Two-and-five-sixths of this Mahler trove - which is to say the Abbado/Berlin recordings - are treasures in the best sense, worth cashing in a couple of 401(k)s for the insights in playing that surpass anything I've experienced in the disc history of the Berlin Philharmonic. The precision, balances, sonority, and comradeship within choirs makes some of Karajan's legacy sound jack- booted and glutinous.

That said, let's clear a piece of comparative junk out of the way - the Westminster reissue of a slapdash, slipshod Fifth Symphony that Hermann Scherchen recorded in 1952 with the "Vienna State Opera Orchestra." Vienna was still recovering from WW2 (and would for another 20 years). The Staatsoper an der Ring still lay in ruins, meaning rank and file players in a pool of 140 did a lot of jobbing under various names for smaller companies (Westminster/Nixa, Epic, Vox, et al.), often with little or no rehearsal. The self-governing Philharmonic skimmed the cream, though it was not very busy in the recording studio from 1947 until the early and later '50s. With mostly left-over players Westminster filled some holes in the catalog, but its product lacked consistency, and predictability in the case of Scherchen, an erratic autodidact. His American debut didn't come until 1964, two years before his death at 74, when he led the Philadelphia Orchestra in a mangled Mahler Fifth one can hear in the orchestra's first self-produced volume of broadcast material.

Here at least he didn't move sections of the score from one movement to another. But this Westminster is chiefly a historical document: the first complete commercial recording of No. 5. The opening solo trumpet calls sound as if the player hadn't eaten for several days, although horns and trombones manage to play robustly later on. Wind intonation is haphazard, the strings lack weight (probably understaffed, given Westminster's budgets), and the reading - well, you remember the car chase in Bullitt? It's not a matter of timings which are conventional if a shade quick, but what happens inside them. There are Mahler Fifths by the carload - the one I've kept is Abbado's thrilling 1993 live performance with the Berlin Philharmonic on DGG; a couple of minutes slower overall than Scherchen's but everywhere to Mahler's advantage. Interestingly, both take 9 minutes for the Adagietto, with Abbado 14 seconds quicker, but the difference in playing and sound makes this Westminster a relic, at its worst a parody.

Of Abbado's most recent Mahler CDs on DGG, all of them live performances, two are remakes - of No. 3 (which he did with the Vienna Phil in the '80s - Schwann/Opus omitted the year before it ceased publication ), and No. 7 (with the Chicago Symphony in 1984, until now my benchmark recording of that piece). With the Berlin, Abbado takes both "Night Music" movements in No. 7 quicker (15:54 vs. 16:37, and 12:58 vs. 14:01), but the first, third, and final movements are virtually unchanged. Within, however, nuances are innumerably expressive and enlightening without sounding mannered in the slightest, which the Berliners play with a subtlety one doesn't hear from the Chicago Symphony of Solti's era. Recorded in May 2001, the year Abbado suffered radical surgery and yet survived, it is the most recent of his three Berlin releases, and sonically all one could ask from two channel stereo. More than any other conductor -and I've heard plenty given that No. 7 is my favorite Mahler Symphony - Abbado finds and sustains a structure in the opening movement that escaped even a colleague as distinguished as Michael Gielen. There are plenty of Sevenths, good ones too (although I don't share R.E.B.'s recent enthusiasm for a 1977 Kondrashin version from Amsterdam on Tahra). But you can jettison all but a couple if space is a problem. Abbado's 2001 Seventh is an abiding source of wonder, and for every Mahler fan an essential purchase.

I didn't much care for the Viennese No. 3 with Jessye Norman as contralto soloist. It seemed almost fractured in the mighty opening movement, fussy in the middle movements, and quite matter of fact in the Adagio finale (not necessarily a drawback) until Abbado prolonged the final chord beyond reason or proportion. In 1999 with the Berlin Phil, a live tour performance in London's Royal Festival Hall, he had tightened and tautened the first movement - it now packs a terrific wallop, although less than the cumulative impact of Michael Gielen's on H”nssler (with a fine orchestra but no Berlin Philharmonic). The second and third movements are still arch, and Abbado continues to prefer a very slow tempo thereafter for the setting of Nietzsche's "O Mensch! Gib act!" But his most excellent singer, Anna Larsson, has no trouble sustaining either his virtual stasis or her intonation. To advantage, he darkens the mood at the end of the fifth movement as Gielen has always done, which makes the finale ("What love tells me") that much more consolatory. Here it lasts 22 minutes vs. Gielen's 24:27 ‚ but dammit, when everything has gone so well (despite the orchestra tiring audibly), Abbado again prolongs the final note disproportionately. This has the same disruptive effect as Daniel Barenboim's sudden, off-to- the-hounds Presto in the last measures of Saint-SaÎns' Third Symphony with the Chicago Symphony, spoiling what was otherwise one of the two best recordings ever made of that work.

Still, I'm keeping the new Abbado (oddly, one Britcrick complained of artificial reverb, writing that the sound lacked punch in the climaxes; but not on my rig). However, it is Gielen's I'll play when I want to commune with Mahler. Abbado offers no filler: the first movement (33:23) is featured by itself on disc 1; the remaining five movements (including 3:20 of rapture by the London audience) fill the second one.

Now for the Ninth, which I don't believe Abbado previously recorded, and which already has two digital performances on DGG by Karajan and the Berlin Phil - made just two years apart in the halcyon 1980s of fat budgets and brisk sales. This new one is superb without equivocation. After the L”ndler scherzo, Abbado's Rondo-Burleske is terrifying in its accumulation of bitterness and anger - Mahler decrying God in effect - after which the long Adagio finale is both a reconciliation and a benediction. Giulini (with Chicago) is the only other Italian conductor I know of who has recorded the Ninth, and a troubled reading it remains, albeit hardly idiomatic. Abbado has made the work his own - it could be a different orchestra than Karajan's that plays so beautifully and immaculately now. He has solved how it goes without stinting on detail or sacrificing structure, and amazingly DG has fit it on a single disc, including almost 2 minutes of applause: a total timing of 81:03! I don't know another performance in this league, and urge that you buy No. 9 as well as No. 7 if Mahler's time has come for you. The recorded sound is so transparently natural one isn't aware of it as such until later.

Which leaves the Naxos Tenth with a pedigree otherwise unmatched on discs: the completion by Joe Wheeler (1927-77), a British composer, undertaken in 1952-53 - unaware that his compatriot Deryck Cooke (with the help of Berthold Goldschmidt) or Chicagoan Clinton Carpenter were using the materials then extant that Alma Mahler allowed to be published in 1923. Both Wheeler and Carpenter (the first one finished in 1947) lacked the 50 pages of additional music found subsequently, chiefly for the second movement and published as part of a follow-up "facsimile manuscript" in 1967. Cooke's first version of 1960 was denounced by the widow until she heard a broadcast of it shortly before her death in 1964. Before Cooke's own death he completed a second performing version, but it was his first one that debuted on discs: a nobly impassioned performance by Ormandy and the Philadelphia Orchestra on CBS/Sony, harshly recorded, especially on top, which sounds best on my mid-fi computer! The Carpenter version was produced in New York City with the help of Remo Mazzetti Jr., who was moved to make his own version in 1983; Leonard Slatkin premiered and recorded it in Saint Louis for RCA Red Seal. The Wheeler went unheard until a "Colorado Mahlerfest" in 1997, again with Mazzetti's assistance. When the latter realized that Wheeler's version, in the words of Jerry Bruck, "came closer to what [Mazzetti] now felt was Mahler's last style [i.e. leaner] than even his own first version," he revised his score substantially. It is Mazzetti's revised version that Jes™s LÛpez-Cobos recorded two years ago in Cincinnati for Telarc.

Wheeler made four revisions of his original as new materials came to light. The last one in 1966 is what the Polish National Radio Symphony recorded for Naxos in 2000 at Katowice. Robert Olson, music director of the summertime Colorado homage and chief conductor of the Kansas State Ballet the rest of the year, is listed in Naxos' credits as "editor" of the Wheeler edition. It turns out to be both beautiful and poignant in the vein of Cooke I - in other words less prolix than subsequent versions which used Mahler's Seventh and Eighth Symphonies as their model. Olson, sorry to say, doesn't do justice to the great outbursts - those sudden dissonant chords in the opening Adagio that return in the finale, or the muffled drum-beats in the fourth movement and finale. They sound perfunctory, hurried; so does the theme that follows. And the orchestra somewhat lacks in numbers what the writing calls for; Mahler did complete both the first and third (Purgatorio) movements, which the widow - by then Alma Mahler Gropius, nÈe Schindler, with Werfel yet to be added - asked her son-in-law Ernst Krenek to edit for performance in 1924. Interestingly, despite her later tantrums, she had first asked Schoenberg to consider completing the Tenth, but by then he was a light-year distant from Mahler; in the 1940s, Schoenberg turned down another offer by Jack Diether of the Bruckner-Mahler Society. So did Shostakovich.

In Olson's favor, he conducts gentler portions of the music with genuine tenderness - the ending is almost as transfigured as Simon Rattle's with the Berlin Phil on EMI (released to celebrate his nomination as Abbado's successor there, although a contract wasn't signed until this past spring). Polish strings may not have the patina of Berlin's, but they've recorded Mahler extensively for Naxos under Antoni Wit, meaning the style is in their bones. Olson has summoned a weight and poignance that move one close to tears at the end of this farewell to life. I prefer Wheeler's version to Mazzetti's, and urge Mahlerians to supplement whichever versions of No. 10 they already have with this budget-priced pearl. As usual in products from Katowice, Beata Jankowska has tripled as producer, engineer and editor: the result here is one of her best recordings to date.

R.D. (July 2002)

ADDENDUM
Since the above appeared on line, two readers have corrected my paragraph about the Mahler Ninth. Longtime friend and colleague John von Rhein, music critic of the Chicago Tribune since 1977, was the first to point out that Abbado recorded it with the Vienna Philharmonic for DGG in the 1980s. Lawrence Kasimow not only remembered the 1980s version (which, truth to tell, I never even knew had been recorded, much less issued) but one also on DGG by Giuseppe Sinopoli with the Philharmonia Orchestra (plus Symphonies 1, 2 and 5). Mr. Kasimow furthermore advised that Arkadia once published a broadcast performance by the late Bruno Maderna (date and orchestra unspecified) - a superb interpreter of the Second Viennese School, whose paterfamilias Arnold Schoenberg and pupil Anton von Webern were unabashed admirers of Mahler (and in the latter case a celebrated conductor).

Apologies for the omissions, but by the time they appeared I was no longer writing record reviews, nor keeping up with new releases in a time of glut and duplications. It's called burn-out, and the hiatus proved in time to be healing. But I gave away most of the catalogs and historical volumes by R.D. Darrell and David Hall (I), which left a gap in my resources.

R.D.  (August 2002)