Why are older books only useful for research near the end of a project?

It was mentioned in another thread that, "Older books are only useful near the end of a project." A member asked "Why is that?" Seemed like it deserved a general response.
There are many excellent 19th and 20th century numismatic books that have established a solid reputation for methodology, accuracy and scope. These are great resources for research, but most are also out of date in one or more significant ways. Maybe new varieties have been discovered, or previous ones discredited, or better understanding of technology has made manufacturing descriptions obsolete.
One of the most common difficulties, however, is the nearly complete absence of clear, reliable sources used by the author to substantiate conclusions. This is especially noticeable in reference books published from about 1900 to 2000. This is also the primary reason I advocate referring to these materials only after most of the data collection, analysis and results are prepared. Experience has shown this approach avoids early assumptions, misdirection and wasted time. The old materials are useful but more as backward confirmation and possible expansion than as primary sources.
For related insights, go to the Numismatic Bibliomania Society website, or NNP and look for the ANA presentation "Concepts of Numismatic Research."
Comments
I search everything early commemorative related. I do start in 1892 and work my way to 1954.
I'm going to have to get new glasses cause I must have read 1,000 or more reference points.
Older books are useful throughout the life of a project. Errors can be corrected as you go along.
a.k.a "The BUFFINATOR"
One reason would be that the newer books should reference older books, which would lead you to the oldest books last. It could be that by the time you get to a really old book, all the good stuff in it has already been framed in a more modern context or corrected, and there's not much left in the old book beyond satisfying one's curiosity as to the thoughts of the day.
Keeper of the VAM Catalog • Professional Coin Imaging • Prime Number Set • World Coins in Early America • British Trade Dollars • Variety Attribution
Old books can easily lead to incorrect assumptions and false conclusions. They are also unlikely to be based on careful research (with some exceptions). Beginning with outdated material is a good way to completely screw up a research project.
Here is an older book, though containing errors, that is very useful to all numismatists.

a.k.a "The BUFFINATOR"
Good information Roger, Thank you. I like old books, and they are a valuable resource. That being said, yes, in many cases the information has since been corrected, modified or shown to be an opinion rather than supported fact. Cheers, RickO
The Breen tome is a good example of something to avoid until most data collection and analysis has been completed. It is an unreliable book - containing much good material and much erroneous and outright lies. Sadly, separating the mess is nearly impossible except on a specialist level, and that separation is possible only because of new research (within the past 20 years). Further, Breen gives few useful source citations. The overall age of the book necessarily omits thousands of auction and private sale events, further limiting the utility of the reliable portions.
Overall, if one starts research with old materials, it will be found to have been a great waste of time and effort for no meaningful gain. To obtain the highest accuracy and research efficiency for numismatics, start with the original sources and work back. Correlation with Breen, Taxay, and others comes near the end of work.
“It is an unreliable book - containing much good material and much erroneous and outright lies.”
Why would you say “outright lies?” What would that have accomplished for Breen?
I can see you using the term “erroneous” cause he sure did make mistakes but “ought lies?"
a.k.a "The BUFFINATOR"
To frame this a different way, nearly all numismatic books are secondary sources (exceptions would be something like first-person memoirs or diaries, maybe others). If you start with secondaries your model could very well be biased from the get-go by the way in which the narrative(s) are presented there. Starting with primary literature is much harder, as it takes time to read, assimilate and build a coherent narrative. Secondaries will get you there faster but it may not be where you meant to go.
a.k.a "The BUFFINATOR"
Trolling Level: Master
Breen regularly made things up - characterizing some of these stories as outright lies doesn't seem all that extreme to me. Some of my favorite Breen stories revolve around his propensity for high-level bullshit.
If you know Jerry Bobbe, ask him to tell you about the time Breen met God (and Mrs. God) on Glastonbury Tor; I also used to love a story Art Kagin told me about Breen giving him details about one of the weddings of King Henry VIII - the story was complete with a list of songs played during the ceremony, a description of who sat where, and what they were wearing (details that don't exist in the historical record).
What is now proved was once only imagined. - William Blake
And in doing so, he undermined all of his other works that contained good information, because once you've been established as a habitual liar, nothing you say or have said can be accepted as truth without further scrutiny.
I know Jerry Bobbe, and this is now on the top of my list of things to ask him about next time I meet him, because Jerry's delivery of the story is probably going to be every bit as entertaining, if not more, than the story itself.
Keeper of the VAM Catalog • Professional Coin Imaging • Prime Number Set • World Coins in Early America • British Trade Dollars • Variety Attribution
It's a classic - I make him tell it almost every time I see him.
What is now proved was once only imagined. - William Blake
I don't know a lot about books or who you're talking about but I do love watching Toshiro Mifune!
Collector of randomness. Photographer at PCGS. Lover of Harry Potter.
RE: "Why would you say “outright lies?” What would that have accomplished for Breen?"
Breen's Encyc has multiple entries that are completely and totally false -- no factual basis at all. They are outright, blatant lies. You'll have to search his psychological profile to discover his possible reasons. These lies have been discussed multiple times on multiple forums, so research of the subject is possible.
"Errors" are of two kinds: unintentional/mistakes; and inventions to cover a lack of diligence. John Dannreuther uncovered a bunch of the latter when examining coin in the NNC several years ago. Many collectors and specialists have identified the former. No author's work can be completely free or mistakes/unintentional errors. But Breen's are especially galling because of his claims to a superb memory, and that he easily reconstructed for pubication materials after losing a nearly complete manuscript (if it actually existed). Dorothy Paschal expressed some concern about this.
Breen's work for JJ Ford, some of which I've seen in manuscript thanks to the late Eric Newman, indicate his willingness to ignore the truth to satisfy Ford. We don;t know how much of this kind of deception was put into the Encyc at the demand of First CoinVestors (or should that be "ConVestors?")
After a while, as with some notable public figures, one runs out of excuses for lies, "mistakes," deliberate omissions, falsification and the continued litany leading to mistrust of anything Breen wrote. Modern researchers, using solid technique, must independently confirm everything Breen claimed regardless of his former good reputation.
Thanks for the information, very interesting.
tom
a.k.a "The BUFFINATOR"
Also, see the thread "Walter Breen's accuracy" and specifically the post by Rittenhouse at the end.
Here he is in all his bearded glory. (My wife does not allow me to grow facial hair!)
a.k.a "The BUFFINATOR"