160 Comments

I think I speak for a lot of us when I say that I appreciate your intellectual integrity.

Expand full comment

Ronald Reagan was absolutely right about "Star Wars." 1) He said the actual tech wouldn't be ready for 20 years (in fact, it was ready much earlier); 2) Early systems, while not exceptionally effective in 1991, were effective enough; 3) The media, calling it "star wars" did Reagan a huge favor by making the technologically difficult understandable; 4) Gorby knew it would work. He knew if the US could build 11,000 airplanes in just four years of war, plus an a-bomb, we could do anything. It literally brought him to the table.

Expand full comment

Great piece, Noah. Two things can be true at once:

1) The procurement process is engineered for risk reduction, leading to long delays, low quantities and cost overruns.

2) The things that survive that procurement process are really damn good.

(Also, your point about the defense establishment not being part of the public conversation is 100% correct. "Absence of evidence is not evidence of absence," etc.)

Expand full comment

Behold! It takes intelligence and self-respect to admit that your opinion has changed, and your previous opinion was naive. Well done.

Expand full comment
Apr 15Liked by Noah Smith

Excellent again. Star Wars was more political in the 80s as technology wasn't ready. With an SB Physics geek and nerd schools early 70s, I've worked in the highest level of non classified optical systems..ie with the smart people (not me).

I also built my own gaming PC in 2002 with nVidia board. Ha. That destroyed mainframes and Sun and Apollo workstations at complex optical calculations.

So now lets think about missile interception. Again, I'm a non attorney no security clearance spokesman.

A. Upon signal enemy launch - counter launch

B. 1st moments need fast large angle analysis. Not precision yet. You get a target cross-section and head for it.

C. Mid range, reduce angle window, optimize TOF (Time of flight).

D. Closure, asymptotically tighten angle space, increase computational cycles.

E. As angle space closes, there are smaller and smaller delta iterations.

1. Ultra fast processors

2. Ultra efficient tight algorithms

3. Sensor plane array technology - cooled for IR, but now mutlispectral.

4. Optical lens design, ie off axis, optimize speed of lenses, ie below f1.2 and lower. Wide angle at the acquisition level.

5. Then tighter FOV (field of view) with precision.

That nice alumni really didn't understand the evolution of the key tech. It didn't exist in the 80s and 90s.

Two proofs we all can see

A. Gaming computers from 1984 to 2024

B. Film cameras of 84 cant compare to your cellphone camera and its sensor today. As an analogy

Expand full comment

I think you've reached the wrong conclusions from the aerial battle between Iran and Israel. As an Israeli I'm very much pro the side which won this round, but I still think the engagement proved that air defence *doesn't* work.

Because:

1. The attack was telegraphed, and deployed against one of the best air defence forces in the world, backed up by an alliance that gave it access to as much air space, military intelligence and fire support it needed. Even so, seven missiles got through. Had this been a nuclear attack Israel would have been wiped off the map - and since nuclear strikes tend to be an all or nothing kind of deal, Iran would have committed more munitions in such a case.

2. Air defence is so much more expensive than air attacks that it can only ever possibly work as a means to buy some time against conventional strikes whole your own army strikes back and neutralizes the enemy to prevent further attacks.

3. In fact, Iran made a limited strike because it does not have faith in its own air defences being capable of stopping Israeli strikes. Israel is similarly unlikely to start an outright war because it does not think it can stop an Iranian attack for long.

So an air war is super scary for both sides, but if it happens, you have to make sure you're the aggressor - being on the defence is a losing game, and the strategic role of air defence is to buy (a little!) time to organize and offence. It's not impossible to hit a rocket with a rocket, it's just impossible to build a system that works at scale and over time.

Expand full comment
Apr 15·edited Apr 15Liked by Noah Smith

I was a young officer in the Air Force during the Obama era. I did not fly, my career field (CE) was heavily stressed in deployment cycles. When I got back from a tour to the Middle East, I remember being frustrated when a colonel, in my performance review, said exasperated, "nobody CARES what you did there. How did you support our [Cold War] mission HERE?"

Sort of a jerk thing to say - but damn he was right. Still stings, definitely I consider my deployed time to be my "real" service and the home station time to be my "training." But I'm wrong. On a deep level, I misunderstood our posture and threats. And deep down, I still kind of do.

Which relates to a fair criticism of why we'd fight a war that we didn't commit to because it wasn't supported by the public and continued another war that wasn't taken seriously enough. But however you feel about how those wars were conducted, it's frustrating for me but also sort of comforting to know that the Big Blue Machine was rolling in the right direction, for the real threats.

Expand full comment
Apr 15Liked by Noah Smith

I love this piece. I don't think it is simply the result of articulate argumentation, either. I believe a good part of what I perceive in it, and the rest of your writing generally, is a mature, ethical mind doing its best to steer humanity towards the best possible outcome. I appreciate your work.

I do wonder if the expert consensus about ICBMs are still correct. No doubt they move extremely fast, but they also, by nature, travel far. That gives our computers more time to calculate their trajectories and destinations.

It is possible that all the nukes in the world will be launched at once, but I suspect that's unlikely. This past weekend hundreds of Iran's missiles and drones were shot down. I expect the US and Europe would face a similar number of nuclear warheads and dummies. I doubt there would be coordination between Russia and China so long as both are not attacked simultaneously. And I doubt either would be attacked by allied forces first, despite the US's utterly deplorable action in Iraq twenty years ago.

Expand full comment
Apr 15·edited Apr 15

Much weapon system criticism comes from people opposed on moral or emotional grounds to everything military in general and Western power in particular.

Ballistic missile defenses would allow the US to fight even well-armed smaller powers with immunity and therefore, in the eyes of some, increase the likeliness of military intervention and American power abuse.

Take any new weapon system, and you will find a think tank claiming that it is useless, but would be destabilising if developed (despite being allegedly useless).

Expand full comment

Just a few comments.

The US also has a ship born anti-missile system called AEGIS (cruisers and currently destroyers) which uses the Standard Missile (https://en.wikipedia.org/wiki/Standard_Missile) to intercept a variety of threats. This system has been in continuous development since the 1980s.

The US DoD often pushes the technical limits when acquiring new weapons systems (e.g. F-35 and USS Ford class aircraft carriers). Sometimes the systems need to get bugs workedout. The military generally updates (fixes) these systems over the years of production. This is why you will see many systems listed as F-18A, F-18D and F-18E or So-and-so-system Block N (where N is a number). Basically the systems get better over time.

Last, I dislike attacking DoD systems for waste, but the DoD budgeting system IS messed up. Activities (think like NAVAIR or NAVSEA) get funded based on what they spent in the previous year, so there is motivation for them to spend every dollar they were budgeted to get the same budget or more the next year (called Use or Lose). At the end of each fiscal year they have a feeding frenzy, to get all budgeted funds spent. It has been this way for many decades and can result in less than optimal sizable expenditures.

Expand full comment
Apr 16Liked by Noah Smith

Good post man. DoD spending is done by professionals, and they know their "must have" and "nice to have if it works" projects. MDA, B21s, and Fords are must haves, so they spend whatevs and let Congress fight over the others. Zumwalts, littoral ships, and F22 were "nice to have". Must have projects sometime turn into revenue as long as we keep the good ones...like the F35, M1 Abrams, Patriot, and Aegis. Looks like Virginia might be there too now thanks to AUKUS. Note that we're not shopping out Ohios to anyone...

And youre right. Detection was always the key. That was on the "must have" menu, which is the one that doesnt list prices...

Keep up the good work bud.

Expand full comment

This is superb. Now do renewables and nuclear power.

Expand full comment

Noah, thank you for highlighting some of the things we're doing right. Particularly today, we need to remember that, as screwed up as we are in some ways politically and socially, some of our institutions actually still function pretty well.

Expand full comment
Apr 16Liked by Noah Smith

This is the last place I expected to see a LazerPig video. +1

Expand full comment

My understanding is that 3 times as many interceptors are fired as there are incoming missiles or drones. Their price is in the millions each vs a much lower cost of what they are shooting down - hundreds of thousands perhaps. It works fantastically but could it be done, day after day, for any extended amount of time? How many could possibly be on each ship? I am looking for some context here. Could $1 billion been spent in one night? Please tell me what I do not get.

Expand full comment

To note that this was a special case. Iran's goal was specifically to make a show of attacking Israel, but NOT hit anything important that would escalate the situation. I'm not sure how far they went - probably used the opportunity to clear a lot of old stocks instead of using the latest and best, and may have touched the scales in other ways as well. But either way, I'd be wary of updating too much from an incident which ended up exactly the way all the parties wanted it to end up.

Expand full comment