Some of them have explained the law of labour-value as an _historical_ law, peculiar to capitalist society, all of whose manifestations it determines;[14] others rightly seeing that the manifestations of capitalist society are by no means determined by such a law, but comply with the general economic motives characteristic of the economic nature of man, have rejected the law as an absurdity at which Marx arrived by pressing to its extreme consequences an unfortunate concept of Ricardo. Criticism was thus bewildered between entire acceptance, combined with a clearly erroneous interpretation, and entire and summary rejection of Marx's treatment; until, in recent years, and especially after the appearance of the third and posthumous volume of _Das Kapital_, it began to seek out and follow a better path. In truth, despite its eager defenders, the Marxian doctrine has always remained obscure; and, despite contemptuous and summary condemnation, it has always displayed also an obstinate vitality not usually possessed by nonsense and sophistry. For this reason it is to the credit of Professor Werner Sombart, of Breslau University, that he has declared, in one of his lucid writings, that Marx's practical conclusions may be refuted from a political standpoint, but that, scientifically, it is above all important to _understand_ his ideas.[15] Sombart, then, breaking openly with the interpretation of Marx's law of value as a _real_ law of economic phenomena, and giving a fuller, and I may say, a bolder expression to the timid opinions already stated by another (C. Schmidt), says, that _Marx's law of value_ is not _an empirical but a conceptual fact_ (Keine empirische, sondern eine gedankliche Thatsache); that Marx's value is a _logical fact_ (eine logische Thatsache), which aids our thought in understanding the actual realities of economic life.[16] This interpretation, in its general sense, was accepted by Engels, in an article written some months before his death and published posthumously. To Engels it appeared that 'it could not be condemned as inaccurate, but that, nevertheless, it was too vague and might be expressed with greater precision.'[17] The acute and courteous remarks on the theory of value, published lately in an article in the _Journal des Economistes_ by an able French Marxian, Sorel, indicate a movement in the same direction. In these remarks he acknowledges that there is no way of passing from Marx's theory to actual phenomena of economic life, and that, although it may offer elucidation, in a somewhat limited sense, it does not appear further that it could ever _explain_, in the scientific meaning of the word.[18]