Refugeed Slaves

Contemporary Impressions

Numerous contemporary observers commenting on the large influx of slaves who were being “refugeed” to Texas during the Civil War. Several pages in the diary of Arthur J. Fremantle commented on the Texas Roads being “alive with negroes” who were being run out of Louisiana because of Union occupation there. And earlier influxes from Arkansas and from Indian Territory had been sparked by Union victories at the Battle of Pea Ridge.

Opinions about this influx ranged from concern about incoming slaves who were unsupervised or dangerous, as mentioned in baum2008, p. 92, to complaints that refugeed slaves were not being targeted by Confederate Slave Impressment while Texan slaveholdings were.

But other observers seemed to see the arrivals in terms of economic opportunity; the Marshall Texas Republican even suggested that refugeed slaves be put to work on civilian public work projects like the improvement of the roads.1 According to an entry in the Encyclopedia of Arkansas, William Williston Heartsill, a Texas cavalryman, made a similar point in his November 1862 entry:

Every day we meet refugees with hundreds of Negroes, on their way to Texas; thus we see that as the war is depopulating and ruining this country, it is building up the ‘Lone Star State.’2

See also this reprinted excerpt of a Galveston newspaper that describes some groups of refugeed slaves who had been recaptured after the fall of Brashear City.

Governor Francis Lubbock also addressed worries about refugeed slaves in his address to the tenth legislature on November 4, 1863:

Since the invasion of Arkansas and Louisiana, many citizens of those states are seeking refuge within our state, accompanied by their Negroes. In consequence of this influx of so large a number of this class, apprehension is entertained by our citizens of a scarcity of provisions as well as other dangers. I see no reasonable grounds for such apprehensions. There is an abundance of bread-stuffs and meat in the state, and labor is becoming very scarce and high, in consequence of the continued increase of our army. All that is requisite in order that this immigration may be productive of good is to take such measures as will keep these Negroes beneficially and constantly employed. Let large drafts for army purposes be made from them. Let companies of local police be organized and see that the remainder be usefully employed, and they will thus prove a source of strength instead of weakness. Much better it is for us to receive them upon these terms than that they should fall into the hands of our abolition enemies to be used against us. The refugee who seeks the last foot of soil unpolluted by the presence of the Yankees is far more entitled to our respect, sympathy and protection, than the wretched cravens and traitors who remain within the enemies lines taking the oath of allegiance in the vain expectation of preserving the property they have not the courage or patriotism to defend.3

How many came to Texas?

I’ve written up a white paper on this question.


A postwar report by the Inspector General of the Freedmen’s Bureau, William E. Strong, estimated that around 125,000 were sent to Texas from other Southern states.4 According to Strong, of that number brought

from Louisiana, Mississippi, and other States, nearly all of them are anxious to return to their old homes, or, at all events, to get out of Texas. Thousands of this number have already returned, and there is a constant stream pouring through the interior of the State in an easterly direction, heading towards Louisiana. The route usually taken by these people is the old San Antonio road leading from Bastor [Bastrop?], on the Colorado, through Caldwell, Madison, Crockett, and from thence running a little north of east to Millan, on the Sabine river, near the Louisiana line. This road is famous as being the first and best route across Texas, and all the refugees get to this road as soon as possible.


berlin1987 cites Asst. Engineer Theo. Herman to Captain E. A. Warren, 4 December 1864, as source of claim that 150,000 were forcibly moved from Mississippi Valley to Texas just in the year after Vicksburg. baum2008 attributes this same number to Magruder, but says that according to Magruder it was the number taken from Arkansas and Missouri alone. sternhell2012 says 150,000 citing berlin1985, while litwack1979, p. 32, also uses the figure to refer to the number “sent out of Louisiana and Mississippi.”5 Thavolia Glymph also says 150,000:

Texas represented the western-most edge of the Confederacy but it was not a sealed-off state. In fact, by the end of the war, it was home to thousands of slaves whose masters had refugeed them to Texas to prevent their escape to Union lines. Some 150,000 slaves were taken to Texas after Union forces captured Vicksburg. These slaves would have brought news of the war, how the Union armies were doing, and especially of the Emancipation Proclamation. Enslaved men confiscated from their masters by the Confederate army to be put to military labor also would have been important links to the outside world.

One potential source for the 150,000 figure may be John Bankhead Magruder.6 After being reassigned to the Department of Arkansas in August 1864, Magruder wrote to one of Arkansas’s Confederate senators in November on the subject of Confederate Slave Impressment. General Kirby Smith had apparently recommended that slaves be impressed in Arkansas to support an upcoming campaign to Fort Smith. Magruder, however, argued against the impressment of Arkansas slaves because their numbers had already been depleted there by flight to Texas:

I am told that over 150,000 negroes have gone from Missouri and Arkansas into Texas, and leaving out all considerations of fairness, I fear that if the few that remain here are impressed, those who would otherwise sow and plant would emigrate to Texas, and through the depopulation of the country we should not be able to support an army through another season, though otherwise successful.

Clearly, then, a high estimate favored Magruder’s purposes. Magruder continued that he had advised Smith to attach some Texas counties to the the Arkansas department for the impressment of slaves, “as I understand he has impressed negroes to work upon the fortifications at Marshall and Shreveport.” Magruder knew that this would not please area slaveholders who probably did not expect further impressments from Smith, but he believed that from “Harrison County to the Indian Territory” there were a considerable number of relocated slaves. He hoped that his correspondent will persuade General Smith “as to the impolicy of impressing negroes in Arkansas,” as well as the great need for slaves to be impressed from somewhere. As evidence of the rhetorical, rather than empirical, basis for Magruder’s estimates, in another letter on December 4 complaining of orders to impress slaves in Arkansas, he asserted that “from 100,000 to 150,000 slaves had gone to Texas from Arkansas and Missouri.”7


Randolph Campbell estimates 32,000 using tax rolls.


Dale Baum expands Campbell’s estimate to a conservative 51,000.

My Questions

  • How many eastern Texas slaveholders removed to the interior, perhaps even beyond the tax assessors’ reach, and how many came from counties impacted by refugeeing?

  • Were there counties that had an above average increase in slaves taxed between 1862 and 1864 but did not have an increase of 1,000 slaves? (I.e., counties that would have been on baum2008, Table 1, Part B if not for his requirement that they have added more than 1,000 slaves.) Did any of these counties experience significant increases relative to their pre-1862 size?

  • How often did refugee slaveholders move once they arrived in Texas? The snapshots by Baum and Campbell may freeze the situation in 1863, but if slaveholders moved on, or had settled elsewhere in Texas first, we might not know it from the statistical evidence.

  • Would it be possible to reproduce tables like the one in baum2008 for other counties, using


WPA Narratives by Refugeed Slaves

Running Away

Examples of slaves who ran away after arriving in Texas include:


Examples of slaves sick on the journey to Texas include:

  1. Marshall Texas Republican, October 7, 1864.

  2. For original quote, see Bell Irvin Wiley, ed., Fourteen Hundred and 91 Days in the Confederate Army: A Journal Kept by W. W. Heartsill, for Four Years, One Month, and One Day, or, Camp Life: Day-by-Day, of the W. P. Lane Rangers, from April 19th, 1861, to May 20th 1865 (1876; Jackson, Tenn.: McCowat-Mercer Press, 1953), available at Hathi Trust.

  3. Message of Governor, November 4, 1863, House Journal of the Tenth Legislature, Regular Session, of the State of Texas, November 3, 1863 - December 16, 1863, ed. James M. Day (Austin: Texas State Library, 1965), 11, link.

  4. William E. Strong to Major General O. O. Howard, January 1, 1866, in “Freedmen’s Bureau Orders,” House Executive Documents 70, p. 312.

  5. Drawing on Litwack, Henry Louis Gates also recently cited the 150,000 figure.

  6. This is the source cited for the figure by kerby1972, p. 255, which is the source for the slightly different claim by levine2013, p. 155. Kerby says Magruder claimed that “General Magruder estimated [in late 1864] that since the beginning of the war the enslaved population of Texas had been swollen by the addition of 150,000 refugee bondsmen,” while Levine writes that “by late 1862, some 150,000 slaves had been refugeed into Texas alone.”

  7. There is also evidence of concern from Smith in November 1864 about too many slaves being taken out of Louisiana, depleting the crops needed to provision troops. See berlin1987, pp. 780–781. Since this period is also the same general time as the letter cited by berlin1987 from Theo. Herman on p. 676 (December 1864), it makes sense that high numbers were being floated at a time of concern about impressment and provisioning. See also Smith letter to Magruder and Magruder letter from December 15.