H.—ANTHROPOLOGY 147 
The only writer on tradition who has touched on this point, so far as 
I can learn, is Prof. Chadwick, who says: ‘ 'The existence of a poem 
‘or story which deals with reminiscences of tribal conflicts necessarily 
* presupposes an absorbing interest in tribal history.’ He goes on toshow 
that this interest could only be due to patriotism, but fails to realise that 
patriots are notoriously indifferent to facts; any fable which gratifies 
their national pride is history to them. ‘The conclusion seems to be that 
since illiterate persons are never interested in history, historical facts can 
never be transmitted by illiterate persons. 
In case, however, we may be thought to have gone too fast, let us ask 
another question: should illiterate persons wish to transmit historical 
facts from age to age, would they be able todo so? Let us first be clear 
as to what we mean by ‘ from age to age.’ We do not speak of our 
parents’ reminiscences as tradition, or apply the term to anything that 
happened within the memory of living men. We apply it only to events 
which may be supposed to have happened in the more or less remote past. 
We must also note that when a tradition is written down it ceases to be a 
tradition, and becomes merely the account of a tradition, unless we can 
be sure that those who repeat it have not been influenced by the written 
record. We may say then that a traditional narrative is one which has 
been handed down for at least a hundred years by people who have 
derived it from purely oral sources. 
Let us now consider what are the sources of history. Apart from 
archeological evidence, which, however valuable, is seldom a guide to 
actual incidents, we may divide them into four classes : 
(z) Accounts written at the time by persons who were present at the 
events which they describe—letters, despatches, memoranda, 
diaries. 
(2) Accounts written by persons who were present, but not till some 
time afterwards—autobiographies, reminiscences, inscriptions. 
(3) Accounts written by people who obtained their information from 
actors or spectators shortly after the event—annals, chronicles, 
proceedings of trials, newsletters, press reports, diplomatic corre~ 
spondence. These would not be accepted as evidence in a court of 
law, but are often very properly accepted by the court of history. 
(4) Accounts obtained by questioning people as to what happened a 
long time before, or accounts obtained at second or third hand. 
These are often recorded as survivors’ tales, conversations, memories, 
gleanings. 
Now it should be clear that the first three are, in varying degree, the 
only genuine sources of history. The fourth may be useful for reconciling 
discrepancies or filling in details, but would not be accepted as a satisfac- 
tory authority for a fact otherwise unknown. I know an old gentleman 
living not far from Leicester who has personal reminiscences of the 
French Revolution of 1848, but the fact would hardly be accepted on his 
sole authority. Second-hand evidence is not admitted in a court of law 
because it is notoriously unreliable. It is admitted by historians, but 
1H. M. Chadwick, The Heroic Age, p. 273. 
