Three Sources on Adwa: Judging Reliability
A Grade 11 student in Addis Ababa is writing a short paper on the Battle of Adwa (1 March 1896) and has three sources. Source A is a chronicle written in the 1920s by an Ethiopian church scholar, in Ge'ez and Amharic, which describes the victory as divinely granted, says 'the enemy's dead were beyond counting', and names the commanders and the order of battle in detail. Source B is a report printed in an Italian newspaper five days after the battle, quoting a correspondent travelling with the retreating army; it gives 4 000 Italian casualties and describes an orderly withdrawal. Source C is a chapter by a modern historian who cites Italian military archives and estimates about 7 000 Italian dead and 1 500 prisoners, and notes that Eritrean askari casualties are usually omitted from both older accounts. The student's draft reads: 'Source B is worthless because Italy lost the battle, and Source A proves that the Ethiopian account is correct.'
1. Which source is most useful for answering 'how many Italian soldiers died at Adwa', and which is least? Justify using date, author, purpose and audience.
Read the worked answer
Most useful: Source C, but as an estimate rather than a fact. It is the only source that names its evidence (military archives), was written after the passions of 1896 had cooled, and its author has no reason to inflate or deflate either side's losses. It also flags a known gap - Eritrean askari casualties - which suggests a careful rather than a triumphalist count.
Source B is chronologically closest (five days) and has a named position (a correspondent with the retreating army), which makes it valuable for sequence and detail. But it was produced under wartime censorship, for a newspaper audience that expected a story of heroic resistance rather than catastrophe, and casualty figures were being compiled while the army was still withdrawing. The 4 000 figure should be treated as a claim, not a measurement.
Least useful for numbers: Source A. It was written about thirty years later, by an author whose purpose was to celebrate a divinely granted victory, and 'beyond counting' is a rhetorical formula, not a count. It is the weakest possible evidence for a numerical question.
2. The student concludes that Source A is unreliable. Explain why it may still be valuable, and state precisely what questions it can and cannot answer.
Read the worked answer
A source can be unreliable for one question and excellent for another; 'unreliable' is not a property of a source in general.
Source A cannot answer: how many died, how many troops were engaged, how long the battle lasted in hours, or what the Italian plan was.
Source A can answer: how the victory was understood and explained by educated Ethiopians in the following generation; the religious framing of the campaign (fasting, prayers, the role of the church); the names and relative standing of the commanders, which the chronicler had a reason to record accurately because his readers knew these families; and the order of battle as it was remembered and taught.
For a paper on military outcomes, Source A is background. For a paper on memory, identity and how Adwa was made into a national symbol, it is a primary source of the first importance. The student should match each source to the question it can bear, rather than assigning a single verdict of 'reliable' or 'worthless'.
3. Critique the two sentences of the draft: 'Source B is worthless because Italy lost the battle' and 'Source A proves that the Ethiopian account is correct'.
Read the worked answer
'Worthless because Italy lost' confuses bias with uselessness. Every source is written from somewhere. A report from the losing side is evidence of what that side believed, what it was allowed to print, and how it explained defeat - all useful for a paper on the war's reception. Bias reduces the value of a source for facts it is motivated to distort (casualty numbers), but leaves it valuable for facts it has no reason to distort (dates, terrain, the direction of the retreat, the weather, the presence of correspondents). The correct move is to use the source where it is not under pressure to lie.
'Proves' is too strong in two ways. First, Source A and Source C do not independently confirm each other; one is a later chronicle and the other a modern reconstruction partly built on European records. Second, 'the Ethiopian account' is not a single thing - there were many Ethiopian accounts, and Adwa was a victory for a coalition of regional armies whose perspectives differed.
The student should also ask what would count as evidence either way: mustering rolls, prisoner lists, Red Cross records, and Italian parliamentary inquiries. Until one of these is checked, the honest wording is 'suggests' or 'is consistent with', not 'proves'.