# Beatrice Houston v. Pat W. Haley

> Court of Appeals for the Fifth Circuit · November 1, 1988 · 859 F.2d 341

URL: https://www.frixlaw.com/law-library/cases/9478253

## Case

- **Full name:** Beatrice HOUSTON, Et Al., Plaintiffs-Appellants, v. Pat W. HALEY, Et Al., Defendants-Appellees
- **Court:** Court of Appeals for the Fifth Circuit
- **Decided:** November 1, 1988
- **Citations:** 859 F.2d 341; 1988 U.S. App. LEXIS 14693; 1988 WL 106239
- **Precedential status:** Published
- **Opinion:** Dissent by Brown
- **Judges:** Brown, Gee, Garwood
- **Cited by:** 13 later opinions in the Frix Law Library

## Citator (automated)

- **Red flag:** Vacated on other grounds by Beatrice Houston v. Pat W. Haley, 869 F.2d 807 (1989).
- Negative treatments: 1
- Distinguished by: 0
- Full citator and citing cases: https://www.frixlaw.com/law-library/cases/9478253

## Opinion text

JOHN R. BROWN, Circuit Judge,
dissenting.
I cannot join, in the court’s opinion in this case. The gains for black political participation for which members of this court and many others persevered in the past two decades have been significant and substantial. It cannot truly be said, however, that equal access to the political process has yet been achieved by blacks — and once achieved, someday, that equal access is unlikely ever to be “assured.” The advances made thus far remain fragile, and in need of jealous guardians. We must reach out affirmatively to protect the advances made thus far and to nurture those that remain to be made.
What’s in a Number
As the majority acknowledges, the Jordan court rejected alternative electoral plans “that would have achieved a significantly higher black voting-age population of (approximately 60%) [,] ... deciding] that ‘a clear black voting age population majority of 52.83% [was] sufficient to overcome the effects of past discrimination’ ” (emphasis added). Having analyzed Jordan in terms of figures for black population of voting age, however, the majority engages in what appears to be a sophisticated Socratic dialogue on the deduction/addition of the 5-9% non-voting age from/to the 53%/47% black/white population resulting, in one instance, in a supposition that in 1980 there was a majority of black voters of voting age. Solving the riddle the majority concludes that the 53.90% black population of all ages created in Ward 4 by the 4-and-l plan somehow “satisfies” the threshold suggested in Jordan. In the worn out, war weary figurative, such a conclusion erroneously compares apples and oranges. I insist — and continue to insist — that we must compare either (i) the Jordan figure for black population of voting age to the Ward 4 figure for black population of voting age, or (ii) the Jordan figure for black population of all ages to the Ward 4 figure for black population of all ages. We cannot sensibly compare a figure for black population of voting age in one case to a figure for black population of all ages in another case.
The black population of all ages in the black majority district created in Jordan was 58.30%. 1 The black population of all ages in the black majority district at issue in the instant case is only 53.90%.
The black population of voting age in the black majority district created in Jordan was 52.83%. 2 The black population of voting age in the black majority district at issue in the instant case was not directly in evidence. Therefore, if Jordan is to establish a threshold to be applied in this case, logically that threshold must be the 58.30% figure for the black population of all ages in the black majority district created in Jordan, and not the 52.83% figure for the black population of voting age in the black majority district created in Jordan which the majority opinion employs in this case. 3
Though there was no direct evidence as to the black population of voting age in the black majority district at issue in the instant case, that is not to say that there was no effort made at trial to address black population of voting age. The black population of voting age was indeed demonstrated, although indirectly. Expert testimony was received on this point from Mr. Louis Armstrong, a professional demogra *350 pher: 4
[Y]ou are talking about a district [Ward 4] that is 53 percent total population [i.e., in which nonwhites comprise 53.8% of the total population of the district]. I’ve not done an analysis on voting age population, but based on my experience across the state of Mississippi in more than 50 cases like this, there is a 5 to 9 percent difference between the [black] voting age population and the total [black] popula-tionl 5 ! And so we’re talking about at the most a 48 percent black voting age population district [i.e. a district in which the proportion of blacks of voting age to the total population of all ages in the district is approximately 0.48]. Then when you add the other socioeconomic characteristics, low income level, the low education level, the low employment level, then you begin to find that blacks participate less in government proportionally than whites based on these socioeconomic characteristics.
This expert’s testimony, presented in tabular form, indicates:
[[Image here]]
[[Image here]]
On this basis, the majority could conclude that it is possible to extrapolate a minimum black voting age population in Ward 4 of .448 X 2451, or 1098. It then speculates that the white voting age population of Ward 4 is 1133 — (1133 X.045), or 1082 (see n. 4). It makes this estimation even though it acknowledges that Mr. Armstrong’s testimony provides “us with little more than a basis for speculation regarding the present black voting age population of Ward 4.” The “need” for making that estimation stems from the majority’s summary dismissal of the 1980 census data for the city of Oxford as a whole which was submitted at trial:
*351 From this census data, we may compute the following table:
[[Image here]]
Citywide in Oxford, blacks comprise a lower proportion of the total population of voting age (16.36%) than they do of the total population of all ages (21.23%). Put differently, the proportion of the total population of all ages whieh blacks comprise in Oxford overestimates the proportion of the total population of voting age which blacks comprise in Oxford. Concomitantly, the proportion of the total population of all ages which whites comprise in Oxford (77.26%) underestimates the proportion of the total population of voting age which whites comprise in Oxford (82.04%),
On the modest and plausible assumption that the percentage of each racial group which is of voting age (C6 of Table 1) does not vary appreciably from place to place within the city of Oxford, then it is also true that (i) the proportion of the total population of all ages which blacks comprise in any given electoral district in Oxford overestimates the proportion of the total population of voting age which blacks comprise in that same given electoral district in Oxford, and (ii) the proportion of the total population of all ages which whites comprise in any given electoral district in Oxford underestimates the proportion of the total population of voting age which whites comprise in that same given electoral district in Oxford. If blacks in Oxford comprise x% and whites comprise (100-x)% of the total population of a district, then blacks would comprise less than x% of the voting age population of that district and whites would comprise more than (100-x)% of the voting age population of the district. This proposition becomes more readily apparent when a portion of the same data — for Ward 4 of the City of Oxford — is presented below in a format similar to that which the Jordan trial court used in the table which appears in its opinion: 6
[[Image here]]
*352 The so-called black majority ward, Ward 4, turns out on the basis of this data not to be a black majority ward at all.
As the majority points out, in language which closely resembles that with which I criticized a now-defunct footnote to the original proposed majority opinion: the cumulative effect of individual births, deaths, attainments of majority, and migrations in and out of Oxford since 1980 cannot reliably be ascertained at present. The majority urges on that basis that 1980 census data is too “out-of-date” for us to draw the conclusion that Oxford’s 4-and-l scheme cannot stand.
The majority also states, however, that “[a] hallmark of our system of government is that a rival candidate need only wait one term to put together a different coalition [of voters] if the elected representative proves to be unresponsive to any group of constituents.” But the democratic guarantee into which the majority implicitly admonishes Houston and others to put their faith has little value if they must wait up to ten years for “fresh” census data with which to seek a redress of the failure of that guarantee.
No doubt the underlying reality which the 1980 census data in the record measured has been altered in some way by the course of human events in Oxford these eight years past. Just what that alteration has been will not be known until the 1990 census results become available. Of course, even that data will not be exact. The knowledge that our data is to some unknown extent out-of-date is not particularly troubling, however. In any measurement system — no matter how sophisticated or finely-tuned — there is always some degree of inherent inaccuracy. When such a measurement system is applied to a dynamic [situation 2], additional inaccuracy is introduced: the data is to some extent “out-of-date” from the instant the measurement is completed and reported. But the knowledge that a yardstick is not precisely one yard long and that a child grows with every passing moment does not preclude using that yardstick to record his growth by marks upon the frame of the kitchen door from time to time. The results of such a measurement are reported to faraway relatives perhaps some weeks or months after the measurement is made, and are held forth as — and accepted as — being reasonably accurate, even though all concerned know the child surely has grown another inch or two in the interim. This knowledge does not make a new measurement mandatory (save, perhaps, in the reckoning of our hypothetical little boy or girl). A new measurement can be and is made only so often as is practical and useful. Consistent with this principle, indulgent parents might even consent to measure their child’s growth weekly or even daily, but the same principle constrains the national census to make its measurements decennially. 7 Fortunately, however, the national census need not instantaneously take account of each growing child's attainment of majority in order to serve as a reliable guide for judicial decision. Just as inaccuracies in the measuring system do not deter us from recording a child’s growth in the fashion described, they need not and should not preclude our reasonable reliance on 1980 census data in 1988.
Within the acceptable limitations of the best available data, it is apparent that in Ward 4 under the 4-and-l plan, black voters remain outnumbered.
*353 As set forth above, in reaching this conclusion I have assumed that the percentage of each racial group which is of voting age (C6 of Table 1) does not vary appreciably from place to place within the city of Oxford. This assumption is no greater or riskier than the expert’s assumption — implicitly accepted by the majority — that the difference between the black voting age population and the total black population in Oxford is the same as the statewide average difference. And certainly it is far more conservative than the assumption that the percentage of the total population of the ward which is black is an adequate and reliable estimate of the percentage of the voting population of the ward which is black.
It bears reiterating that there is no magic in either my figures or those employed by the majority. Jordan mandates no specific percentage of black voting age population in a district, but rather mandates a percentage “sufficient to overcome the effects of past discrimination.” 603 F.Supp. at 814. The determination of what percentage is sufficient is to be made on a case-by-case basis. The divergence between my calculations and those of the majority serves primarily to reaffirm the need for caution before concluding on the basis of one set of figures that all is well — no more and no less. But that reaffirmation is loud, clear, unmistakable, and commanding.
Stopping the Buck
In deciding whether relief is warranted, federal courts are not confined to the “seven factors” enumerated in Thornburg. Section 2 clearly states that “totality of the circumstances” is the test. Given, however, that “Mississippi has a long history of de jure and de facto race discrimination, that racial bloc voting is common in Mississippi, and that political processes have not been equally open to blacks,” 8 we should look quite literally and expansively to the totality of the circumstances.
In addition to the seven enumerated factors which guide the majority’s analysis, the majority considers an eighth factor: “whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group.” 9 This factor appears in the Senate Report on the 1982 amendments as one of two “[additional factors that in some cases have had probative value as part of plaintiffs’ evidence to establish a [§ 2] violation.” 10 The majority bypasses without comment the other member of that pair, namely “whether the policy underlying the state or political subdivision’s use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.”
This court has previously characterized this factor as “of ... diminished importance ... under the results test” and expressed “doubt that the tenuousness factor has any probative value for evaluating the ‘fairness’ of the electoral system’s impact.” Jones v. City of Lubbock, 727 F.2d 364, 383 (5th Cir.1984). This factor does retain some of its vitality even under the results test, however. S.Rep. No. 417 at 29 & n. 117, 1982 U.S.Code Cong. & Ad.News at 207 & n. 117. The District Court certainly considered this factor — apparently on an equal basis with the seven enumerated factors and the responsiveness factor. It began by acknowledging that
[t]he strength of a given political jurisdiction’s policy in favor of a particular electoral scheme is admittedly a more rele *354 vant consideration under the intent standard than under the results standard of section 2. See [United States v.] Marengo County [Commission], 731 F.2d [1546,] 1571 [cert. denied, 469 U.S. 976 , 105 S.Ct. 375 , 83 L.Ed.2d 311 (1984)]. However, the strength of the underlying policy is still relevant because a finding of discriminatory intent may provide circumstantial evidence that the particular electoral device produces discriminatory results. Furthermore, a tenuous justification for a policy may also indicate that the policy itself is unfair.
663 F.Supp. at 355-56 (emphasis added). The District Court then went on to conclude that “the policy of electing aldermen by wards is not tenuous [because it] is mandated by state statute” — namely, Miss. Code Ann. § 3374-36 (1942). 663 F.Supp. at 356. But the court need not — nor should it — restrict itself to looking to the policies underlying the decision of the responsible city authorities to implement a 4-and-l system, but instead may also look to the policies underlying the decision of the Mississippi legislature to mandate that the city authorities decide in certain specified circumstances to implement a 4-and-l system. While the city authorities may have had a justification for their actions that certainly could not be described as “tenuous,” this does not resolve the question whether the legislature had an adequate justification for its action. We must foreclose the possibility that local and state authorities could evade the review of the justification for an at-large system that the “tenuousness” factor requires — even under the results test — simply by shifting ultimate responsibility for the existence of such a system from the local to the state level.
The conflicting sets of statistics here leave room for reasonable observers to disagree as to which set accurately portrays reality in present-day Oxford. Given the history of the area at issue here, however, the court should be conservative on the side of making certain that black access to political participation is protected and nurtured. This does not mean that an entitlement to a black election winner exists or that we need find ourselves sliding uncontrollably down the slippery slope which the majority forbodes. It means simply that where, as here, (i) there are no equities that favor the present system over the proposed alternative (ii) the present system much less-reliably overcomes the effects of past discrimination than the proposed alternative, and (iii) the administrative burden of implementing the proposed alternative is minimal, then there is no reason for judicial reluctance to mandate — in furtherance of Congressional intent — that , the present electoral system be supplanted with that superior alternative.
I must, therefore, respectfully dissent.
. See 604 F.Supp. at 814 or 819.
. Id.
. The 58.30% figure from Jordan is the one that the District Court in this case used. See 663 F.Supp. at 348 n. 3.
. Armstrong further elaborated on his qualifications. He holds a B.S. degree from Jackson State University and has also taken graduate courses in public administration at that same institution. He is a member of the Public Administration Society of America and is also certified by the District Courts for the Northern and Southern Districts of Mississippi as an expert witness to provide and interpret census and other demographic data for Mississippi.
. Standing alone, that statement is ambiguous. It could have either of two meanings:
[[Image here]]
. See 604 F.Supp. at 814 or 819. The headings used here are slightly different than those in the Jordan table. Columns Cl through C6 of my Tables 2A and 2B correspond to columns 2, 4, 5, 6, 7 and 8 of the Jordan table.
. If it were true that one set of census data was intolerably inaccurate before the next census, then it seems likely that our decennial census would instead take place more frequently — perhaps every five years, for instance. Indeed, some census measurements do take place more frequently than decennially, now that census data is used for many purposes, public and private, that go far beyond the allocation of seats in the House of Representatives. That allocation continues to be made decennially, however. The 1980 census data will continue to form the basis of that allocation until the 1990 census data becomes available, in 1992 or so. This is so despite the fact — which is common knowledge — that population shifts since 1980 have increased the population of the South and West relative to the Northeast. If decennially-gathered data is nevertheless sufficiently accurate for allocating House seats, it surely should be sufficiently accurate for our present purposes in this case.
. Mississippi Republican Executive Committee v. Brooks, 469 U.S. 1002, 1004-05 , 105 S.Ct. 416, 417 , 83 L.Ed.2d 343, 344 (1984).
. S.Rep. No. 97-417, 97th Cong., 2d Sess. 29 (1982), reprinted in 1982 U.S.Code Cong. & Ad. News 177, 207.
. Id. I observe, however, that the "Senate Report expressly states that ‘the election of a few minority candidates does not "necessarily foreclose the possibility of dilution of the black vóte,” ’ noting that if it did ‘the possibility exists that the majority citizens might evade [§ 2] by manipulating the election of a "safe" minority candidate.'" 478 U.S. 30, 75 , 106 S.Ct. 2752, 2779 , 92 L.Ed.2d 25, 62 (emphasis added). The possibility of evasion of § 2 through just enough grudging "responsiveness” by white incumbents also exists. Therefore, this "responsiveness” factor should be discounted appropriately. Section 2 guarantees blacks more than just crumbs from the whites’ political table.

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/cases/9478253. Public record. Not legal advice.