Planets continue to tangle, inconveniently close to the Sun.
See the end note about enlarging illustrations. Arrows through the moving bodies show their movement (against the starry background) from 2 days earlier to 2 days later. The Moon is exaggerated 4 times in size.
Around April 11, the involved planets were Mercury, Mars, Saturn, and Neptune, and they were on the morning side of the Sun. This time, the grouping has formed on the evening side. Bright Venus and Jupiter are high and easily seen, but those involved in the tangle are Mercury, Uranus, and the extremely slender crescent Moon. (Remember that the Moon and Sun were members of the ancient set of planêtai, “wandering stars.”)
The cluster of planets is even more tantalizingly low, going below the horizon only minutes after the Sun does, but perhaps it is of even more interest for the “mind’s eye.” The motions of these “stars” at their greatly different distances cause an extraordinary number of events to be listed for the single day (by Universal Time, UT) of May 17 on [age 38 in Astronomical Calendar 2026, where more details are given about each event. (A “trio” is a gathering of three bodies tight enough to fit within a 5-degree-wide circle. “Moon” and “Pleiades” refers to their centers.)
1 UT trio of Moon, Mercury, and the Pleiades
1 trio of Mercury, Uranus, and the Pleiades
2 Moon 4.4° N of Mercury
4 Moon 1.00° N of Pleiades
4 trio of Moon, Uranus, and the Pleiades
4 trio of Moon, Mercury, and Uranus
5 Moon 5.2° N of Uranu
14 Moon at perigee; only 17.7 hours after new Moon
18 Mercury 3.4° SE of the Pleiades
Intelligent life on the Third Planet
Yet again, an election shows the need for proportional representation (as advocated by many groups, such as the Electoral Reform Society) and for approval voting (as advocated by me and by the Center for Electoral Science),
Approval voting is about how to count votes; proportional representation is about how to assign seats on the basis of the counted votes.
Proportional representation is simpler to see. Under the usual system, if Party A were to get 51 percent of the votes in all of the constituencies in a country, and Party B 49 percent, then A would get 100 percent of the seats and B zero. In countries that use PR by various arrangements, those parties would get 51 and 49 percent of the seats.
In the May 7 British local elections, there was the usual crop of unfair results, as pointed out by the group Make Votes Matter. For instance in one city, Wigan, “Reform UK” (the party of extreme anti-immigrant xenophobia) got 46% of the votes and 96% of the seats in council.
I found myself looking at a ballot with a list of 13 candidates: 3 each for Green, Liberal Democrat, Labour, “Local Conservative”; and one for “Reform.”
With instruction: Vote for no more than 3 candidates.
This is not quite the same as the common “vote for one only,” but it is similar in effect.
There is no way, short of giving all your votes to one of the candidates you approve, to avoid weakening them all and making victory more likely for one of those you disapprove. So you have to choose between supporting the one you like most and the one who has more chance of winning – either way, weakening the other.
The instruction should be: Vote for as many as you like. It is impossible for this to work out unfairly.
__________
This weblog maintains its right to be about astronomy or anything under the sun.
ILLUSTRATIONS in these posts are made with precision but have to be inserted in another format. You may be able to enlarge them on your monitor.
One way: right-click, and choose ”View image” or ”Open image in new tab”, then enlarge. Or choose ”Copy image”, then put it on your desktop, then open it. On an iPad or phone, use the finger gesture that enlarges (spreading with two fingers, or tapping and dragging with three fingers). Other methods have been suggested, such as dragging the image to the desktop and opening it in other ways.
Sometimes I make improvements or corrections to a post after publishing it. If you click on the title, rather than on ‘Read more’, I think you are sure to see the latest version. Or, if you click ‘Refresh’ or press function key 5, you’ll see the version change to the latest.
These days I find myself a little lazy (perhaps falling into a trap?) in utilizing AI as a companion tool in my arsenal to provide research and production. I’ve long known Guy’s aim in promoting approval voting as the “better way” to achieving sensible political and crowd-sourced policy. So I asked Claude.ai about this posting. I list it here in its raw form:
The prompt to Claude.ai (Opus 4.7):
This is an observation by a friend in the UK about voting methods. Examine his recommendation and how it could benefit US voting.
Response:
Your friend is making a case for two electoral reforms—approval voting and proportional representation—using a recent UK local election as his motivating example. Let me focus on his approval voting recommendation, since that’s the one he’s personally championing, and consider how it could benefit US elections.
His core argument is elegant: when you’re told “vote for no more than 3” (or the more common “vote for one only”), you’re forced into a strategic dilemma. If you genuinely like several candidates, splitting your support weakens all of them. So you end up either betraying your true preferences to back the most viable candidate, or sticking with your favorite and helping someone you dislike win. Approval voting—”vote for as many as you like”—eliminates this trap entirely. You simply mark every candidate you’d find acceptable, and the one with the most approvals wins.
For US elections, this addresses several persistent pathologies.
The spoiler effect is the most obvious one. American voters routinely face the Nader/Stein/Perot dilemma: support a third-party candidate you actually prefer, or vote “strategically” for a major-party candidate to avoid throwing the election to your least-favored option. Under approval voting, you could approve both your favorite and your acceptable backup without penalty. This would meaningfully lower the barrier to entry for third parties and independent candidates, since voters wouldn’t have to “waste” a vote to support them.
Primary elections are where approval voting might shine brightest in the US context. Crowded primary fields—think the 2016 Republican primary with 17 candidates, or the 2020 Democratic primary—often produce winners with small pluralities who may not actually represent the broadest support within the party. A candidate who is everyone’s second choice but nobody’s first currently loses badly; under approval voting, that broadly acceptable candidate might win, which arguably produces nominees with wider general-election appeal.
Approval voting also tends to reward candidates who build coalitions rather than ones who energize a narrow base by attacking opponents. If you need to be acceptable to as many voters as possible, scorched-earth campaigning becomes counterproductive—you’re trying to be on lots of ballots, not just dominate a faction.
It’s also remarkably simple to implement compared to ranked-choice voting (which the US has been adopting in places like Maine, Alaska, and New York City). Approval voting requires no new ballot design beyond removing the “vote for one” restriction, no rounds of elimination, no complicated tabulation. Voters already understand checkboxes. Fargo, North Dakota adopted it in 2018 and St. Louis adopted a variant (approval voting in the primary, top two to the general) in 2020—both jurisdictions have run elections without the confusion that sometimes accompanies ranked-choice rollouts.
A few honest counterpoints worth mentioning: approval voting can encourage “bullet voting” (approving only your favorite) if voters fear that approving a second choice hurts their first; it doesn’t capture intensity of preference (a candidate you love and one you merely tolerate count equally); and it can produce winners who are broadly acceptable but no one’s actual favorite, which some argue is a feature and others a bug. The Center for Election Science, which your friend links, generally argues these concerns are outweighed by the benefits, while ranked-choice advocates like FairVote disagree.
On proportional representation, the US application would be much more limited—most American elections fill single offices (president, governor, senator, mayor), where PR isn’t really applicable. It would matter most for state legislatures and the House of Representatives, where multi-member districts could replace the current single-member system. That’s a far heavier lift constitutionally and politically than changing how ballots are counted.
The cleanest takeaway from your friend’s piece: approval voting is the rare reform that’s both theoretically sound and practically easy to adopt. It doesn’t require constitutional amendments, new voting machines, or voter re-education campaigns. It just requires removing two words from the instructions on the ballot.
Your example of a party winning 51% of the popular vote and 100% of the seats in a legislature is not the worst possible outcome.
Here in the United States, in the run-up to November’s congressional elections, many state legislatures are redrawing congressional district maps to give the greatest possible number of seats to the political party that controls each state legislature. Both Republican and Democratic legislatures are playing this game, but since the Republicans have less popular support than the Democrats, the Republicans stand to benefit more. It is possible that the Republican party could win significantly less than half of the national popular vote but still hold majorities in both the House of Representatives and the Senate. If this happens the legitimacy of our government will be questioned.
If a state used proportional representation, it would act as a single region and assigns seats to party candidate lists proportionally to the popular vote. This is what is already done in many countries.
At present I don’t see any realistic path toward proportional representation, approval voting, or any other structural electoral reform in the United States. I think we just have to try to save the democratic institutions we already have, imperfect though they may be. Here in California the pro-Democratic party gerrymander map was at least put on the ballot for everybody to vote on (Republican state legislatures have pushed through their gerrymanders with rushed legislative votes). I voted against the California gerrymander on principle, but I’m glad I didn’t have to personally decide its fate. It passed with a large majority of votes. A Democratic majority in the House of Representative would be better for our nation’s long-term prospects than continued Republican hegemony, even if winning those seats is a messy process. The California initiative said we will reinstate our non-partisan redistricting commission after the 2030 census. Hopefully by then we will have started cleaning up the huge mess left behind by Trump and his fascist minions.