One of the most grown-up review sites around
One of the most grown-up review sites around

Search MusicWeb Here


International mailing

  Founder: Len Mullenger
Classical Editor in Chief: Rob Barnett


Some items
to consider


Leticia Gómez-Tagle (piano)

Gloria Dei Cantores
All Night Vigil

Whispering Mozart

absolutely essential release

The best recording I will hear all year

Long awaited recordings

Absolutely enthralling

Left me longing for more

A superb 90th birthday recording

A top recommendation.

Superb advocacy

Tchaikovsky Sibelius


Complete ballet

Janáček’s Glagolitic Mass

Part One

Leoš Janáček (1854-1928) composed his Glagolitic Mass (Mša glagolskaja) in 1926-1927 at the end of his life and during the most productive period of his compositional career. It is rightly considered one of the supreme choral masterpieces of the twentieth century and has been very lucky in recordings. However, it was not until 1994 that the whole work was actually recorded. Prior to then it was heard in its published form of 1930, which was shortened and simplified from the original composition — largely due to the perceived inability of its first performers to execute the difficulties of the score. Thanks to Janáček scholar Paul Wingfield, whose edition restores the Glagolitic Mass to its original form, it can now be heard in all its glory. It is important to remember that the standard version, even with cuts and simplified scoring, is still a very powerful work although, for obvious reasons, the original has steadily gained adherents and is now being performed throughout the world.

The main differences between the two versions are as follows: the original reinstates the Intrada movement at the beginning of the mass as well as in its usual place at the end. This gives the Mass a perfect arch form with Věruju (Credo) as its centerpiece. There are meter changes in both the Úvod (Introduction) and especially in the Gospodi, pomiluj (Kyrie) where 5/4 meter was originally conceived and then switched to easier 4/4. The Věruju movement contains the largest changes, with the orchestral middle section, the "Raspet" (an orchestral commentary on the Crucifixion), longer and more complex in the original version and containing a wild section with three sets of timpani omitted in the revised version, as well as off-stage clarinets. Finally, the Svet (Sanctus) movement is extended thrillingly near the end, reaching ever higher and higher. For a detailed analysis of the Glagolitic Mass and its versions, please see Paul Wingfield’s study, Janáček: Glagolitic Mass, Cambridge University Press, 1992..

In 1953 Janáček pupil Břetislav Bakala with his native Brno forces first recorded the Glagolitic Mass in its published edition. Even though it is a powerful account of the score performed by someone who was close to the composer, its outdated sound prevents it from consideration in this comparative review. The recordings to be considered here go back to the early 1960s when the work finally received the universal acclaim that it always deserved. First, recordings of the standard version as published in 1930 will be compared; then the three that restore the original score (Wingfield ed.).

Timing (Standard version of Mass)

Ančerl, Czech Phil (1963) Supraphon (SU 3667-2 911) 39:59

Kubelík, Bavarian Radio Sym (1965) DG Originals (463 672-2) 38:13

Rattle, City of Birmingham Sym (1982) EMI GROC (CDM5 66980 2) 39:03

Mackerras, Czech Phil (1984) Supraphon (33C37-7448) 39:55

Shaw, Atlanta Sym (1990) Telarc (CD-80287) 38:35

Tilson Thomas, London Sym (1990) Sony (SK 47182) 41:07

The timings given above for the six recordings being compared indicate that most performances come in under 40 minutes, with Kubelík’s being the fastest and Tilson Thomas’s the slowest. This is relevant as will be seen below.

The recording from which I learned the Mass was Rafael Kubelík’s. I first heard it in 1966 and it was a mind-blowing experience. I decided then and there that I didn’t need any other! How time has changed my opinion. Of the six recordings, I can easily eliminate Simon Rattle’s from the start due primarily to the muffled, even mushy recorded sound. Everything sounds in a blur and the soloists are not particularly good, either. Felicity Palmer is operatic, in the wrong sort of way: too heavy with a pronounced vibrato and screechy at the same time, particularly in Slava (Gloria) movement. John Mitchinson, the tenor, is okay, but also too "operatic" for my taste. If I were to describe Rattle’s interpretation in one word, it would be "stolid." But the real culprit is the recorded sound. The chorus is often indistinct and the important timpani parts are all but buried in the murk.

Going back to the Kubelík. It has two things going for it, wonderful soloists and clear recorded sound. I would place Evelyn Lear and Ernst Haefliger near or even at the top of the competition. Lear soars in the Slava, as does Haefliger in the Věruju movement. All four vocalists shine in the Svet and Agneče Božij (Agnus Dei) movements. The recording, itself, is very good for its age. Here the timpani are crisp and clear. However, the very clarity of the sound exposes for me what ruins this recording: sub par brass, especially in both the Úvod and Intrada movements. It is terribly out of tune and in the Intrada the musicians can barely hit the notes. A comparison with either Ančerl or Mackerras will soon reveal the difference! Kubelík’s chorus is fine, if not as idiomatic as the Czechs for Ančerl and Mackerras. As a whole, I find his faster tempos to be detrimental to the ceremonial nature of the music. The Intrada is downright perfunctory, but then I imagine he wanted to get it over as soon as possible given the state of his brass players.

I don’t want to dismiss Robert Shaw’s or Michael Tilson Thomas’s recordings out of hand, because both contain some wonderful things. Shaw’s has the advantages and disadvantages of Telarc’s sound. His soloists are also good, with Christine Brewer especially fine. The main problem is in the interpretation, or lack of one. Everything seems rather matter of fact and metrical. He starts the Věruju faster than any of the others and makes it seem perfunctory. On the other hand, he has what might be the best of all organ solos in the seventh movement. This was "piped in" separately from a church and that could explain why the sound is somewhat clearer and better than elsewhere in the recording. Overall, the sound is big with a full bass response, but at times the chorus disappears in it when it should be heard. Comparison with the much older Ančerl recording will demonstrate this. What makes Shaw’s recording indispensable is the Mass’s disc-mate: one the best recordings I’ve ever heard of Dvořák’s infrequently performed Te Deum. It blazes where the Mass should but does not. A real pity.

Tilson Thomas’s is in many ways a frustrating account. For me, it has the best soprano soloist of the lot in Gabriela Beňačková and the worst tenor in Gary Lakes. Just listen to the Slava, for instance. Beňačková is perfect in every way — not too heavy or operatic, but with a real Slavic tone. Lakes, on the other hand, sounds as though he would rather be singing Puccini or some romantic ballad — totally unidiomatic. You don’t want crooning in Janáček! Thomas’s interpretation tends to be more romantic than the others and drags things out, such as the ending of the Gospodi pomiluj movement. Then he is too smooth in the Svet and rather featureless. The chorus and orchestra, however, are exemplary. The recorded sound is a little too "live," and the important timpani parts — especially in the Intrada — get swallowed up in a roar of sound.

As you may have guessed, I am saving the best for last. In this case I have to choose two, both with the venerable Czech Philharmonic and Prague Philharmonic Chorus. It is not only because the Czechs have the music in their blood, but also because of their recordings as recordings. While the recordings made by Karel Ančerl and Charles Mackerras have much in common - the timings - there are enough important differences between them that collectors should have both. Ančerl’s comes from 1963 and the one from Mackerras from twenty years later, yet the sound of the earlier recording is in no way inferior to the later one. Ančerl is recorded up close, with the chorus and soloists practically in your face; Mackerras at a greater distance and at a lower volume level. In fact to make the Mackerras recording sound as full as the Ančerl requires an upper adjustment on the volume control and a slight reduction in the treble and increase in the bass. Once this is done, the performance comes across just as powerfully as Ančerl’s.

The chorus and orchestra perform their hearts out for both conductors. You can understand every word and the orchestral interjections are thrilling. I marginally prefer Ančerl’s soloists, but the soloists in both recordings are fine. Ančerl’s soprano, Libuše Domanínská, and legendary tenor, Beno Blachut are very idiomatic; Mackerras’s Elisabeth Söderström is very good, too, but better in the Svet than in the Slava where her voice tends to shrillness at times. František Livora, while certainly doing a creditable job, is not in the Blachut class. Neither soprano eclipses Beňačková in the Tilson Thomas recording, though. The alto and bass, with their less important roles, are fine in both recordings. The important organ solos, however, come across much better in the Mackerras. This is largely due to the sound of the organs, rather than the organists’ performances. Both organs have a reedy quality that is attractive in this score, but Ančerl’s also has “theatre” organ quality that is not present in the Mackerras and detracts from the performance. I don’t know if this due to the recording or the actual sound of the organ, but it is the reason I have strong preference here for the Mackerras. That said, both recordings project the organs powerfully if without quite the presence of Shaw’s.

Now to specific differences in the individual movements:

Úvod: Both open the great mass with fervor. Ančerl’s tempos are slightly broader than Mackerras’s but the overall effect is similar. The orchestral brass is splendid in both, but the timpani register better in the Mackerras and are especially telling later in the movement.

Gospodi pomiluj: Here it is Ančerl who is slightly faster, but again there’s very little in it. I prefer Domanínská to Söderström here, because of her more idiomatic, Slavic timbre, but both sopranos are excellent. Ančerl’s chorus seems to have more bite at times, perhaps because of the closer recording.

Slava: Ančerl is powerful, but arguably too slow — especially at the end where his timpani are heavy and mushy and tempos really drag; Mackerras’s slightly faster tempos seem perfect to me and the timpani and organ at the end have real bite. As I noted above, I prefer Domanínská to Söderström in this movement, too.

Věruju: Ančerl’s tempos are spot-on - not as fast as Kubelík’s nor as slow as Rattle’s - and his tenor solo is wonderful. The chorus is terrific, clear and incisive; characterful woodwinds, but that “theatre” organ sound again! Mackerras’s tempos are also fine as is his tenor, but his chorus is not as clear as in the Ančerl where the close-up sound allows for better diction. Woodwinds are again special and the orchestral interlude is terrific with a better organ than Ančerl’s.

Svet: Both conductors set ideal tempos, if Mackerras’s seem slightly slower in the beginning but faster from the "plna sut nebesa …" dance section. The soloists in both are excellent; I like Söderström better here than I did in the Slava movement. Mackerras keeps the performance exciting up to the end with blazing brass and timpani. Note that he includes fourteen bars near the end that were cut from the standard published score. Ančerl, who of course used the only available version, sticks to the standard score. His is also tremendously exciting, even though his timpani make less impact at the end — due in part to the age of the recording.

Agneče Božij: While both are excellent in this movement, Mackerras has the advantage of a more modern recording and greater dynamic range. This is especially important the last time “pomiluj nas” is sung much more softly than before. Ančerl’s closer sound and louder dynamics rob this ending of some of its mystery.

Organ Solo: As noted above, I have a preference for Mackerras. Ančerl’s is still very powerful — a big, clear sound where you can hear everything that’s being played. It’s just the tone of the organ that bothers me. Mackerras is also powerful and exciting, and clear. You just have to make sure you have the volume turned up, as mentioned above.

Intrada: Ančerl’s tempo at the beginning seems somewhat slower than the others — certainly a far cry from the Kubelík speed demon! However, Ančerl brings out the ceremonial majesty better than anyone else, with really blazing brass. You can actually hear all of the trumpets’ notes. Only the timpani are a bit too distant. Mackerras chooses slightly faster tempos than Ančerl and his performance loses something of the majesty of the other. However, it makes up for this in its excitement. While the trumpet parts do not come out as clearly and powerfully as for Ančerl, the brass as a whole is superb. And the important timpani really excel. In the end, it’s a toss up.

Thus, overall, both Ančerl and Mackerras have a great deal to recommend them, and I cannot see being without either. Of the other recordings I have not considered here, I have only a distant memory of Bernstein, Jílek (also with Beňačková) and Kempe, and haven’t heard Chailly, Dutoit, Košler, Masur, Neumann. In part two, I will consider the available recordings of the "original" version of the Mass as edited by Paul Wingfield. To my knowledge there are three, all excellent in their own way: Mackerras, Danish RSO (Chandos), Boulez, Chicago SO (, and Mackerras, Czech Phil on DVD (Supraphon). I have not heard the Svárovský (Ultraphon) recording reviewed here in November 2000, which according to Marc Bridle incorporated some of the Wingfield revisions (unspecified) [This recording is coupled with The Eternal Gospel - see review].

Part Two

It was with great anticipation and no little excitement that I read that the original version of the Glagolitic Mass had been recorded. The fact that Sir Charles Mackerras was the conductor was not surprising; what was unexpected was that he recorded the work not with Czech or even Vienna forces, but with the Danish National Radio Symphony for Chandos. As it turned out, this landmark recording was a revelation. It still stands as a safe recommendation for this version of the work, even if two live recordings have appeared since — both of which are even more exciting. The latter recordings are by Pierre Boulez and the Chicago Symphony ( and Mackerras again, on DVD with the Czech Philharmonic (Supraphon).


Mackerras, Danish National RSO (1994) Chandos (CHAN 9310) 40:29

Boulez, Chicago Sym (2000) (CSO CD05-2) 42:58

Mackerras, Czech Phil (2005)* Supraphon DVD (SU 7009-9) 43:03

*The actual performance took place on 21 March 1996

The timings given here are somewhat misleading because of longer pauses between movements in the live recordings, and the timing on the back of the Mackerras DVD is actually 45:27 which includes more than two minutes of applause.

All three of these recordings have a lot to offer, and I wouldn’t want to be without any of them. As I noted above, the Mackerras/Chandos recording, which was recorded in the concert hall but apparently not as a live recording, is a fine performance. It is well played and sung, even if the soloists are not as good as those in the other recordings. Moreover, it follows the original text to the letter, e.g., in the Svet movement the soloists sing "Blagoslovl’en gredyj v ime Gospodn’e," rather than the more usual "vo ime," of the revised version (which both Boulez and Mackerras/Supraphon follow). However, "safe" is not what one necessarily wants when it comes to the Glagolitic Mass. Boulez and Mackerras/Supraphon offer more exciting renditions of the Mass, even with a few missed notes (Boulez) and some suspect intonation (Mackerras) at times.

One does not associate Boulez with the music of Janáček and, indeed, the concerts from which this recording was made were the first time that he had ever conducted the work. The excitement of presenting a work he had not conducted before comes across in the recording, even if the interpretation is not as idiomatic as that of a Czech conductor or Mackerras. It is splendidly performed, with superb soloists and outstanding recorded sound. As one would expect, the Chicago brass blazes throughout. This might not be a first choice, but it is a necessary supplement.

Mackerras, of course, as one of the greatest Janáček interpreters, does not disappoint. One would think that after all the performances he has conducted of this work in both editions he might be bored by it. Not at all. The Supraphon DVD gives us what is arguably his greatest performance of the work, though it is rumored that he has also recorded it in Brno. That account, to my knowledge, has yet to be released. Listening to the soundtrack of this Czech Philharmonic account without the video is thrilling enough. The visuals, with excellent camerawork, only add to one’s pleasure. Where this recording especially outshines the others is in the glorious singing of the Prague Philharmonic Choir. For me, the highlight of the whole performance is the final choral entry in the Agneče Božij. The singing is incredibly soft, yet with firm underpinning by the basses, which takes the breath away.

Now to specific differences in the individual movements: (Mackerras/Chandos will be referred to Mackerras 1; Mackerras/Supraphon DVD as Mackerras 2)

Intrada: Mackerras 2 is lighter and faster than either Boulez or Mackerras 1, if not quite as imposing as either. All three are very well played.

Úvod: All versions are notably faster in the original scoring with its more complex meters than in the revised version. Boulez is somewhat smoother than either Mackerras recording, and Chicago’s low brass really projects here.

Gospodi pomiluj: The superiority of the Czech choir shows here with its diction especially clear. Again the Boulez brass is very imposing, and he adopts a slightly slower tempo. His oboe and that of Mackerras 2 are really beautiful. As for the soloists, both here (soprano only) and elsewhere, Boulez and Mackerras 2 are superior to those in Mackerras 1.

Slava: Mackerras sets perfect tempos in both versions, while Boulez is much slower and smoother. Yet his tempo still works, especially with the radiant singing of soprano Elzbieta Szmytka. Mackerras 2’s Eva Urbanová is also wonderful here; both are considerably better than Mackerras 1’s Tina Kiberg, whose singing can be overwrought at times. The same is true for the tenors, with Leo Marian Vodička (Mackerras 2) and Stuart Neill (Boulez) superior to Mackerras 1’s Peter Svensson, who tends to strain at higher levels. Orchestrally, Mackerras 2 excels with his very incisive timpani, whereas Boulez overwhelms with his stupendous brass and thundering organ. Some of the differences here are due to the closer, drier recording on Mackerras 2. The Boulez recording is fuller and more reverberant, but still clearer than Mackerras 1.

Věruju: All three begin this movement very well and the tenor soloists are more evenly matched here. The big differences come in the Raspet (Crucifixion) orchestral section. The flute solo is exquisite in both Boulez and Mackerras 2; not quite at that level in Mackerras 1 but still very good. Mackerras 2’s clarinet is reedier, typically Czech, than the others. However, in both Mackerras recordings the clarinet trio is played "offstage" as requested in the original score; the Boulez sounds upfront and as loud as his solo clarinet. The wild section with the three sets of timpani is presented variously in the three recordings. In Mackerras 1, one is not as aware of the overlaying string melody as in Boulez where the strings seem to predominate, and Boulez takes this section at a slower tempo than Mackerras in either version. In Mackerras 2, the strings definitely take a back seat to the timpani, which are very powerful, and there is a cymbal clash near the end that is not audible in the other recordings. All three impress, but perhaps Mackerras 1 achieves the best balance. Still both Mackerras 2 and Boulez in their own ways are mightily impressive here. As is true elsewhere, the organ makes a greater impact in the Boulez recording than in the Mackerras where the organs have a more distant presence. As to the bass soloist, first heard in this movement, Mackerras 1’s Ulrik Cold sings with a pronounced and rather unsteady vibrato compared to Peter Mikuláš (Mackerras 2) and Nathan Berg (Boulez), both of whom are superior.

Svet: Boulez takes this movement at a slower tempo than Mackerras in either version, but, as in the Raspet section of Věruju, it works. The solo violin, however, is best in Mackerras 2, much more of a presence than in the other recordings. The superiority of the soloists in both Mackerras 2 and Boulez once again makes itself known here. Peter Mikuláš is especially good in his solo. The "Plna sut" dance section is also taken at a faster clip in the Mackerras versions, but Boulez compensates with outstanding brass playing. One can really appreciate the tuba in his performance. As noted above, only Mackerras 1 follows the textual changes of the original version.

Agneče Božij: The choral entrance in all three versions is warm yet clear, beautifully sung, but Mackerras 2 is far superior to the others on the choir’s last, extremely quiet appearance. This is a world apart from the other performances and a highlight of the Mackerras 2 version. The soloists in their turn all do themselves proud; Ulrik Cold in Mackerras 1 is better here than he was elsewhere, if not as good as the other basses. The alto, who made a brief appearance in the Svet, is more of a presence here. Randi Stene in Mackerras 1, Nancy Maultsby (a mezzo) in Boulez, and Bernarda Fink in Mackerras 2 are all excellent in the solo and duet with the soprano. The horn, which accompanies the tenor in this movement, is especially noteworthy in Mackerras 2.

Organ Solo: While all the organ soloists impress, Boulez’s David Schrader seems at first to be the obvious choice due to the upfront recording of the organ. It is very present with a great deal of bass. However, on repeated listening, I have come to prefer the Mackerras versions’ more distant, yet full organ sound. My favorite at the moment is Per Salo on Mackerras 1 largely because I prefer the sound of the organ there to the reedier one as played by Jan Hora on Mackerras 2. However, none of these surpass that by the unnamed organist in Robert Shaw’s recording of the revised version.

Intrada: The second appearance of this movement is no mere copy of its appearance at the beginning of the Mass. Especially in the Mackerras versions, it is much more emphatic this time round. Mackerras 1 seems slightly slower and heavier than Mackerras 2, but still very good. Mackerras 2 makes a terrific impact, with its blazing trumpets and incisive timpani. In both performances he ends the work with a tremendous thwack, but in the later one he has a slight pause before the timpani’s final notes. Boulez is also excellent, with great trumpets and lower brass as usual, even if his timpani are neither as present nor his ending as emphatic as that of Mackerras.

To sum up, if I were to choose just one recording of the Glagolitic Mass to take to my desert island it would have to be the Mackerras DVD — even if I had only audio capability there! Having said that, however, I would really miss the 1963 Ančerl and 1984 Mackerras Czech Philharmonic recordings of the revised version. To these I would want to supplement at least the Boulez and for individual sections or soloists some of the others as noted above.

John L. Wright


Addendum to Glagolitic Mass Comparative Review

When I wrote my comparative review of recordings of Janáček’s Glagolitic Mass, I referred to the recording by Leos Svárovský with the Czech Philharmonic Chorus of Brno and the Czech Symphony Orchestra of Brno (Ultraphon UP 0011 2 231) by way of an earlier MusicWeb review by Marc Bridle. Because I had not heard that recording I could not use it in my comparison, but it was very favorably reviewed. Thanks to your founder, I have now heard that performance and must conclude that it is in the same class with Ančerl and Mackerras. According to Bridle, some of the revisions of the reconstructed original score by Paul Wingfield were included in this performance. After listening closely to the performance several times, I have not been able to detect any of these revisions. Nonetheless, this recording is clearly one of the very best of the published version, right up there with Ančerl and Mackerras. The singing and playing is marvelously idiomatic and the soloists are all superb. Overall, I prefer them to Mackerras’s and, with the possible exception of the tenor, to Ančerl’s. Ančerl’s Beno Blachut is indeed hard to beat. The sound, moreover, is for the most part the best of the three. The work is vividly recorded with a good perspective (not too close as Ančerl sometimes is, or as distant as Mackerras sounds--requiring some knob twiddling). The organ could have greater presence at times, but the important timpani are very powerful especially in the earlier movements. For example, in the Slava section they are really incisive; and later in the movement there are two instances of timpani crescendi that are not audible in other recordings. Did the conductor add them? If so, they are effective, if not necessary. I noticed in the later movements and especially in the final Intrada (which is nearly as majestic as Ančerl’s) that the timpani sound more muffled. I assume this is due to the recording and not the performance. This is a minor cavil in what may possibly be my current favorite recording of this version of the Mass. It in no way replaces those of the longer, Wingfield version, however. The overall timing for the Svárovský recording is 41:23, 1½ minutes longer than Ančerl or Mackerras, but not enough to make a notable difference.

John L.Wright

Leos Janacek Glagolitic Mass; The Eternal Gospel/Vecne Evangelium
E. Drizgova - Soprano, H. Stolfova-Bandova - Contra-Alto, V. Dolezal - Tenor, J. Sulzenko - Bass, M. Jakubicek - Organ, Czech Philharmonic Chorus of Brno and Czech Symphony Orchestra of Brno, conductor Leos Svarovsky

UP 0011 2231 FP DDD £10.50 postage paid




Gerard Hoffnung CDs

Advertising on

Donate and get a free CD


These prices cannot last!
Compare Toccata discs
Presto £13.50 +£2 pp
Amazon £15.53 post free
MusicWeb £10.50 post free

New Releases

Naxos Classical


Musicweb sells the following labels
Acte Préalable
(THE Polish label)
Altus 10% off
Atoll 10% off
CRD 10% off
Hallé 10% off
Lyrita 10% off
Nimbus 10% off
Nimbus Alliance
Prima voce 10% off
Red Priest 10% off
Retrospective 10% off
Saydisc 10% off
Sheva £2 off
Sheva Contemporary
Sterling 10% off
Toccata Classics

Follow us on Twitter

Subscribe to our free weekly review listing

Sample: See what you will get

Editorial Board
MusicWeb International
Founding Editor
Rob Barnett
Seen & Heard
Editor Emeritus
   Bill Kenny
Editor in Chief
MusicWeb Webmaster
   David Barker
MusicWeb Founder
   Len Mullenger

Return to Review Index

Untitled Document

Reviews from previous months
Join the mailing list and receive a hyperlinked weekly update on the discs reviewed. details
We welcome feedback on our reviews. Please use the Bulletin Board
Please paste in the first line of your comments the URL of the review to which you refer.