from the few who do not commit themselves, and those who trace Heine's
poem direct to Brentano, and Oscar F. Walzel to be referred to later,
all commentators, so far as I have looked into the matter, say that he
did. Adolf Strodtmann said[44] it first (1868), in the following
words: "Es leidet wohl keinen Zweifel, dass Heine dies Loeben'sche
Ballade gekannt und bei Abfassung seiner Lorelei-Ballade benutzt hat."
But he produces no proof except similarity of form and content. Of the
others who have followed his lead, ten, for particular reasons, should
be authorities: Franz Muncker,[45] Karl Hessel,[46] Karl Goedeke,[47]
Wilhelm Scherer,[48] Georg Mücke,[49] Wilhelm Hertz,[50] Ernst
Elster,[51] Georg Brandes,[52] Heinrich Spiess,[53] and Herrn. Anders
Krüger.[54] But no one of them offers any proof except Strodtmann's
statement to this effect.

Now their contention may be substantially correct; but their method of
contending is scientifically wrong. To accept, where verification is
necessary, the unverified statement of any man is wrong. And, that is
the case here. Elster's note is of peculiar interest. He says: "Heine
schloss sich am nächsten an die Bearbeitung eines Stoffs an, die ein
Graf Löben 1821 veröffentlichte." The expression "ein Graf Löben" is
grammatical evidence, though not proof, of one of two things: that
Loeben was to Elster himself in 1890 a mere name, or that Elster knew
Loeben would be this to the readers of his edition of Heine's works.
Brandes says: "Die Nachahmung ist unzweifelhaft."[55] His proof is
Strodtmann's statement, and similarity of content and form, with
special reference to the two rhymes "sitzet-blitzet" that occur in
both. But this was a very common rhyme with both Heine and Loeben in
other poems. How much importance can be attached then to similarity of
content and form?