Historical Methodology: A Few Thoughts
Posted: Mon Jun 13, 2022 6:19 am
Since so much of what is discussed on Discuss Mormonism is about history, I thought that a few thoughts I posted in the Celestial Kingdom in relation to the Kinderhook plates might have wider interest and relevance...
A few thoughts on historical method...
First, historical explanation is a process of inference to the best explanation. We seek to piece together the fragments of the past in the way that offers the best overall explanation. The more data an explanatory model can account for, the better for that model.
Second, what we mean by explaining the data here is that we can explain the various historical sources and why the events came out the way they did--why they came out that way in particular and not any of the other countless ways they could have.
Third, a model is to be judged not only on how much it explains, but also on how simply and elegantly it explains it. So, if a model can both account for more data than another and can do it more simply than another model, then it is, hands down, the better model.
Fourth, historical sources are by their very nature "messy." As in scientific measurements, there is always a level of randomness and "noise" in the historical data. For instance, despite the overwhelming mass of sources reporting that Joseph Smith practiced polygamy, there is enough contradiction between sources and counter-evidence (such as Emma Smith's denials that her husband practiced polygamy) that some intelligent people have been able to put together entire books arguing that Joseph Smith did not practice polygamy. Yet, you'll notice, while this historical "noise" has been enough to trip up some non-historians, it has not fooled any actual historians. No model will perfectly account for all the data, but it doesn't have to. The preferable model is the one that provides the bestexplanation rather than than an unattainably high standard of perfect explanation of all noise. So, historians recognize that despite the noise in the historical record on Joseph Smith and polygamy, the model of Joseph Smith as a polygamist is by far a better explanation than the model of Joseph Smith as a monogamist.
_________________________________________________________________________________________________
I'll add here that probably the two biggest errors I've seen non-historians (and even sometimes historians) make in the interpretation of historical data are:
1) Not realizing that historical data often conflict.
Because people don't typically realize this, they often believe that having any data points in favor of an interpretation make it true or that having any data points against it make it false. Rather, as noted above, because the data do often conflict, we look for the overall pattern of the data to determine which model works best.
(However, since we have no way of determining at the outset what data may be mere "noise," we should treat all historical data as potentially significant; and data that seem at first to be anomalies or noise can eventually serve as evidence for further refining an explanation. The error comes just in assuming that historical models have to have perfect consistency--by the nature of differing human perspectives, bias, the imprecision of language, and the decay of memory, they never do.)
2) Treating historical sources (records, testimonies, etc.) as authorities rather than as evidence.
The difference between these two approaches is stark.
If I view historical sources as authorities, then I compare different accounts of an event, select the one I deem most authoritative, and then reproduce that person's account as the definitive narrative of the event. And if there is no account that tells me outright that person A did X, then I am simply silent on the question, believing it cannot be known.
If, however, I view historical sources as evidence, I comb through all accounts of the event--and other historical documents related to it, treat minutiae in all these sources as potentially significant, and engage in detective work that, usually, accepts no one account as definitive but infers details of the event from a number of the sources. Think detective work. You may have reason to believe that one witness is completely honest, but honesty is no guarantee of accuracy. On the other hand, another witness may be completely dishonest, but you can infer that his lies will almost certainly be biased in favor of his own interests; so taking his story into account in the context of his bias, you may actually be able to infer to what actually happened. The great philosopher of history Robin Collingwood gave examples of this. And he wrote cogently about treating sources as evidence rather than as authorities and provided a brilliant exposition of historical method in the epilogues to his The Idea of History. (An application of this in biblical studies is where scholars use the "criterion of embarrassment" as a way of sifting out truth--if an author with a known bias offers a narrative or detail that cuts against the grain of his own bias, then you can be reasonably sure he didn't make it up).
An example of using historical sources as evidence, rather than as authorities, may be found, for instance, in the study of Joseph Smith's translation of Isaiah in his Bible revision (JST). Both the Book of Mormon and the JST vary from Isaiah in the King James Version. So, as an undergraduate, I wanted to know: Did Joseph Smith consult the Isaiah variants in the Book of Mormon to tell him how to edit Isaiah in the JST? Did he actually open up the Book of Mormon and dictate the Isaiah variants there to his JST scribe? In discussing this question with a noted authority on the JST I was told, "I don't think we'll ever know that. I don't think we can second guess the Prophet Joseph Smith." This scholar's reasoning appeared to be that since there was no historical account of whether Joseph Smith used Book of Mormon Isaiah in the JST--i.e., because there was no "authority", no one who explicitly said whether Joseph Smith did this or not--there was just no way of knowing.
That was a very silly assumption. Although we had no account from a putative historical "authority" on the subject, we had the actual texts of the Book of Mormon and the JST. All it took to figure out what Joseph Smith had done was to compare those texts. Extensive comparison of Isaiah in the 1830 Book of Mormon with Isaiah in the earliest Book of Mormon manuscripts showed that the copyists and typesetter had introduced 22 errors into the text of Book of Mormon Isaiah. Comparing these with JST Isaiah showed that 21 of those 22 errors showed up in the JST. Case closed.
This pattern of textual errors is as distinctive as a fingerprint: there is no other way the 'fingerprint' of Isaiah in the 1830 Book of Mormon was going to get onto JST Isaiah than from Joseph Smith importing it from 1830 Book of Mormon Isaiah. Thus, even thought we have no "authority" giving an account on the process used, the process can be uncovered quite clearly. In fact, this evidence is so powerful that even if we had a historical account that said Joseph Smith did not use Book of Mormon Isaiah in the JST, we would have to reject the alleged authority since we could prove it false from the evidence provided by the Isaiah text itself. (We see something similar with regard to the Kinderhook plates--the text of the GAEL finds its way into Joseph Smith's translation from the Kinderhook plates, showing that he used the GAEL as a tool for translating from those plates--although in that case we also have eye-witness testimony that dovetails with the textual evidence.)
Collingwood's epilogues in The Idea of History are the best introduction I know to historical methodology, and I'd highly recommend them to anyone interested in how to do history and do it well.
Don
A few thoughts on historical method...
First, historical explanation is a process of inference to the best explanation. We seek to piece together the fragments of the past in the way that offers the best overall explanation. The more data an explanatory model can account for, the better for that model.
Second, what we mean by explaining the data here is that we can explain the various historical sources and why the events came out the way they did--why they came out that way in particular and not any of the other countless ways they could have.
Third, a model is to be judged not only on how much it explains, but also on how simply and elegantly it explains it. So, if a model can both account for more data than another and can do it more simply than another model, then it is, hands down, the better model.
Fourth, historical sources are by their very nature "messy." As in scientific measurements, there is always a level of randomness and "noise" in the historical data. For instance, despite the overwhelming mass of sources reporting that Joseph Smith practiced polygamy, there is enough contradiction between sources and counter-evidence (such as Emma Smith's denials that her husband practiced polygamy) that some intelligent people have been able to put together entire books arguing that Joseph Smith did not practice polygamy. Yet, you'll notice, while this historical "noise" has been enough to trip up some non-historians, it has not fooled any actual historians. No model will perfectly account for all the data, but it doesn't have to. The preferable model is the one that provides the bestexplanation rather than than an unattainably high standard of perfect explanation of all noise. So, historians recognize that despite the noise in the historical record on Joseph Smith and polygamy, the model of Joseph Smith as a polygamist is by far a better explanation than the model of Joseph Smith as a monogamist.
_________________________________________________________________________________________________
I'll add here that probably the two biggest errors I've seen non-historians (and even sometimes historians) make in the interpretation of historical data are:
1) Not realizing that historical data often conflict.
Because people don't typically realize this, they often believe that having any data points in favor of an interpretation make it true or that having any data points against it make it false. Rather, as noted above, because the data do often conflict, we look for the overall pattern of the data to determine which model works best.
(However, since we have no way of determining at the outset what data may be mere "noise," we should treat all historical data as potentially significant; and data that seem at first to be anomalies or noise can eventually serve as evidence for further refining an explanation. The error comes just in assuming that historical models have to have perfect consistency--by the nature of differing human perspectives, bias, the imprecision of language, and the decay of memory, they never do.)
2) Treating historical sources (records, testimonies, etc.) as authorities rather than as evidence.
The difference between these two approaches is stark.
If I view historical sources as authorities, then I compare different accounts of an event, select the one I deem most authoritative, and then reproduce that person's account as the definitive narrative of the event. And if there is no account that tells me outright that person A did X, then I am simply silent on the question, believing it cannot be known.
If, however, I view historical sources as evidence, I comb through all accounts of the event--and other historical documents related to it, treat minutiae in all these sources as potentially significant, and engage in detective work that, usually, accepts no one account as definitive but infers details of the event from a number of the sources. Think detective work. You may have reason to believe that one witness is completely honest, but honesty is no guarantee of accuracy. On the other hand, another witness may be completely dishonest, but you can infer that his lies will almost certainly be biased in favor of his own interests; so taking his story into account in the context of his bias, you may actually be able to infer to what actually happened. The great philosopher of history Robin Collingwood gave examples of this. And he wrote cogently about treating sources as evidence rather than as authorities and provided a brilliant exposition of historical method in the epilogues to his The Idea of History. (An application of this in biblical studies is where scholars use the "criterion of embarrassment" as a way of sifting out truth--if an author with a known bias offers a narrative or detail that cuts against the grain of his own bias, then you can be reasonably sure he didn't make it up).
An example of using historical sources as evidence, rather than as authorities, may be found, for instance, in the study of Joseph Smith's translation of Isaiah in his Bible revision (JST). Both the Book of Mormon and the JST vary from Isaiah in the King James Version. So, as an undergraduate, I wanted to know: Did Joseph Smith consult the Isaiah variants in the Book of Mormon to tell him how to edit Isaiah in the JST? Did he actually open up the Book of Mormon and dictate the Isaiah variants there to his JST scribe? In discussing this question with a noted authority on the JST I was told, "I don't think we'll ever know that. I don't think we can second guess the Prophet Joseph Smith." This scholar's reasoning appeared to be that since there was no historical account of whether Joseph Smith used Book of Mormon Isaiah in the JST--i.e., because there was no "authority", no one who explicitly said whether Joseph Smith did this or not--there was just no way of knowing.
That was a very silly assumption. Although we had no account from a putative historical "authority" on the subject, we had the actual texts of the Book of Mormon and the JST. All it took to figure out what Joseph Smith had done was to compare those texts. Extensive comparison of Isaiah in the 1830 Book of Mormon with Isaiah in the earliest Book of Mormon manuscripts showed that the copyists and typesetter had introduced 22 errors into the text of Book of Mormon Isaiah. Comparing these with JST Isaiah showed that 21 of those 22 errors showed up in the JST. Case closed.
This pattern of textual errors is as distinctive as a fingerprint: there is no other way the 'fingerprint' of Isaiah in the 1830 Book of Mormon was going to get onto JST Isaiah than from Joseph Smith importing it from 1830 Book of Mormon Isaiah. Thus, even thought we have no "authority" giving an account on the process used, the process can be uncovered quite clearly. In fact, this evidence is so powerful that even if we had a historical account that said Joseph Smith did not use Book of Mormon Isaiah in the JST, we would have to reject the alleged authority since we could prove it false from the evidence provided by the Isaiah text itself. (We see something similar with regard to the Kinderhook plates--the text of the GAEL finds its way into Joseph Smith's translation from the Kinderhook plates, showing that he used the GAEL as a tool for translating from those plates--although in that case we also have eye-witness testimony that dovetails with the textual evidence.)
Collingwood's epilogues in The Idea of History are the best introduction I know to historical methodology, and I'd highly recommend them to anyone interested in how to do history and do it well.
Don