had been edited before with Jorge Duran).
Back to the table