In a letter to his father of April 1781 Mozart wrote that one of his symphonies (almost certainly either the Paris or No. 34) went ''magifique'' at a Viennese concert with 40 violins, ten violas, eight cellos, ten double-basses, six bassoons and the rest of the woodwind doubled. Mozart evidently relished large orchestras, though he didn't always get them—the Prague orchestra that gave what was probably the first performance of No. 38 numbered only a handful of players. But as in his recent recording of three of Haydn's ''Paris'' symphonies (12/92), Ton Koopman opts to play these sonorously scored symphonies, all in C and D major, with a string band of no more than 14 (5, 4, 2, 2, 1). The inevitable lack of string weight and power is most noticeable in the Prague and the Jupiter, in every sense the biggest works here. Whatever the size of the orchestra at the first performance, the violin writing in the first movement of the Prague simply demands more body of tone than the Amsterdam players, for all their finesse and agility, can muster—above all at that magnificent, searing final climax (12'04''). Both outer movements of the Jupiter suffer similarly in this respect and in the slow movement important violin detail is often submerged beneath Mozart's sustained, high-lying wind writing (notably in the agitated C minor theme from 1'25'').
The same problem intermittently affects the earlier symphonies, especially the Paris, with its delight in grand, brilliant orchestral effects, in sheer richness of sonority Nor is there the compensatory clarity of texture you might expect with a tiny string band: the recordings, made at live concerts in Tokyo, are over-reverberant, sometimes slightly unfocused. The tuttis can be opaque (I noticed this particularly in No 35) and too much contrapuntal detail is blurred in the first movement of the Prague and the finale of the Jupiter. In fact, comparing Koopman in the Jupiter finale with Bernstein and the VPO (DG, 1/87) and Walter and the Columbia Symphony Orchestra (Sony, 9/91) revealed how much more lucidly etched were the contrapuntal textures in the large-scale modern-instrument performances, thanks partly to the recorded sound partly to the older conductors' unerring ear for balance.
For all these provisos there are many good things in these readings In actual execution the Amsterdam orchestra is as fine as any period group today. If string melodies are sometimes afflicted by the baroque bulge, articulation is wonderfully delicate and precise. The distinctive soft-grained timbre of the woodwind (an especially characterful first oboe) often makes for ravishing sonorities, as in the Minuet of the Jupiter. And the valveless horns and trumpets can bray thrillingly at climaxes—at the end of the first movement of No. 34, for instance, or in the codas of No. 31's outer movements. Rhythms in the quicker movements have a fine stride, except, perhaps, in the Jupiter first movement, which can sound a shade harassed The outer movements of both the Paris and the Linz, especially, are done with immense verve and brio. Minuets truly dance, even though I wouldn't always want to hear that in the Haffner taken as smartly as it is here. The Andante of the Jupiter apart, slow movements are decidedly con moto, after the fashion of period performances: I like the easy, two-in-a-bar sway of the 6/8 Andante of the Paris, though in the 6/8 slow movements of both the Linz and the Prague I thought the tempos just too swift for Mozart's elaborate textures and chromatic harmonies to make their full effect.
Needless to say, the performances contain a liberal smear of unmarked trills and turns, as much a Koopman hallmark as violent gear changes between minuets and trios are with Harnoncourt. And you may or may not care for the added timpani rolls in the first movement of the Linz (from 1'44''). I don't. You may also regret that Koopman fails to make second-half repeats in those movements where Mozart requests them, most crucially in the finale of the Jupiter; but the characteristically generous Erato playing time is fair compensation. The Japanese audience is less bronchial than many, though we are treated to around 20 seconds of rapturous applause after each symphony.
These are always intriguing readings, sometimes with a flair and spontaneity that are hard to recreate in the studio; and if you're undeterred by the exiguous string band and the resonant recording the two discs will make an attractive and economical proposition. My own period-instrument recommendations, though, would be the polemical Norrington (whose late Mozart series on EMI has so far embraced No. 38, coupled with No. 40, 11/92, and No. 41, coupled with No. 39, 6/91) and the more traditional Gardiner, whose Philips series now includes all the symphonies from No. 35 onwards (9189, 2/91 and 11/92).'