Interpreting Agricultural Census Data

29 April 2014

A historical monograph published by a university press invokes the agricultural schedule of the 1850 U.S. census and asserts: "Considering [this individual's wealth], the census statistics regarding cattle and hogs seem unusual. ... No cattle, milk, or beef was listed (probably to escape paying taxes on them), and there were sixty-five hogs. The value of livestock was $1,310, most of it represented by horses."*

What is your reaction to the authors' interpretation of the census data? Would your interpretation differ?

(*The source will go uncited in this particular case. The quote is used for analytical purposes; the object is not to rap the author's knuckles.)

Submitted bykdauon Tue, 04/29/2014 - 14:02

I think the most salient issue is the idea that the census return was falsified to evade taxes. No local agency assessing property tax should have had access to the individual-level census data. In a small community where the census enumerator (assistant marshal) might be known to the subject individual, it is conceivable that the individual falsified their census return for consistency with falsified local tax assessments. That, on the other hand, brings to mind absurd images of scores of cattle being herded into hiding whenever any official came near. It just seems to be an odd assertion to make without any further evidence.

There is also the question of whether "No cattle, milk, or beef was listed" refers to columns that were left blank or columns with zeroes written in them. I have seen several agricultural schedules in which enumerators appear to have omitted certain questions entirely, with no farms on a sheet having any value listed for a certain column - even in a region where a value might be expected. Cattle and their products were certainly more common (than hemp or cane sugar, for instance) and perhaps less likely to be skipped. Still, I would consider an empty cell to be relatively weak as negative evidence compared to a zero.

If I recall the instructions correctly, the "Value of live stock" column was to include the value of all types of livestock on hand, even those not specifically enumerated on the sheet. Unless the author has access to other evidence of the farm's full complement (perhaps an estate inventory), I'm not sure they can safely assume that horses were the majority of the $1,310 amount. Even the comparative number of horses and hogs might be misleading without knowing their ages and conditions.

More generally, I would want to see some context to interpret the relationship between overall wealth and predicted livestock count. Did the subject individual live in an area where cattle were often a significant part of farming operations? Was the individual returned as a farmer in the population schedule, or did they have an additional occupation? What was the value of real estate given in the population schedule?

I love the agricultural schedules for the snapshot of farm life that they can give us, but I think that these difficulties of interpretation underscore that these returns should be a starting point, not a standalone source, for understanding a subject's farming activities.

Interesting observations, Kevin. You include issues we've never had raised when using this problem for classroom instruction.  We'll reserve our comments until tomorrow, so that others can add that own analyses.

Kevin, you nailed the salient point: the idea that the census return was falsified to evade taxes. Given that this university-press monograph was a study of a frontier pioneer, assertions that he gave false data to evade taxes is a significant interpretation of his character—and one that could affect many other conclusions the authors would then proceed to make about him.

Among all your superb observations, there is only one point on which we'd quibble. The second sentence ("no local agency assessing property tax should have had access to the individual-level census data") is a good springboard for discussing the differences between historic and modern mindsets and laws. In their era, there was no privacy insofar as census information was concerned. From 1800 through 1880, censuses were made in multiple copies and one copy was publicly posted, locally, to better ensure accuracy. Moreover, various census takers were indeed known to consult the local tax records to complete their returns.

 

Submitted byyhoitinkon Tue, 04/29/2014 - 16:33

I agree with Kdau's observation that the conclusion about evading taxes seems modern and not based on evidence. 

I have an observation about the value of the lifestock. According to this post,1 8 seven month old pigs sold for $91.25, or about $11.40 each, in 1850. Older and bigger pigs might have sold for more, but even at this price 65 hogs would be worth $741, more than half the value of the cattle. You would have to research local pricing conditions at that time but this rough estimate shows that the conclusion that the valuation was mainly based on the horses might be premature.

From a researchers point of view, I don't like the way the author mixed evidence from the census with his own opinions without any way to tell where one ends and the other starts. For example, I don't know if the phrase "most of it represented by horses" is based on something in the document or if it was a conclusion that the author drew. The paragraph would have been much more useful for researchers if the author had used editorial brackets for his own comments and/or notes in the middle of sentences to show which part was based on the actual source.

Sources:

  1. Alan MacRae, "The Farmer of 1850 Sells His Eight Pigs," peicanada.com The Voice of Rural Pei (http://www.peicanada.com/past_alan_macrae/columns_opinions/farmer_1850_sells_his_eight_pigs : accessed 29 April 2014)

 

Yvette,

The point you raise here is a major difference between conventions and standards in the discipline of history versus those in the discipline of genealogy. In history (my own academic background) we're taught to interpret and paraphrase. Each paragraph is, itself, a mini-essay that is ideally based on a study of multiple sources. We report the facts from each source in our own words, with the expectation that we will be attempt to be true to the tenor of the original. Because any individual sentence and paragraph may offer material from multiple sources, we're taught to place reference notes at the end of a sentence—or, preferably, at the end of a paragraph—in which case multiple sources would then be cited.

Genealogy's standards are much more precise. Because genealogists are micro-historians rather than macro-historians, errors on small details can affect the validity of all work done past that point. Therefore, in genealogy's expository writing, any sentence that both reports "facts" and offers interpretation is expected to make a clear distinction between what the record actually says and the interpretation that the writer is drawing from the record. If a writer were quoting from the document, then any addition of opinion should indeed be in editorial brackets but, then, good writing is expected not to be a patchwork of quotes.

All things considered, writing standards in genealogy follow the same basic rules for good expository writing and careful paraphrasing, but it is exquisitely precise in its placement of citations. In genealogical writing, a reference note would be placed at the end of any material that came from a specific source, rather than waiting for the end of a paragraph. A well-constructed sentence would indeed make a clear distinction between the content of the record and the author's own thoughts. That typically is accomplished by placing the reference note number in the middle of a sentence (or paragraph) to make the point that the assertions the writer just made are faithfully taken from the cited source, while the assertions in the rest of the sentence (or paragraph) is context or interpretation added by the writer.

 

Submitted byJadeon Tue, 04/29/2014 - 21:42

I think Kevin's and Yvette's observations are sound.

In addition I see the author as mixing apricots and grapes.  Why would "wealth" imply that the person would own all types of livestock?  What if a herd were lost to weather or disease in the past year or two?  Did the person in question just recently purchase some land together with existing chattels (that would have been amassed by someone else)?  Did other persons in the particular schedule have more of a mix of stock?  Why would absence of sheep not be noteworthy?  Has the author found and shared an assessment regarding suitability of the area lands for particular purposes?

I would find the account a trifle more credible if the author stated what the value of a horse ought to be.  I did a survey of one township's assessments in Ohio in the 1840s, and horses were consistently valued at $40, cows at $8 -- pigs not assessed.  The number of hogs could be a mix of grown swine and weanling shoats, which would have different values.  The author does not give us any details that would enable us to follow along to the conclusions.

I do not love "it's true because I say so" writing.

 

Submitted byErickdmon Tue, 04/29/2014 - 21:55

I would need more information on the subject before drawing any conclusions. I've used the agricultural schedules quite a lot, and they are no more or less accurate than other schedules (population, slave, manufacturing).  All are subject both to the information provided by whomever the census taker found to ask, and the assertiveness (or lack thereof) of the censustaker.  There is certainly no reason to think that there was an effort to avoid taxes, although I suppose it may be possible that the person being enumerated may have thought there was a connection.

All that being said, on its face, this type of enumeration might suggest that this farmer (or rancher?) focused his attention on raising swine for resale or slaughter, and horsetrading.  If the schuedule shows there was little cleared land, it would suggest a good habitat for hogs, which were often turned loose in the woods in those days. But I don't know where he lived - for all I know he lived on the prairie where there were no woods.

In summary, we need more information in order to properly interpret the agricularual census entry, and to determine if the article of the monograph is off base in his assumptions.

Erick Montgomery

Submitted byEEon Thu, 05/01/2014 - 09:11

In reply to by Erickdm

You're definitely right, Erick. We always need more info. The authors of the monograph did tell us much more about the individual, across his whole lifetime. On the other hand, the issue involved in the little passage we excerpted for "Tuesday's Test" is one that hinges upon context from elsewhere that we, as researchers, need to know in order to avoid similar misinterpretations.

On another issue: your Woodrow Wilson related article in the latest NGSQ is superb. Congratulations!

Submitted byEEon Wed, 04/30/2014 - 11:43

Great work, everybody! All of you have probed areas that most researchers miss. We'll add one more—the issue that propted this question in the first place.

The authors in question, like many history researchers, made a common but inaccurate assumption: To use a census, all we have to do is find the individual in question, copy down the data, integrate it into our narrative, and then explain what we've decided about it. Sheez! How complicated can it be to read census names and figures! Right? Wrong!

With censuses, the accuracy of our conclusions will depend upon whether we also study the instructions given to census takers. It can also depend upon whether we study the analyses made by (and published) the Census Bureau after each decennial count. In this specific case, the 1850–1900 practice was summed up in the compendium for the 1900 census, in the preface of the Bureau director:

"In 1850, as in all succeeding census years, there were many individuals who made use of public lands or of the unfenced and unused lands of private individuals for herding sheep and cattle. ... No attempt was made to enumerate or estimate the number of animals on ... unfenced lands, or to ascertain the number of ranches, farms, dairies, or similar establishments with which these animals were connected."1

So: are you now rethinking conclusions you may have made about cattle ownership and management on your farms in your own research?
---------------
1William R. Merriam, Director, Bureau of the Census, Census Reports, vol. 5, Agriculture, Part 1, Farms, Livestock, and Animal Products (Washington: U.S. Census Office, 1902), "Introduction," xv.

Note: If you haven't seen it already, you may be interested in "Census Instructions? Who Needs Instructions?" QuickLesson 9 at https://www.evidenceexplained.com/content/quicklesson-9-census-instructions-who-needs-instructions. 

Submitted byyhoitinkon Wed, 04/30/2014 - 13:03

Awesome reminder of how understanding the conditions under which the source was created is essential to interpret the data. In the Netherlands, all communal grounds had been divided by 1850, often centuries before, so it did not occur to me that that might not have been the case in the US. It should have, as I've done enough research into Dutch settlers in the US in the 19th century. 

Were communal grounds still in existence throughout the US at that time or was that typical for some states?

Yvette,

Communal grounds were not the common pattern. Outside New England (which means, of course, the vast majority of the country), communal anything was not the pattern! The U.S. of A. was definitely a country built on the spirit of individual entrepreneurship.

Submitted byyhoitinkon Thu, 05/01/2014 - 13:42

In reply to by EE

Having grazing rights is not what I would call the American dream, no :-) The fact that this custom was limited to New England also explains why I haven't run into this before: 'my' Dutch emigrants all settled further west.

Submitted byJadeon Wed, 04/30/2014 - 13:48

Thus far, I am not aware of relatives who would have utilized unsettled lands during the 1850-1880 time period for which the agricultural schedules are readily available, but this is certainly a great heads-up.  Some were living then in areas where there were still Town Lots modeled after schemes in New England whence the earlier settlers had come, and the Hog Reeve or bailiffs sometimes found themselves in Court testifying as to residents' abuse of this land, or animals escaping from the pound.

It had not occurred to me that the instant example might have concerned unsettled or abandoned land, or formally designated grazing lands.

Thank you for this!

 

Submitted byEEon Thu, 05/01/2014 - 08:23

In reply to by Jade

Jade,

New England communities did have a different settlement pattern, a different land-ownership system, and a different lifestyle from most of the continental U.S. Aside from the original thirteen colonies and the states cut from them, the other regions in 1850 had much land that had never been purchased from the government. (Some of the western states still do.) Also, there were so many wild, large animals that fencing went around crops, while domestic animals were allowed to roam free.

Another significant factor in the mid-1800s was that, in the public-land states, most owners had cleared only a portion of their land, which meant that cattle ranging in the woods weren't necessarily ranging on "unsettled or abandoned land." They could well be on the owner's property, but sending census takers into what could be hundreds of acres of woods to try to find all of each owner's cattle was a bit more than the Census Bureau was about to take on—and more than most landowners would have tolerated.

Submitted bymjneillon Wed, 04/30/2014 - 15:09

The author needs to provide some context for his statements regarding the livestock. Saying something is "unusual" without discussing what was the more common practive gives the reader no point of reference.  It would have strengthened the argument to have made even a brief comparison to other enumerations within the same township. National averages and summaries are fine, but the most accurate interpretation of this type of data is often made at the micro-level. Also without knowing where the property was located, it is difficult to determine if cattle farming was done for family use only or was commonly done on a larger scale. 

And I'm always skeptical of valuations whether or not the tax man finding out. Valuation is more subjective that item counting. It also does not matter if the tax man found out about the valuations or not. If local farmers had the perception that information was "share," then that perception impacts their answers and it really does matter if the information was shared with taxing officials or not.

Submitted byEEon Thu, 05/01/2014 - 08:26

In reply to by mjneill

Good points, Michael John. For the record, the area was Florida. And 1850 was not long after John Randolph stood in the halls of Congress and declared No man would move to Florida—not from hell itself!

Submitted byJadeon Thu, 05/01/2014 - 11:25

Dear Editor,

I did omit mention of an owner's unfenced lands.  Here again, the reader needed the Ag. Schedule details on the owner's lands, and for the writer to present more detail as to conclusions on land use.

None of my relatives went into Montana and the Dakota Territory until the 1890s+.  The one who took up land in Wyoming in the 1880s has not been found in a US Census enumeration there (a newspaper report suggests interest in oil; he lived in a series of Homes for Disabled Civil War Veterans).  I have not seen examples of how the land and livestock descriptions were configured for these late arrivals.

I am not certain that I completely believe the Census Bureau's assertion that the Ag. Schedule enumerators did not attempt to assess numbers of turned-out stock.  The Deputy Marshalls were not assessing state/local tax valuations, and there was no reason for the owners to conceal stock unless obtained by some nefarious means.  However, in geographic areas where taxes were levied on certain livestock, it is possible that the enumerator and the assessor were at least sometimes the same person.

From the information we are provided in the instant case, we do not know whether specific stock or any at all was taxed.  Is this an assumption on the author's part?

Florida is a surprise.  My relatives who settled there were all around Lake Worth and Jupiter Inlet (the postman delivered by boat), except for the smattering of late-in-life arrivals post-1910.

The more we look at this the more questions arise.  I am glad to have this example for chewing on a bit more.

Thank you!