Toynbee’s Annex ‘note’ “Sicilian Light on Roman Origins”(“Study of History vol. VIII” pp. 704-707) is a good example of the logical process by which we come to knowledge of the past. As an Annex ‘note’ it is not a detailed scholarly essay. Rather, a parenthetical comment to the main body of the text to which it is ‘annexed’ (more like a long annotated footnote). Accordingly, it is cryptic and not intended to ‘prove’ pre se; rather ‘indicate’ an hypothesis about the origins of Rome and a direction of research for those interested.
However, while I find the ‘Rome origin’ argument fascinating, to my mind, the real value of the piece is the eloquent representation of the process Marc Bloch called “The Historian’s Craft”.
Because Toynbee is cryptic, the ‘note’ is challenging and not easily read. In two previous articles, I have explicated Toynbee’s “hypothesis” and supporting evidence that Rome “may have been founded by Elymian Sicilian colonist” (see “Sicilian Light...” and “Science, Irony...” in related articles box) . The purpose of this article is to gleen from this classical scholar and master historian the general epistemological nature of historian’s claims about knowledge of the past.
This article is not history per se. Rather, philosophy (i.e. epistemology) – How do we know what we think we know about the Past?
Observations in the Present lead to Logical Inferences about the Past
Remnants FROM the Past
The only objects that historians and other researchers of the past can ‘observe’ are remnants from the past: documents, potshards, arrowheads, tools, fossils, bones, glacier cores, tree rings, etc.
Logical Inferences ABOUT the Past
Based on observable remnants of existing objects FROM the Past...
We make logical inferences ABOUT the Past non-existing phenomena (societies, climate, geological events, etc.)
Classification of historic disciplines based on types of Remnants:
History – Document remnants (writing on pots, manuscripts, diaries, newspapers, letters, books, deeds, receipts, etc.)
Archeology – Material cultural remnants (potshards, arrowheads, graves, building foundations, jewelry, art works, tools, etc.)
Geology – Fossilized remnants (plants, animals, etc.)
Paleoclimatology – Nature remnants (ice sheets, tree rings, corals, shells, microfossils)
These and all other historic disciplines posit knowledge claims ABOUT the past logically inferred from observable remnant objects in the present preserved FROM the past.
Thus, for example, if archeologists discover (observe) a structure constructed with precisely carved very hard stone, based on the observation of the stone’s characteristics (hardness and carvings), they infer that the people who constructed the building must have had metal tools (copper or iron) to carve the stone.
They have no direct observable evidence of metal tools. They have not found (observed) metal tools at the location of the structure; however, there is no other logical (inferable) explanation for how very hard stone can be carved with precision other than with metal tools.
Preponderance of Evidence
Further, when considering knowledge of the past, one must emphasize that such knowledge claims are not ‘indubitable’; they lack the certainty of knowledge claims made by natural scientist (chemist, physicist, biologist, etc). Because natural scientists study existing observable phenomena, they can ‘verify’ their knowledge claims with repeatable experiments (controlled observations) and gather more evidence as needed to increase the certainty of their knowledge claims.
Researchers into non-existent past phenomena obviously have no opportunity for controlled experimental observations, and are limited to serendipitous (fortunate discoveries by accident) evidence. Historians et al take what objects fortune/luck has preserved and been discovered.
Accordingly, the inferences they make about the past are based on (in legal parlance) the “preponderance of the evidence” and “beyond a reasonable doubt”; or (in mathematical parlance) the “balance of probabilities”.
The historian gathers all the possible evidence (remnants from the past) available and then must make judgments (about the past) with the highest probability of truth (reasonable doubt) based on the preponderance of the evidence.
Preponderance implies that all the evidence may NOT support the historian’s inferred conclusions. Indeed, some evidence may contradict the conclusion. The historian must base his inferences on the greatest amount of evidence in support of his conclusions, while recognizing the possibility that the conclusion may be erroneous, and new evidence found in the future may result in a rejection of the inferred conclusion.
For example, consider Toynbee’s phrasing:
1.) “...the possibility that the Romans may have been descended from [Sicilian] immigrants...”
Note: the tentativeness associated with words like ‘possibility’ and ‘may’
2.) “As it appears to the writer of this Study (i.e. Toynbee) the evidence proves conclusively that the Sicels’ mother tongue was virtually identical with Latin”.
Note: he claims on the one hand “evidence proves conclusively;” yet qualifies his ‘conclusiveness’ with “appears” and “virtual”.
He is expressing his judgment about the evidence’s preponderance and probability; but recognizes that other scholars may disagree. Indeed, he provides a footnoted reference to those other scholars who do not agree with him.
In short, ultimately, there is a degree of subjectivity in the concepts of preponderance and probability and Toynbee is demonstrating that while he entertains a degree of confidents of his conclusions; his conclusion is not based on the concurrence of all the evidence; rather, the preponderance of the evidence.
3.) “The relics [remnants] of the Ligurian language have been presented and discussed by (historians) Conway and Whatmough. Whatmough’s judgment is that Ligurian is not a member of the Italic family of languages. The present writer (i.e. Toynbee) ventures the opinion that Ligurian, as well as Sicel, will turn out to be an Italic language of the Latin sub-group”
Note: the phrases “ventures opinion” and “will turn out to be”; he has an opinion about the Ligurian and Sicel language, but recognizes that there is currently not a ‘preponderance of evidence’ to support the ‘probability’ of the truth of his opinion. Yet, he thinks that in the future more research (e.g. archeological discoveries) will bring forth supporting evidence.
4.) “At the eastern end of the Italian Riviera, we find four place-names...that are identical with [those] in Elymian Sicily...This can hardly be accidental; we cannot reject the inference...and the probability that the names in Liguria were derived from Sicily...”
“The hypothesis that most naturally suggest itself is that these Elymian settlements along the Eastern Riviera were part of the older colonizing movement...”
“On the analogy of the Elymian settlements along the Riviera, is it too rash a conjecture to guess that the cluster of Latin-speaking communities [on the Tiber]...”
Note the terms he uses: inference, probability, hypothesis, analogy and conjecture. There is not enough evidence to ‘conclusively’ demonstrate that the Elymians were the founders of Rome, but there is enough evidence that the hypothesis cannot be rejected and should be entertained.
Hypothesis – says it all!
Toynbee’s use of the word “hypothesis” is the key to his thinking not only on the subject of Rome’s origin; but more generally and more importantly, the key to understanding the logic of historiography.
‘Hypothesis’ is a proposed explanation of observable verifiable remnant facts. Again, the historian begins the quest of knowledge about the past by first making observations of remnants from the past. After the observations are made, then a tentative (probable) explanation is posited (i.e. a hypothesis) based on the preponderance of the observable evidence.
For example, in the case of Liguria, Toynbee cites the remnant evidence:
- Latin was spoken in both Liguria and Sicily
- Place-names were identical in both Liguria and Sicily
- Grecianized word for ‘mountain’ was used in both Liguria and Sicily
How does one explain those remnant’ facts’?
He posits the hypothesis:
Elymain Sicilians (i.e. Sicilians from the area were the place-names occur), whose Latin language was influenced by the Greeks in Sicily, emigrated to and establish a colony in Liguria.
He has not, nor does he claim to have proved that Elymains established a colony in Liguria. He has posited a ‘tentative’ ‘hypothetical’ explanation of the observable facts.
Further research with a mind to gathering more ‘observable remnant facts’ is necessary before a judgment can be made about the ‘preponderance’ of the evidence and the probable truth beyond a reasonable doubt of the Past.
For example, corroborating evidence supporting the hypothesis:
Edward A. Freeman, the renowned classical scholar, in his seminal work History of Sicily From The Earliest Times Vol I (1891), posits that the Elymians had a history of seafaring colonization. In fact that is how they happen to be in Sicily. He writes:
"The Elymians were likely to have been colonist in the strictest sense, in the same sense as the Phoenicians and the Greeks. That is they were strangers from some other land, who found a corner which the Sikans had failed to occupy or from which they could be driven out. ( p. 198)
This would be considered corroborating evidence supporting Toynbee’s hypothesis that the Elymians were a seafaring colonizing people.
However, there is also evidence NOT supporting the hypothesis:
Toynbee says that both the Ligurians and the Sicilians were Latin speaking people. However, while there is general agreement among scholars that the Sicel Sicilians spoke Latin, there is no evidence that the Elymian Sicilians spoke Latin. Freeman writes;
“Of the language of the Elymians we have no certain remains...” (i.e. there is no certainty that they spoke Latin p.198)
In short, the historian is obliged to:
- Gather all the evidence available both supporting and not supporting the hypothesis about the events of the Past;
- Make a judgment about the preponderance of the evidence either supporting or not supporting beyond a reasonable doubt the probability of the truth of the hypothesis.
Toynbee’s essay concisely illustrates that process, and more generally is an excellent example of the historiographic method that came to be known as the Critical Method so eloquently explicated by the great historian Marc Bloch in has book “The Historian’s Craft” (see “Southern Question...” article on Bloch in related articles box).