Levine and Mackerras round off their respective series with collections of early symphonies, and the first comment to make is that in none of these performances of generally trivial boyhood inspirations do I feel any sense of routine. None the less, as in previous issues from these rival performers I consistently prefer Mackerras to Levine for his extra variety of expression, whether in rhythm, texture or dynamic, with a keener feeling for Mozart style, not least in the slow movements and minuets, which he regularly takes faster than Levine. Levine by comparison, with his great Viennese orchestra, is always lively in Allegros, but seems too smooth, even glib at times, with rhythms sprung less infectiously.
There is a discrepancy too between these offerings in what is covered, and I commend the regularized lists above as being far easier for identification than the labelling on the discs. Levine's set uses the Kochel numbers from the original edition, and puts the revised numbers in brackets, which is customary, but then omits all mention of the regular numbering. The usual numbers for the early symphonies may be misleading over chronology, but at least they have the advantage of being easier to identify. That is an irritating omission, and Telarc in the Mackerras series do have the numbers very clearly displayed, but then confusingly they list the revised K numbers first, and the usual ones only in brackets. It does not help that the second of the three discs contains no fewer than five of the symphonies latterly lumped together as K73. How much clearer to use the conventional K numbers. Admittedly, the monumental AAM/L'Oiseau-Lyre set using period instruments gives priority to the revised K numbers in the same way, but it has been a great relief to go to the Marriner set, which as ever in the Philips Complete Mozart Edition, is a model of detailed clarity in identifying of items, with even that information readily supplemented by the valuable ''Compactotheque'' volume issued along with the Edition.
So far as I can work out, the symphonies not in the Levine set have been omitted on the grounds of not being fully authenticated, as for example K81, ''No. 44'', which is quite probably by Leopold Mozart. On a practical level, the Mackerras discs are better-filled than Levine's, with 18 symphonies on three CDs, as against 20 symphonies on five discs, not available separately. Both these issues are at full price, though almost at once the whole Mackerras series is being collected into a ten-CD set at a special price, working out a degree more expensive than mid-price. Against the obvious advantage of the Mackerras series there, Levine is more generous over repeats, though even after the AAM set I do not warm to repeats being observed in the da capos in Minuets, as Levine does too. With the Minuets in particular I much prefer Mackerras's brisk manner, and his flowing Andantes are consistently more stylish. An outstanding instance is the G minor Andante of K22 in B flat, No. 5, where Levine sounds heavyhanded in bringing out the foretastes of romanticism, and Mackerras, faster and lighter, underlines the harmonic surprises far more clearly, as well as elegantly.
Consistently, it is Mackerras who finds light and shade in these early works, though some may feel that, with warm reverberation characteristic of this Prague orchestra's recording venue, the scale is too large. Marriner, recorded in the 1970s, takes a more traditionalist view, with speeds for Andantes and Minuets slow like Levine's, but in clarity of texture and immediacy the old analogue recording is arguably the best of the three, even if violins sound edgy at times. Broadly, each of the Mackerras issues can be warmly recommended, particularly when they provide a useful option for those who simply want to sample early Mozart, and not to have the whole series. The first disc containing seven symphonies from No. 1 onwards makes an excellent sampler. The DG recording for Levine is a degree more intimate, but the textures are more opaque, even though, as with Mackerras, the harpsichord continuo is well-balanced.
The Spivakov disc, despite brisk and refreshing performances of an attractive group of early symphonies, is a serious disappointment, entirely because of the washy recording. Though the phenomenal articulation of quavers by the first violins in the finale of No. 28 is clear enough, the rest of the orchestra is absurdly distanced, with trumpets and horns a blur, and the oboe sounding as though it is off-stage. I have rarely known reverberation of this degree so damaging to textures, a question of microphone placing I suspect. I am surprised to find that the venue was not a church but a school hall. I hope these talented musicians will receive more helpful treatment in their future recordings.'