In 1956, Arlington County held a referendum that a local newspaper at the time called “controversial.” The question was whether future bond referendums should be limited to property owners.
By a vote of 17,681 to 15,1950, Arlington voters said “yes.”
Seventy years later, that long-forgotten referendum — and the legal cases it set in motion — may matter once again in Virginia.
That’s because the Virginia Supreme Court threw out the results, ruling that the very question people were voting on was unconstitutional. This case — Carlisle v. Hassan — serves as a precedent if the Virginia Supreme Court decides later this spring that the results of the April 21 referendum on redistricting don’t count.
The “yes” side of that question so far has spent more than $40 million trying to persuade Virginia voters to approve the proposed constitutional amendment to allow this unusual mid-decade redrawing of congressional lines — re-redistricting, as some call it, since we’ve already had one redistricting off the same census numbers.
What little polling is available suggests that the “yes” side has the advantage, although not in a decisive way. The early voting numbers also show that more people are voting so far in this special election than in last year’s gubernatorial election, when turnout was on the high side for a state election.
None of this may matter, though, because the seven most important votes in this election won’t be cast until well after April 21: Those are the votes of the seven Virginia Supreme court justices.
You’ll recall that before early voting began, there were multiple legal challenges to the referendum. Two made their way to the Virginia Supreme Court (via Tazewell County). The court said it would hear those cases after the election. That may seem odd to those outside the legal profession — shouldn’t the court rule beforehand? — but the court’s decision rests squarely on precedent.
In 1912, a Lynchburg man felt that a particular constitutional amendment being put before voters that year was procedurally flawed. He sued to stop the referendum. The Virginia Supreme Court ruled that it would not do so, likening the process of approving an amendment to legislating. Courts don’t step in to rule a bill unconstitutional when the General Assembly sends it to the governor; they wait until the bill is law and someone challenges it. In the case of a referendum, voters are essentially acting as the governor in deciding whether to approve or reject a measure — therefore, the proper time to challenge the constitutionality of the vote is after it has occurred, the court found. You can read more about that case, known as Scott v. James, here. In that case, the referendum passed but James Scott apparently never challenged the outcome.
The Virginia Supreme Court, in its order letting the April 21 vote go forward, cited Scott v. James as the rationale, but that raises a question: Has this or any other court ever invalidated an election for something other than fraud or other irregularities? Thanks to Rebecca Green, co-director of the Election Law Center at the College of William & Mary, I can report that yes, there are such examples.
None are quite on point with the current set of facts, but they all serve as precedent that, yes, courts can and have thrown out election results.
The only Virginia example may be that Carlisle v. Hassan case over the 1956 Arlington referendum. Here’s the background, pieced together from court rulings, government documents and old newspaper clippings from the now-defunct Daily Sun newspaper in Arlington.
In 1956, the General Assembly passed a bill that allowed counties with a county manager form of government (at the time this may only have been Arlington and Henrico) to hold referendums on whether to restrict bond referendums to “freeholders,” the legal term for property owners.
That November, Arlington held such a referendum, which passed.
The following year, Arlington began making plans for a major infrastructure package — it envisioned eight separate bond questions on the ballot in May 1958, from libraries to schools, sewers to parks, sidewalks to streets, county buildings to storm drainage. In preparation for that vote, the county electoral board hired Patricia Lowe to figure out which voters were property owners. She spent three days doing so in November 1957 and submitted an invoice for her services, which the electoral board dutifully sent to the commonwealth’s attorney to certify the payment. It’s unclear why the prosecutor had authority for this but he may have also been serving as county attorney. What matters is that William Hassan refused to sign off on the invoice because he said the referendum-approved provision restricting the vote to property owners was unconstitutional. The electoral board sued.
On March 10, 1958, the Virginia Supreme Court ruled that Hassan was right — Arlington voters had approved an unconstitutional law and the provision was struck down, just as courts from time to time strike down other laws as unconstitutional.
This may be the only time where the Virginia Supreme Court has overturned an election result (absent fraud and such), but it does reinforce the precedent under the 1912 Scott v. James case that the court can do so — and sets a precedent for actually doing so.
Meanwhile, there’s at least one other instance of the Virginia Supreme Court taking up a case dealing with an election result. In that case, the court upheld the results but further established the precedent that it has the final say.
That case — Harrison v. Day — came just a year after the Arlington decision. In 1958, Virginia voters approved a constitutional amendment that allowed counties to borrow money for school construction from the Supplemental Retirement System without going to a referendum. When Rockbridge County applied for such a loan to build a high school in the northern part of the county, the state comptroller refused to approve the paperwork. Sidney Day Jr. contended that the amendment had not been properly approved because the governor hadn’t included the full text of the amendment in his official proclamation that the amendment had been approved. Attorney General (and future governor) Albertis Harrison sued the comptroller to seek the release of the Rockbridge funds. The Virginia Supreme Court ruled in November 1959 that the governor’s proclamation was fine, that the amendment had been properly approved and therefore Rockbridge was entitled to its money. The details of this case seem quite technical, but the principle remains: The court felt within its power to rule the amendment had been improperly adopted had the facts led there; they just didn’t.
There are also cases of courts in other states invalidating results. In 1996, Nebraska voters approved a state constitutional amendment that instructed members of the state’s congressional delegation to “use all of his or her delegated power” to seek a federal constitutional amendment imposing term limits. As part of that state amendment, Nebraska voters required future ballots to say “DISREGARDED VOTERS [sic] INSTRUCTIONS ON TERM LIMITS” beside the names of any U.S. House members who didn’t pursue term limits. A federal appeals court later struck that down as a violation of the U.S. Constitution and invalidated four sections of the Nebraska constitution.
California provides another example: In 2008, voters approved Proposition 8, an amendment to the state constitution that banned same-sex marriage. In 2010, a federal judge ruled the amendment a violation of the U.S. Constitution’s 14th Amendment guarantees of due process and equal protection, setting off more legal action that went on for years.
I haven’t been able to find any cases that exactly match the facts involved in the current cases challenging the Virginia referendum, but I suppose if there were, then we wouldn’t need the Supreme Court to weigh in.
The two strongest legal arguments to invalidate the vote involve language and timing. Republicans argue that the language voters will encounter on the ballot is misleading because it says this is a vote to “restore fairness,” which is very much the Democratic view but not the Republican one. The state constitution also requires that the legislature pass an amendment twice, with an election in between. Democrats say they’ve done so, because they first passed the redistricting amendment in a special session last October, before the November vote for governor. Republicans say that because early voting was already underway, there has been no intervening election. Is election day a specific day or has early voting stretched it into an election season?
Those are some of the questions the Virginia Supreme Court will take up. I have no idea how the court will rule, but the Virginia Supreme Court has four times now (1912, 1958, 1959 and 2026) made clear that it has the power to invalidate an election if either the law and constitution isn’t followed. The fact that it’s only done so one time doesn’t mean there won’t be a second.
For more on redistricting (and other politics)
I talk about redistricting in our latest Cardinal News podcast; find it here or wherever you get your podcasts.
I’ll give an update on the early voting trends (and whatever other political news happens) in West of the Capital, our weekly political newsletter that goes out on Friday afternoons. Sign up here:
