Epistemology, Kant, and Rand

We will begin with an excerpt from Ayn Rand, ‘For the New Intellectual,’ on Kant.

The man who formalized this state, and closed the door of philosophy to reason, was Immanuel Kant. Kant gave metaphysical expression to the psycho-epistemology of Attila and the Witch Doctor and to their primordial existential relationship, shutting out of his universe the existence and the psycho-epistemology of the Producer. He surrendered philosophy to Attila—and insured its future delivery back into the power of the Witch Doctor. He turned the world over to Attila, but reserved to the Witch Doctor the realm of morality. Kant’s expressly stated purpose was to save the morality of self-abnegation and self-sacrifice. He knew that it could not survive without a mystic base—and what it had to be saved from was reason. Attila’s share of Kant’s universe includes this earth, physical reality, man’s senses, perceptions, reason and science, all of it labeled the “phenomenal” world. The Witch Doctor’s share is another, “higher,” reality, labeled the “noumenal” world, and a special manifestation, labeled the “categorical imperative,” which dictates to man the rules of morality and which makes itself known by means of a feeling, as a special sense of duty. The “phenomenal” world, said Kant, is not real: reality, as perceived by man’s mind, is a distortion. The distorting mechanism is man’s conceptual faculty: man’s basic concepts (such as time, space, existence) are not derived from experience or reality, but come from an automatic system of filters in his consciousness (labeled “categories” and “forms of perception”) which impose their own design on his perception of the external world and make him incapable of perceiving it in any manner other than the one in which he does perceive it. This proves, said Kant, that man’s concepts are only a delusion, but a collective delusion which no one has the power to escape. Thus reason and science are “limited,” said Kant; they are valid only so long as they deal with this world, with a permanent, pre-determined collective delusion (and thus the criterion of reason’s validity was switched from the objective to the collective), but they are impotent to deal with the fundamental, metaphysical issues of existence, which belong to the “noumenal” world. The “noumenal” world is unknowable; it is the world of “real” reality, “superior” truth and “things in themselves” or “things as they are”—which means: things as they are not perceived by man. Even apart from the fact that Kant’s theory of the “categories” as the source of man’s concepts was a preposterous invention, his argument amounted to a negation, not only of man’s consciousness, but of any consciousness, of consciousness as such. His argument, in essence, ran as follows: man is limited to a consciousness of a specific nature, which perceives by specific means and no others, therefore, his consciousness is not valid; man is blind, because he has eyes—deaf, because he has ears—deluded, because he has a mind—and the things he perceives do not exist, because he perceives them. As to Kant’s version of morality, it was appropriate to the kind of zombies that would inhabit that kind of universe: it consisted of total, abject selflessness. An action is moral, said Kant, only if one has no desire to perform it, but performs it out of a sense of duty and derives no benefit from it of any sort, neither material nor spiritual; a benefit destroys the moral value of an action. (Thus, if one has no desire to be evil, one cannot be good; if one has, one can.) Those who accept any part of Kant’s philosophy—metaphysical, epistemological or moral—deserve it.

If one objects to Rand’s summary, here is one from a Kantian, though I quote Rand’s quote of him (Henry Mansel) from ‘Introduction to Objectivist Epistemology:’

With him [Kant] all is phenomenal [mere appearance] which is relative, and all is relative which is an object to a conscious subject. The conceptions of the understanding as much depend on the constitution of our thinking faculties, as the perception of the senses do on the constitution of our intuitive faculties. Both might be different, were our mental constitution changed; both probably are different to beings differently constituted. The real thus becomes identical with the absolute, with the object as it is in itself, out of all relation to a subject; and, as all consciousness is a relation between subject and object, it follows that to attain a knowledge of the real we must go out of consciousness.

The error of Kant was this: he wished to justify synthetic judgments. But he did so wrongly. He concluded that synthetic judgment was justified because the nature of human consciousness forces anything of which we are conscious to obey our synthetic judgments.

Now, first, he vastly overestimated the prevalence of synthetic judgments. For example, arithmetic is not synthetic, as he claimed, but analytic (as shown by Frege and Peano; see any textbook on Peano arithmetic or cardinal arithmetic). His insistence that Euclidean geometry is a synthetic judgment to which the human mind would force all phenomenal experience to conform is just wrong (we in fact have theories which do not involve Euclidean geometry except as an approximation; general relativity shows us that the geometry of the world may be curved).

But second, Kant could not guarantee that human consciousness would remain the same, except by an appeal to synthetic judgment. Since human consciousness is part of reality, what guarantee is there that our synthetic judgments of today will be the same tomorrow? The only justification is to appeal to a certain uniformity of reality. But that uniformity of reality would justify synthetic judgments regardless of Kant’s bizarre metaphysical and epistemological dancing.

Part of the problem appears to be that Kant wanted to set synthetic judgments as having the same level of certainty as analytic judgments. But this is simply not possible. Analytic judgments are true ‘by definition’ (although they still can give us surprisingly long chains of interesting results: every result in mathematics is of the form ‘If X holds true, then Y must hold true as well to avoid there being a contradiction,’ and the results of mathematics are anything but obvious). Synthetic judgments are judgments about relationships between concepts which are not true by definition, but which apply to what we observe in the real world. Such judgments are justifiable if we take it as axiomatic that there is some uniformity to the world. But Kant’s entire justification also depends upon there being uniformity (uniformity of the constitution of human consciousness, for Kant). We therefore see that Kant’s position is 1. Unnecessary and 2. Subject to falsification by empirical observation (as when we conclude that the world is not in fact described by Euclidean geometry – or at least, that it may not be described by Euclidean geometry).

Indeed, Rand herself explains how synthetic judgment ought to work, as well as how all synthetic judgments are of a tentative nature:

It is precisely the “open-end” character of concepts, the essence of their cognitive function, that modern philosophers cite in their attempts to demonstrate that concepts have no cognitive validity…. [They] offer, as an example of man’s predicament, the fact that one may believe all swans to be white, then discover the existence of a black swan.

Rand prefers to say that as we bring in new observations, we may need to modify and clarify our concepts. Yet this is, in practice, no different from doing synthetic judgment with analytic concepts, as we will explain below.

According to Rand’s approach, our concept of ‘swan’ is based on trying to distinguish a collection of entities according to some common characteristics. Later, we may find that we need to modify the characteristics in a way that is most sensible. Her rules for the sensible way to do this are: ‘concepts are not to be multiplied beyond necessity,’ and ‘concepts are not to be integrated in disregard of necessity.’

In the case of black swans, it is objectively mandatory to classify them as “swans,” because virtually all their characteristics are similar to the characteristics of the white swans, and the difference in color is of no cognitive significance.

In keeping with this approach, Rand denies that a concept needs to have a set of absolute characteristics; she prefers to maintain that a concept includes only a ‘common denominator’ that suffices to distinguish one group of known entities from all other known entities, a characteristic that she calls the ‘contextual’ nature of definitions. As more entities are discovered, or as known entities are analyzed further and more properties are discovered, the concepts may need to be changed around. By doing so, she avoids the problem of analytic and synthetic logic.

The problem, of course, is that full, non-contextual definitions are possible, even though psychologically, Rand’s description is probably pretty spot-on for how people actually think in day-to-day life (but note well that mathematics, for example, involves closed, non-contextual definitions). We will refer to such definitions as ‘closed.’ They consist of all characteristics that an entity falling under the concept must have; everything with those characteristics, and only those things, will fall under the closed concept (by contrast, Rand’s open-ended concepts only involve enough characteristics to distinguish one class of known entities from other known entities). But this does not change her basic program. This is because the process of modifying open-ended definitions in context is essentially similar to the process of modifying synthetic judgments involving closed definitions.

For example, a closed definition may be made which describes swans, but which does not refer to colour. We then synthetically conclude that, since all entities which fit our description are white, that these characteristics go together. Later, we discover that there is a bird which fits our description but is not white. We then have a choice: should we include the color white in our closed concept? The reasons for not doing so are the same as the reasons for modifying Rand’s open-ended concept.

The point is, however, that synthetic judgments are contextual. They are made in the context of the known entities. We can have closed concepts, rather than Rand’s open-ended ones; this simply moves the contextual nature of our knowledge from the open-endedness of the concept itself, to the synthetic judgments we make about the relations between different concepts.

This is what Kant wished to avoid: that synthetic judgments would be open-ended and contextual. But that is what they are.

Yet it would be equally erroneous to conclude that all possible synthetic propositions are equally valid. The justification for synthetic judgments comes from the uniformity of nature (see, e.g., Russell, ‘Our Knowledge of the External World’). Corresponding to Rand’s two criteria for open-ended concepts:

  1. Concepts are not to be multiplied beyond necessity
  2. Concepts are not to be integrated in disregard of necessity

we may make a criterion for synthetic judgments:

  1. Synthetic judgments are not to connect closed concepts unless they are necessary based on our current knowledge
  2. The closed concepts which we use on a practical level should be the simplest ones that suffice for usefully categorizing known entities, that is, they should select out the minimum number of distinguishing characteristics for this category among currently known entities

It seems that anything done via open-ended concepts can be done using closed concepts and synthetic judgments. And the reverse is true as well. The justifications, in each case, will be similar.

Nonetheless, it is understandable that Rand wished to avoid talking about analytic and synthetic. First, this is a throwback to Kant (though the distinction between analytic and synthetic is nonetheless a useful one). Second, her approach is more psychologically natural, even though in my opinion it is not as natural from a perspective of pure reason. It is for this second reason that I prefer to consider closed concepts and synthetic judgments, especially since my own field of mathematics is one which uses only closed concepts and analytic reasoning.


NOTE: If you prefer, you may substitute ‘deductive’ for ‘analytic’ and ‘inductive’ for ‘synthetic,’ when referring to logic or reason.

Contra Hegel

A brief note:

Contra Hegel and Sartre, ‘To be P’ does not include the notion of ‘to be not P,’ and the predicate P does not contain any positive notion of what ‘not P’ is. Though it is true that we may contrast things which are mutually exclusive, the positive form of P does not contain anything about the possible forms of things which are not P.

To demonstrate this is quite simple, I believe. The notion of ‘not red’ does not in any way give us a notion of blue or green. We might imagine a mind that could perceive visual sensations only in shades of grey; ‘not white’ would in such a mind be contrasted with various greys and black, but would not in any way give rise to the notion of red. In a mind with no hearing, ‘not visual’ would in no way give rise to a notion of audible sensations. We can ourselves imagine that perhaps other colors exist of which we have no positive conception, or other senses of which we have no positive conception.

The idea of Hegel/Sartre and others, that a thing contains the notion of its negation, seems really to stem from the fact that we often see predicates P1 and P2 such that P1(X) and P2(X) are contradictory, so that P2(X) implies Not P1(X). For example, a square cannot be a triangle. But this is because three does not equal four. The notion of a square does not in any way contain the positive notion of a triangle hidden within it, so that it can ‘deny’ being a triangle. Nor does the notion of ‘white’ have the notion of red in it, to deny being red. It is merely that these forms are different, and existing in one form means not existing in any of the others. But this tells us nothing about what possible forms there may be. Until we conceive both forms, we have no conception of their contradiction.

In such a manner, the entire foundation of Hegelian philosophy falls at a stroke, since it is dependent upon the sophistic notion that evrry concept contains every other concept within it via negation.

Bullpups 2: Not Just Shorter OAL (and a correction)

I touched on this in the previous blog on Bullpups, but the main point I want to get at here is that the advantage of a bullpup is NOT simply to reduce the overall length (OAL) of a rifle, as some people have stated (notably, this opinion has been expressed on InRangeTV by Karl and Ian – two people whose views on guns I greatly respect, but in this case, I disagree). But first, a correction, or maybe supplement, regarding OAL to barrel length ratios:

Bullpups and ballistics

The ballistic purpose of a longer barrel in the bullpup is not necessarily just to give a short rifle the ability to act as a DMR (designated marksman’s rifle).

I stated in a previous blog that this was a big point for bullpups, i.e., that the point of the long barrel was that it gave more accurate shots. But as it turns out, I think I was somewhat off on this. It is probably true in theory, since DMRs generally required somewhat longer barrels (a 10.5 inch barrel is unlikely to be very good as a DMR). However, most military bullpups are 3 MOA guns (their shots spread out over a diameter of three inches at 100 yards), which is acceptable by the standards of a frontline combat rifle, but not necessarily good for a DMR. But why, then, is the smaller OAL/barrel length ratio desirable? Why not just go with a short barreled rifle in a conventional design? The reason IS largely about ballistics, but not necessary ballistics in the sense of accuracy.

We can really give three reasons:

  1. You COULD configure a bullpup to be a DMR, it’s just a lot aren’t made as DMRs (at least, not the ones that are available to civilians in the US). This shouldn’t be surprising – many AR-15s, for example, are 3 MOA rifles, even with their greater overall length. The difference is that you can go out and buy $3000 AR-15s that are designed for long-range marksmanship, whereas most bullpups are only available in one configuration. If someone wanted to design and sell bullpups configured for long range accuracy, that would probably be quite feasible. But more on long range accuracy later when we discuss the physics of bullpups vs. conventional rifles.
  2. Some bullpups actually do get really small OALs by using short barrels AND the bullpup design, for example, the mini Tavor (x95) with the 11 inch barrel. This achieves OALs which are simply impossible with a conventional rifle (unless you really sacrificed ergonomics).
  3. The big advantage of having a longer barrel: better terminal ballistics. Because the bullpup has a standard length barrel, it will fire bullets going faster than the conventional short barrel rifle. This translates into more ‘stopping power,’ which is a measure of how likely it is that, when you hit a target, the target is disabled or killed (morbid, yes, but that is the relevant consideration). This is a BIG consideration, especially in militaries. Now, as aforestated, most military rifles are only required to be 3 MOA guns, whether conventional or bullpup. However, a bullpup gives much better stopping power for the same OAL when compared to a conventional rifle. Note that you can get the same stopping power/terminal ballistic performance out of a conventional rifle, but it will have a substantially longer OAL (probably 4 to 8 inches or so).

Now that we’ve corrected this issue about ballistics, let’s discuss the physics of handling a bullpup vs. a conventional rifle.

Moments of Inertia (Advantage: Bullpups!)

Stone’s study (see previous blog) found that bullpups, with 83% confidence, result in better scoring, when shooters have to quickly engage multiple targets. Why would this be? I think there’s a simple reason: moments of inertia.

When holding a rifle, whether conventional or bullpup, and trying to move it around to aim it at multiple targets, the shooter is essentially moving a rod-shaped object around, with a pivoting axis that goes through the shoulder or possibly the centre of the shooter’s torso. If the rod had the mass distributed equally along its length, the moment of inertia (if the base of the rod were pressed against the axis of rotation) would be one third the mass, times the square of the length. In other words, as you lengthen the rod, you REALLY knock up the moment of inertia.

Now, a bullpup is shorter than a conventional rifle, so if the weights were spread along the lengths evenly, the conventional rifle would have a somewhat higher moment of inertia. However, conventional rifles like AR-15s tend to have almost no weight in the stock, which means their moment of inertia is even greater than what we would get approximating them as an homogenous rod.

Why does this matter? Well, when target switching, it’s easier to use something with a lower moment of inertia, because it requires less torque to get the same angular acceleration, or in other words, it takes less effort and strength to move around the item with the lower moment of inertia.

Jerry Miculek, who is a famous competitive shooter of conventional AR-15s, has a video on YouTube entitled ‘Battle of the Bullpups,’ and at one point, he comments on how the bullpup moves from target to target so much more easily than he is used to, that is throws of his aim. Now, throwing off his aim is bad, but it is easy to see that this is simply because he is used to having to manhandle a gun with a larger moment of inertia.

This is why bullpups might be expected to do better in close quarters combat against multiple opponents, or in situations where the shooter needs to acquire targets from disparate directions very quickly: it is easier to move the bullpup around. This may also mean that the gun is easier to fire from the opposite shoulder (although there is still the issue of brass flying in one’s face).

Note that the bullpup will have the smaller moment of inertia if the two designs have equal masses (weights), but that even if the conventional rifle is lighter, the bullpup is still likely to have the better moment of inertia, because the length term gets squared in the formula for moments (and because the stock has so little of the weight in a conventional rifle).

Another advantage to this smaller moment of inertia is that it makes it easier to hold the rifle at the shoulder for long periods of time without tiring, including holding it one-handed. This might be necessary when performing overwatch on a given area, or when needing to cover an area while manipulating things with the other hand.

The potential downside to a smaller moment of inertia is that this makes the rifle a bit more sensitive to any twitching on the part of the shooter when aiming at long ranges without supporting the rifle. Higher moments of inertia mean that a slight torque from twitching will not move the aim point as much as the same torque with a lower moment of inertia. This brings us to…

Firing From Support (Advantage: Conventional!)

Let’s say you put a bipod (or even a monopod or tripod) on the front end of a rifle, and try to fire from prone. Or perhaps, lacking this, simply rest the front of the handguard on something (in the best case scenario, a bench rest; in the real world, probably a log or something similar), and try to hit a target at a long range. We will assume, for the sake of comparison, that both the conventional and bullpup rifles have got the same mechanical accuracy (i.e. if you could hold the rifle perfectly still, they would fire bullets just as consistently to the same place), and both have the point of rest equally distance from the end of the barrel. We will also assume both are affected equally by the support (so either both have free floated barrels (see the appendix), or the pressure on the barrel has the same effect on accuracy).

Here’s the problem with bullpup designs: in this particular context, the bullpup gives two disadvantages.

First, the trigger. A bullpup trigger will never be as nice as a conventional trigger, although it can probably be made almost as nice, to the point of making little effective difference.

Second, and this is the bigger issue: geometry gives an advantage to the conventional design. This is because its greater OAL actually confers an advantage here. As the shooter adjusts the butt of the gun to aim it, the butt is further from the bipod or point of rest. This means a movement in the butt actually gives less of a change in the angle than with the bullpup (imagine two levers, one that is one foot long, and one that is one yard long. If you move the ends of both levers an inch, the one foot lever goes through a much larger angle than the one yard lever). The same goes for movement of the firing hand, since the firing grip is closer to the tip of the barrel in the bullpup, and hence closer to the bipod or other support.

As a consequence, the bullpup design will be somewhat more sensitive to things like the shooter twitching while firing. Plus, the trigger being worse means the shooter is more likely to apply unwanted torque to the grip area, and in any case, the grip is closer to the point of rotation. The upshot is that it will be harder to make supported shots like this using a bullpup than a conventional weapon.

So, for a rifle designed for making long-distance shots from a supported position, the conventional design is better!


Basically, I suspect bullpups are better for close quarters combat, but conventional guns are better for long range shots from supported positions. This seems to tally with Stone’s result on shooters doing better with bullpups when needing to switch between targets at 25 yards, and with other results about the efficacy of bullpups and conventional rifles. It also tallies with the preference that longer-range shooters have for their conventional AR-15s (which is helped by the simple fact that it’s easy and cheap to get a free-float AR-15, and not so much to get a free-float bullpup).

What I think is ultimately also worth noting, though, is that both designs can do both types of shots reasonably effectively. The bullpup may give a slight edge in close quarters (say, within 50-100 yards), but not exactly a game changing edge – the side with better tactics and training will still probably win. And similarly, the conventional design may be better for long range shots, but again, not enough to be game changing, especially since most military rifles are 3 MOA anyway, and thus unlikely to be all that accurate at longer ranges (300-600 yards) anyway.

The upshot is that for a lot of purposes, I think bullpups come off slightly better. This is especially true for things like close and mid-range combat in the military, police work, and home defense. However, as Stone’s study found, many shooters, especially if they are used to a conventional design, do not realize they are doing better with the bullpup design. Furthermore, the difference is marginal enough that for most militaries, other considerations will probably overshadow them.

Appendix: Free float?

A free floated barrel might sound fancy, but it really just means that the barrel is secured at the chamber, and nothing presses on the middle or end of it. As it turns out, having things that touch the barrel is bad for accuracy, so free float barrels are generally preferred by those who want high accuracy at longer ranges. However, many military rifles are not free floated. This is because you can still achieve perfectly reasonable accuracy (3 MOA) without free floating. Free floating is particularly helpful in preventing pressure on the handguard (whether from gripping the gun or from resting it on an object or bipod) from messing with accuracy, however. Obviously, this is mostly an issue when talking about longer ranges.

Appendix: 3 MOA?

Instead of measuring the angle which a shot deviates from the aim point using degrees or radians, it is convenient to use ‘minutes of angle.’ Roughly, one minute of angle corresponds to one inch at one hundred yards, two inches at two hundred, etc. A 3 MOA gun means that if you fire a group of five shots with no operator error and with the gun held perfectly in the same position for all five shots, those shots will all land within 3 inches of one another at 100 yards.

Most militaries expect their battle rifles to fire 3 MOA. This is perfectly adequate because most combat takes place within 300 yards, and even if your bullets are spreading around a disk of nine inches at 300 yards, that’s still enough to hit a human-sized target most of the time if you aim at their centre (one of the big takeaways from WWII was that most combat takes place within 300 yards). While marksmen and snipers obviously want better than 3 MOA, for most people, it’s just not a big practical issue.

Note that measuring what MOA a rifle of achieves is often done in a manner that strikes me as statistically suspect, because it will be done by firing a bunch of five shot groups, and just seeing how each one spreads. However, this has many statistical problems. For example, one gun might have four groups of five rounds that have a 1/2 MOA spread, and a single group with a spread of 36 MOA. How should that be reported? Another rifle might consistently fire 5 MOA groups instead. Does it get labelled a 5 MOA rifle, while the previous one gets called a ‘usually 1/2 MOA’ rifle, despite the one group being truly terrible? Or we might get an example where in each group of 5, we have 4 shots that are within an inch of one another, with a single ‘flier’ that pushes the group to 3 inches. To my mind, that should rank differently than a rifle that consistently fires bullets that sit around the circumference of a 3 inch circle. But the conventional measure of accuracy treats them the same.

So, trying to measure MOA with five shot groups is not exactly a statistically precise affair. To alleviate this, some people prefer to use things like the mean radius of deviation of the shots. This strikes me as a much more useful number.

The advantage of giving an MOA based on the maximum spread, however, is that it gives a worst-case-scenario estimate. For whatever reason, this is the generally-used measure of accuracy among shooters.

Bullpups part 1: answers to some objections

I want to talk about bullpup carbines vs. traditional carbines, and why, in my opinion, the bullpup is slightly superior (but also why it doesn’t see wider support in the West). My basic thesis is this: while both designs have upsides and downsides, people tend to overemphasize the downsides of the bullpup, and ignore or even deny some of its strengths, which ultimately make it a (slightly) better design.

First, some background: what is a bullpup vs. traditional carbine? In a traditional carbine, the magazine and chamber are in front of the firing hand (think of an M-16), whereas in a bullpup, the magazine gets inserted behind the firing hand (for example, the Israeli Tavor, or the Australian Steyr AUG). The result is that the gun can be made about six inches shorter while keeping the same barrel length.

Stone’s Study

Throughout this article, I will be referring to a study by Stone et al., on the performance of shooters with bullpups. This is available at http://journals.sagepub.com/doi/abs/10.1177/0018720813509107. Here is the full information:

Biomechanical and Performance Implications of Weapon Design:
Comparison of Bullpup and Conventional Configurations
Richard T. Stone, Brandon F. Moeller, Robert R. Mayer, Bryce Rosenquist, Darin Van Ryswyk, Drew Eichorn First Published November 8, 2013

The Usual Analysis, and a Rebuttal

Usually, when people talk about bullpups, they give a list of pros and cons that looks something like the following (I’m taking this rather negative list from https://anarchangel.blogspot.com/2005/03/why-bullpups-are-persistently-bad-idea.html) (in deference to this source, keep in mind that these objections were written in 2005):

PROS: the gun is slightly shorter (but, those who dislike bullpups will say this actually doesn’t matter).

CONS: everything else, including, but not limited to:

  1. Triggers are bad, because they are more complex.
  2. If the gun explodes, it’s slightly closer to your face because the chamber has been brought back six inches.
  3. Mag changes (reloads) are more difficult and slower, and less ergonomic: the magazine is harder to see, and is further away from the shooting hand which makes it harder to line the mag up without looking. ‘Magwells should either be in your dominant hand, or just in front of it; because it is far more difficult to manipulate anything dexterously that is located behind your dominant hand.’ Also, you have to move your hand further to get to the mag well, which slows the reload.
  4. The magazine being behind the firing hand makes it harder to fire the gun while prone.
  5. Charging the rifle can be more difficult.
  6. Because the brass ejects so close to the shoulder, you can’t use the gun ambidextrously, or else the brass will hit you in the face (this is one of the most common objections.
  7. Bullpups have a centre of mass that is too far to the rear. From the aforementioned blog: ‘The balance point on most bullpups is in between your hand and your shoulder when mounted, which is unnatural. We have a natural tendency to try to balance things between our hands, not between our hand and shoulder…. This balance will tend to make a bullpup tend to shift its butt under recoil, unless it is very tightly mounted to your shoulder; particularly during rapid fire. This tendency is somewhat countered by the position of your support hand so far forward on the barrel,  by the fact that the overall leverage moment of the muzzle is lower (the muzzle isn’t as far from either your shoulder, or your dominant hand), and by the fact that most bullpups have straightline recoil. A conventional rifle is balanced in between your dominant and support hands, and there are good reasons for that. A human being naturally handles things that balance in the palm, or in front of your dominant hand, better, because we naturally want to balance things between our hands.

Let’s address each of these points in turn. Then we will address why the ‘pros’ list should really be much LONGER than it usually is.

  1. Trigger complexity: Granted. It is possible to simplify the trigger workings in a traditional rifle more than in a bullpup. But once the trigger is made good enough, there’s diminishing returns, isn’t there? Bullpup triggers can be made so close to feeling as good as traditional triggers that it doesn’t really matter. And this ‘bad trigger’ problem does not stop the bullpup from actually being more ergonomic and accurate to shoot (see below).
  2. Again, granted, and this is probably the most severe issue with bullpups. Some rifles do try to deal with this, however; e.g. the Kel-Tec RDB has a specially reinforced shield to help protect the shooter. On the other hand, if the gun is kept in good condition, this is probably not going to be a major issue.
  3. Here I categorically disagree (and this is something that also applies to 5 and especially 7 as well, so I will answer all three objections here): bullpups are easier to handle than traditional rifles. See, by putting the centre of mass nearer to the shoulder, this means that the firing hand gets a much greater leverage advantage when trying to support the rifle during magazine reloads or even just while aiming. It’s much easier to hold a bullpup stable while doing a mag change, though this is less of an issue if you use the proper, safety-conscious civilian reload method of raising the barrel to sixty degrees above horizontal before reloading. If you want to reload while keeping the gun aimed, though, the bullpup makes this much easier, by giving your dominant hand so much leverage. After all, the gun forms basically a heavy lever with the shoulder as fulcrum. If the centre of mass of the lever is behind your grip, as in a bullpup, that makes it much easier to hold onto than if it is in front, as in the traditional carbine. A very short amount of practice makes it so you can reload without looking – and in fact, it’s not any harder to figure out where your hands are relative to one another when you slip the magazine in behind your main hand as when you put it in front. Nor is the mag well any further from your main hand with a good bullpup design than it is with a traditional design – it’s just behind, instead of in front. In addition, you can place your forearm right next to the mag well, which means you can press the magazine against your forearm to feel whether it is aligned correctly. This is not possible when the mag well is in front of your shooting hand. The only part of this objection that is valid is that you have to move your support hand further when doing the reloading.
    What’s more, Stone et al found that, when comparing the AR-15 and FS2000 reloading times, although the reloading time on the AR-15 style platform was faster than with the FS2000 (bullpup), there was no significant difference in reloading times. And in addition, the FS2000 has a particularly tightly-sealed mag well. It was deliberately made to form a very tight seal and keep out all dust. This means that other bullpups might be faster to reload than an FS2000. Regardless, the point is that the difference in reloading times is, at worst, marginally slower for bullpups, and this is more than compensated for by factors we will mention later.
  4. Somewhat valid, but probably not that big a deal, I would say. This is especially true if the gun is issued with a vertical foregrip that extends into a bipod. So it is an easily fixed problem.
  5. I don’t think this is true. In fact, the side-mounted charging handles on my bullpups are easier to reach than the ambidextrous AR-15 charging handle. Of course, these side-mounted charging handles are not ambidextrous, but this is not an inherent issue to the bullpup design: the Desert Tech MDR features a side-mounted charging handle on both sides. As the blog even acknowledges, this is more an issue of individual rifle design than of bullpup vs. conventional. Finally, as long as the gun locks back on an empty mag and has a bolt release for reloading, the charging handle won’t usually need to be used when reloading.
  6. This is probably the most common objection to bullpups: they can’t be switched from shoulder to shoulder, and hence, if you have to lean around cover in certain directions, you are out of luck: either you get smacked in the face by brass, or you die because you’re too exposed trying to fire from your main shoulder. Also, you can’t issue the gun to a left-hand shooter without doing massive changes to the bolt and ejection.
    The problem with this objection is that it only applies to bullpups with poorly-designed side ejection. Even with the most basic side ejection, a bullpup still CAN be fired from the other shoulder, as long as the shooter takes care not to put their face too far forward. But there are also numerous solutions to this problem, leading to truly ambidextrous solutions:
    The F2000 and Kel-Tec RFB both send empty brass out the front of the gun through an ejection tube. Granted, this may make clearing any jams a bit more complicated.
    The Tavor simply uses a brass deflector to ensure that the side ejected brass goes sideways or forward, and does not hit the shooter in the face. After-market deflectors are also available for the Steyr.
    The Kel-Tec RDB ejects the brass downards right next to the butt of the gun. Again, this might introduce extra complexity when trying to clear jams; also, as presently designed, it could use a dust cover to help keep the inner workings clean. However, this latter is something that could easily be fixed in design.
    Perhaps the best solution is the Desert Tech MDR’s approach, which ejects brass through a side port, but uses a relatively simple system to eject the brass forward along a diagonal, together with a robust yet compact brass deflector to ensure that the brass will not hit the shooter, regardless of which side the rifle is on.
    The point is, the problem of ambidexterity and support side shooting is one with many good solutions. It is not a problem that is inherent to the bullpup design. And even conventional rifles can have issues with ejected brass potentially hitting the shooter when doing support side shooting (though usually in the arm, rather than the face).
  7. To return to the ergonomics, I find that the reasoning here is somewhat specious. The idea that humans want to have the point of balance be between our hands sounds obvious, until we consider that this only really makes sense if we’re holding something parallel to our chests, while the gun is pretty much perpendicular to the chest. In that position, we actually want to have the point of balance as close to the shoulder as possible (think about how people carry heavy poles, for example). This is precisely what the bullpup offers. Such a centre of mass requires far less strength to maintain the position of the rifle, which translates, even for strong shooters, into less strain. This is backed up by the results of Stone’s study, which we will discuss in more depth below.

The upshot here: most objections to bullpups turn out to be blown out of proportion, or else entirely moot.

The True Bullpup Pro/Con List: Stone’s Study

Stone et al studied 48 shooters, with conventional and bullpup weapons. For the 556 NATO weapons, they used an M&P AR-15 copy, while the bullpup was an FN FS2000. The FS2000 is generally considered to have an unpleasant trigger for a bullpup, and to have rather difficult magazine changes. The study showed that after performing the challenges, the participants rated their abilities with a conventional rifle to be better than with bullpups.

But the numbers told a different story.

First of all, they did find that the FS2000 was slower to reload. This may be simply due to the FS2000 having a very tight mag well, and the fact that unlike most AR-15s, the FS2000 bolt does not lock back on an empty mag, nor does it have a bolt release, meaning the weapon must be charged after reloading – I am unsure, but I would assume that the M&P 15T Stone’s study used had a bolt release, since that is fairly standard on AR-15 platforms, which would alone make for a faster reload. In any case, the M&P 15T example that I looked at (technically, a picture of one online) clearly had a bolt release. However, whatever the reason for the FS2000 being slower to reload, they found that this difference was minimal and unlikely to be significant.

But that was not what was so surprising. The surprise was that these shooters, despite being more confident on the traditional rifle design, were actually better with the bullpups! They were both more accurate and more biomechanically stable. To quote from the study:

The results showed that the bullpup configuration resulted in more biomechanically stable users who, due to their increased stability, were able to shoot more accurately in an equivalent period of time. These results were in spite of the fact that the participants in the study demonstrated a preference for the conventional configuration.
Stated differently, the results show that bullpup weapons can be fired more quickly with the same overall accuracy. At 25 yards, the results of this study indicate differences of a few inches in accuracy between the weapons, given the same amount of time to fire. These few inches could easily mean the difference between a life saved and a life lost in an armed confrontation. At distances greater than 25 yards, a few inches could easily translate to a few feet.

Now let me propose a different list of pros and cons for bullpups vs. conventional rifles.

Bullpup advantages:

  1. Short overall length (OAL) for barrel length.
  2. Better ergonomics translating into more stability and accuracy.

Conventional rifle advantages:

  1. Simpler engineering and trigger design.
  2. Slightly faster reloading.
  3. Possibly slightly better when going prone.

Now, many people are going to perhaps question the second bullpup advantage listed here, but let us take things in order, starting with the conventional rifle advantages:

Trigger design/complexity:

It’s true that the conventional rifle can have a ‘nicer’ trigger. However, as stated before, this is going to be a marginal difference. And Stone’s study, with one of the most notoriously badly-triggered bullpups, shows how this difference really plays less of a role than one might think, since the bullpups proved more accurate and not significantly slower.

Faster reloads:

Yes, you can reload a traditional rifle faster. But do you need to? Sure, there are times when a faster reload is advantageous, but I cannot think of many cases where a tiny difference in reload time will be all that important, especially in a fireteam where other team members can provide cover while one member reloads. And even while alone, Stone’s results show that the bullpup can be aimed faster and more accurately. This seems like a bigger advantage than a faster reload.

Better going prone:

This seems like the main insurmountable advantage to the traditional design. However, it can be largely alleviated by using a foregrip that has extendable bipod feet or an extended monopod on the bottom (this appears to be something which the Israelis use sometimes on their X95 bullpup). And balancing a conventional rifle on its magazine like a monopod is hardly ideal either, while having the magazine used as an ersatz monopod behind the firing hand on a bullpup is still an option, albeit even less ideal.

Now the bullpup advantages:

Short OAL for a long barrel:

This is an advantage which really seems to get understated a lot, perhaps because it is to some extend situational. I have heard from a guy who was in, I believe, the Singapore military, who, if I am remembering rightly, talked about seeing trainees with conventional weapons have a much harder time in CQB (close quarter battle) training than those armed with bullpups, because it is so much easier to get a longer weapon caught on something in cramped conditions. However, it’s true that if you are just talking about shooting at a range, or even in a competition – even a competition that seeks to simulate tactical conditions, such as the two and three gun matches exhibited frequently on InRange TV (a truly excellent channel about guns, by the way – my disagreement with them on bullpups is despite the fact that Karl and Ian know far more about guns than I do) – you may not realize the advantages that having a short OAL and a long barrel provides.

The thing is, a short OAL is good for use in close combat. But you can get a short barreled rifle (SBR) in a conventional design, and it is about as short as you can make a bullpup (past 25-26 inches, there’s not much point in making the carbine shorter, since you would lose any place to put your support hand). So why not just use an SBR? For civilians in the US, the answer is simple: an SBR requires a class III stamp from the ATF, and while anyone who can pass a background check can get this, it is an extra expense and a royal pain, really. But what about for military/LE?

Well, the SBR is fine for close quarters combat. But the advantage the bullpup provides is that it gives you an SBR-sized package with a DMR (designated marksman)-sized barrel, and hence SBR size combined with DMR ballistics. In other words, you can swap from easily clearing a house or fighting in a jungle with extensive undergrowth, to taking a shot at the farthest distances a conventional battle rifle is capable of, with ease. This also means the bullpup will be putting more energy into the bullet, and hence in general will have better stopping power, even at close range. The SBR will have a lower energy bullet, and it will lack accuracy at ranges where the bullpup still operates very well.

Bullpups have better ergonomics and accuracy:

This is the point of which I think most people are simply unaware. The traditional argument usually is bullpups having good length vs. bad triggers, bad reloads, and bad ergonomics. But I think the idea that bullpups have bad ergonomics is just wrong. They actually have good ergonomics. Why, then, do people think otherwise? Well, if they’re American shooters, perhaps it is because they grew up shooting traditional rifles. Many shooters in America start with basic .22 rifles and hunting rifles, which are much closer to the traditional carbine than do a bullpup. So it is possible that they are simply used to the feel of such a rifle. After all, if we are used to doing something a certain way, then even if some other approach is better, it will still feel wrong when we do it.

What Stone et al found, of course, was simple:

American shooters think they do better with traditional carbines, but in fact they are more stable and more accurate when shooting bullpups.

I mention ‘American’ here because the participants in Stone’s study, insofar as they were trained, were more likely to be trained on traditional carbines than on bullpups – Stone’s paper acknowledges that there was a strong pro-conventional bias among his participants as a result of both national bias and bias from previous training. If this is true for American shooters, it is probably as true or more true for shooters from other nations, who are likely less biased.

This is a major plus for the bullpup, if the result replicates (and assuming the results were not obtained by some statistical sleight-of-hand). Of course, as with any study, one must be careful and make sure that the results replicate (i.e. that when we try it again, we get similar outcomes as the first time). We must also consider that sometimes it is possible to ‘hack’ the data in ways that give a desired outcome. Therefore, Stone’s results should not necessarily be taken as definitive. And in fairness, Stone’s study is fairly recent (2013 publication) and not very well known. But to my mind, this is the definitive argument in favour of the bullpup. While it does not mean the bullpup is so superior to the traditional rifle as to render the latter obsolescent in any way, it does mean that the design is likely to be more ergonomic and to give rise to higher levels of stability, and thus accuracy, in shooting, which gives the bullpup design a definitive edge.

But Militaries Aren’t Using Bullpups?

It is worth noting that the US military has never shown interest in adopting a bullpup, and that at least two nations which did adopt bullpups are moving away from them (the UK and France). However, there are a few points to be made here.

  1. France is moving away from the bullpup because the FAMAS is outdated, not because it was badly designed. And their choice of a conventional rifle in the HK 416 is more or less motivated by logistics – they no longer have the ability to domestically procure rifles, and the HK 416 is what is available. The same sort of applies to the UK, which also had a very bad experience with the L85, not because it was a bullpup, but because until HK fixed it with the A2, it was a wretched gun. The HK 416 also is a well-tested platform. The upshot is that these nations are going for tried and true, easily available weapons, and those happen to be conventional, in part perhaps because the US likes conventional weapons.
  2. The US military has often been on the wrong end of weapons advancement when it comes to small arms. They refused the high-capacity lever action rifles (e.g. the Henry) in the late 1800s. They did adopt self-loading rifles as standard in WWII before anyone else had done so, but immediately thereafter, they insisted on a full-power cartridge over an intermediate (this would prove to be a bad mistake), then insisted on the M14 over the FN FAL (another mistake). Not to mention that weapons procurement is just in general notoriously political in the US – consider the naval ship that senator Trent Lott pretty much forced the navy to build, the F-35, and the many other politically-motivated projects in which the US military is forced, or chooses, to engage.
  3. Militaries need to be concerned about things like durability and reliability, as well as resistance to dirt and the like – these are issues which are more about the particular firearm than about conventional vs. bullpup design. The M4 platform, and related platforms, are extremely reliable not because of its conventional design, but because of precision machining and the fact that over time, the design has been substantially refined. By contrast, many bullpups are relatively new designs. If the M-16 had been given a bullpup design, which would have been perfectly feasible, then perhaps we would see a bullpup with the same sort of reputation for reliability and excellence as the M4 sees today.
    In any case, apart from a marginal increase in complexity in the trigger system, bullpup design does not in itself have much effect on reliability or durability.
  4. The US is loathe to adopt foreign designs. A big part of the reason for the M14 winning over the FN FAL was because the latter was a Belgian design. The bullpup is ‘tainted’ with being an Austrian/British/French/’generally furreign’ design.
  5. Militaries like to do things the way they’ve always done them. There’s a good reason for this in part: if it isn’t broken, then trying to improve it could actually make things a lot worse. Since conventional M4s and M4 SBRs are more than adequate, then, why try to change things? Which leads to…
  6. Militaries have far better things to spend money on, than changing an infantry platform that is performing adequately. Even though I am a bullpup fan, if I were in charge of US military procurement today, I would not try to replace the M4 with a bullpup, because it would be a waste of money. Yes, we could get something slightly better in a bullpup; no, that doesn’t justify the expense, when what we have works well enough to do the job.
    I would argue, though, that the marines missed a great opportunity for a bullpup when they chose to replace the M4s with the M27 IAR. A bullpup would have provided the same marksmanship and long-range suppressive fire in 556 NATO that the M27 gives, while keeping the closer quarters battle capabilities which seem like they might be quite important for marines. But perhaps the M27 has enough parts in common with M4s that it makes more sense logistically. There might have been some unfortunate ‘real riflemen shoot conventional rifles, and every marine is a real rifleman’ sort of mentality that influenced the decision, though.
  7. Plenty of militaries, including large or ‘respected’ militaries, have adopted bullpups. For example, the Chinese, the French, the UK, the Australians, and the Israelis. And of these, only the French and the UK are rejecting them, and this is because of logistics, not because of any flaw in the design of bullpups. The other nations continue to use them, apparently quite happily.

I am, therefore, unimpressed by the fact that the US refuses to use bullpups. And when we analyse those nations which are choosing to replace their bullpups with something else, the reasons are inevitably logistical (they’re trying to update their weapons, and the major production facilities from whence they can purchase are manufacturing M4 or HK 416 derivatives) rather than based on a shortcoming of the bullpup design per se. A former British special forces operator with whom I was able to talk was greatly in favour of bullpups, for example, despite having extensive experience with many weapon platforms.


I do not believe that there is a definitive winner in the bullpup vs. conventional carbine debate, at least not if we are looking for one design to be clearly superior to the other the way that, say, a breech loader was superior to a muzzle loader. However, the evidence does suggest that at worst, the bullpup is as good as the conventional rifle, with a couple of minor weaknesses (usually overstated) and a couple of minor strengths; at best, however, the results of Stone’s study suggest that the bullpup has a major advantage in ergonomics and potentially accuracy. This is, however, not a decisive enough advantage to render the bullpup worth choosing over a conventional design when faced with either:

  1. A bad bullpup design vs. a good conventional design. While a bad breech loader was probably better than a great muzzle loader (as long as it was functional and not outright dangerous), a badly-made or designed bullpup is probably worse than a well-made, well-designed conventional rifle, since all the accuracy and ergonomics in the world will not help if the gun will not shoot.
  2. Logistical demands such as those now faced by France, which make it easier/cheaper to procure conventional rifles.

The French WC victory and ethnicity

Following the French 2018 World Cup victory, people on both the far left and the far right began tweeting things along these lines: ‘Congratulations to Africa on winning the world cup. Isn’t immigration wonderful/horrible?’ In response, a number of people said ‘They’re FRENCH, not Africans! Ethnicity doesn’t matter! They’re French citizens so they’re French! Many were born in France! FRENCH!’

I find both positions rather sad and inaccurate. Let me explain.

My grandfather was born in Finland. His dad had been in America before moving to Finland, where he fought for Finnish independence (and was wounded). My grandfather grew up in Finland, and never lived a day in his life in Sweden (though he has of course visited) – yet he is proudly Swedish, not Finnish. Of course, he has nothing against Finns. He calls himself a Swede-Finn (and so did many of the others in his city, which had a large ethnically Swedish percent). He served in the Finnish merchant marines.

Then he came to America, where he served in our military for two years before going into civilian life as an electrical engineer, and marrying an Irish-American woman. He is of course an American citizen. But his Swedish ethnic identity has always been important to him.

Perhaps in part because this kind of appreciation of one’s ethnicity has always seemed normal to me, when I look at people try to erase this aspect of a person’s identity, and reduce all such things to citizenship, I find it very sad. Citizenship is very important, and should be what matters for political rights, etc., but ethnicity is a fascinating thing about a person. Why would anyone want to erase that aspect of themselves or anyone else? If I were to move to, say, Japan, and become a citizen there, I would nonetheless be quite upset if anyone described me as simply ‘Japanese.’ This would be unfair both to me and to the ethnic Japanese.

So when I look at these French players, yes, they’re French, but they’re not the same as the French who are descended from the Franks. Or the other ethnic groups that have been in France for hundreds of years. And I think it is unfair to erase that difference, just as it is unfair to simply call them Africans.

In the film (which everyone should watch) ‘The Free State of Jones,’ the two main characters (a white man and a black woman) have a baby together. She asks, ‘Is she black or white?’ and he answers ‘Neither, she’s something new.’ As demographics change and people move and intermingle, people become something new, yet those new things have their roots in older things. We should acknowledge where these new types of people came from, while also acknowledging that they are something new. Neither the woke left and alt right ‘They’re Africans which is great/awful,’ side, nor the ‘THEY’RE JUST FRENCH’ side, is doing these players, or France, or Africa, justice.

Is it really racist?

What I want to discuss is how some policies, while they might seem racist, if you analyze the impact by race, might have nothing to do with racism (in the sense of believing one race inferior) at all.

Let me start with an analogy. Did you know that for drivers 18-25, males pay more in car insurance than females? Surely this is a sign of a deep sexism at work! Surely, the insurance companies hate men!

… Actually no, it has nothing to do with an hatred of men, and everything to do with statistical analysis. See, insurance companies have to judge customers, not based upon the customer as an individual, but rather, based on a few broad characteristics. And, statistically, males are WAY more likely to get into an accident. So, it makes sense to charge males more. You will, statistically, need to do so, in order to make a profit.

More generally, insurance companies do not charge everyone the same price for the same coverage, and this is based on statistical analysis, not upon bigotry.

Now, if you can grasp that this is not about bigotry, perhaps consider the following scenario: A Republican in charge of voting districts engages in gerrymandering. This is absolutely wrong. But let’s say they do this to minimize the effect of black voting. Is this necessarily racist? As in, done because this person dislikes black people or considers them inferior?


See, gerrymandering is wrong. But if you ARE going to gerrymander, you do it so that your party wins. And if you see that some factor correlates with voting against your party – including race, since blacks tend to vote Democrat – you will include that factor, when you calculate how to gerrymander.

It’s statistics, and no different than what insurance companies do.

Starship Troopers: Some Thoughts

Warning: as writing goes, this is somewhat disjointed, because I cover a number of different aspects, sometimes without much segue.

I do not believe the Starship Troopers glorifies war. Or if it does, it does so in a good way, because it does not glorify senseless violence, but rather, to paraphrase the book itself, putting oneself between the violence of war and one’s home.

I do not know what Heinlein himself believed, or whether Rico’s opinions were really his. I imagine he agreed with a lot of Rico’s opinions, but perhaps not all; whether he really thought that limiting the voting franchise to military veterans was a good idea, in particular, is questionable. He probably did think that it was best for the individual to sacrifice themselves for the whole, which is a somewhat fascist notion, or rather, the fascists have such a notion (though it is worth noting that liberal democracies have had the same notion – including when said liberal democracies were forced to fight fascists).

Here is the thing: we evolved to sacrifice ourselves so our genes might continue. This is why people sacrifice themselves for their offspring (some other animals do so in even more spectacular fashions: see, for example, the spiders in which the male commits suicide, being eaten by his mate, so that his offspring have a better chance of survival). But thanks to our intellects, humans are in the rare (evolutionarily speaking) situation, in which we can die to try to promote the survival of our ideals. This is probably a by-product of the evolution of our intellect. Yet, it might end up being the key to the evolution of an intelligent species.

Rico is, of course, correct, that any group which limits its own reproduction, is going to be overrun by other groups that do not limit reproduction. The morality of how that reproduction is obtained is irrelevant. We see this now, with Western nations having low birthrates, and high numbers of non-Western migrants, causing a large demographic shift. Unfortunately, not only European genetics are thus being crowded out, but also Enlightenment ideals, as the people with the most reproduction belong to the most conservative, right-wing religious groups. If humanity ever achieved an equilibrium, it s still true that other species could throw this off with ease, should they exist, and should humans fail to reproduce rapidly.

Thus, Heinlein does not glorify war, though he does glorify the war, and perhaps rightly so: any group that wishes to survive, whether that group is defined by family ties, ideological ties, national ties, or some other kind, must face the possibility that they will be competing with another group that will refuse to be persuaded, and engage in what von Clausewitz called ‘the pursuit of policy by other means,’ i.e., war for survival.

The notion that the Terran Federation should never have any rebels strikes me as ridiculous, both on the face of it, and because his explanation is nonsensical: ‘A revolutionist has to be willing to fight and die – or he’s just a parlor pink. If you separate out the aggressive ones and make them the sheep dogs, the sheep will never give you trouble.’ But the Terrans do not do this. It does not seem, first of all, that a system could ever be created which would guarantee that this would be done, but secondly, since their system of Federal Service is entirely voluntary, there is no reason why an aggressive individual with revolutionary tendencies might not simply choose revolution over Federal Service. And in fact, there are doubtless many individuals who have the aggressiveness, but end up kicked out of Federal Service, and may feel themselves treated deeply unfairly, while also having strong political aspirations – Hendrick, for example, is just this type. Here, Heinlein appears to have created a rather unrealistic aspect of his world. Hendrick and his type would be prime candidates for rebels. Now, it may be that rebellion is impossible, because the Terran Federation controls all the weapons, but at the same time, an effective insurgency does not require parity of weapons; it requires only that the government be unwilling to engage in wholesale slaughter of civilians in order to put it down, together with a modicum of popular support.

This brings us to the next issue: the book intends to glorify the infantry, yet the MI really do not seem to be infantry. They use extraordinary amounts of equipment (namely, the powered suits), which functionally make them more of a mix of a tank and a ground attack helicopter, tactically speaking, mixed with a small amount of the additional flexibility that infantry have. Yet it seems incredible that the Terran Federation would not employ numerous less-well-equipped soldiers as grunts, with the MI retaining a role more that of support and special forces mixed into one. This is doubly perplexing, since the recruiters complain of having to find make-work for many of the volunteers: would not not make sense to send them as, if nothing else, canon fodder, given that the Federation has little concern over whether the volunteers survive Federal Service? Tactically, they could be very useful. We see, near the end of the book, how thinly spread the MI troopers are; numerous unsuited but armed troops could have easily made a big difference. Even today, tanks and helicopters and drones do not win wars; the lightly equipped grunt remains the focus of the battle (If you don’t believe me, you can ask Nicholas Moran).

In addition, the training of the MI seems absurd. They undergo a course with a 10% pass rate, which seems to rival the toughest special forces courses in the real world. Yet surely, to use a powered suit effectively, requires nothing more than what we would expect from a tanker or pilot today? Now, while these soldiers are required to have an high level of physical fitness, they are not held to nearly the standards of, say, a SEAL. The training may be more intended to psychologically weed out some of the recruits; if so, very well. But the weakness of the entire system (including that volunteers can quit service even if it is 30 seconds before a jump) is shown in the fact that the Federation spends much of the book losing the war, largely due to a lack of troops. History shows us that, while it may be nice to have only the most elite troops as soldiers, it is often more effective to bolster the numbers, especially when there are an excess of volunteers, and not enough jobs for them. Some motivated, though perhaps not quite so psychologically tough, individuals, who are required to serve their term once they sign on, might have proved decisive, or at least beneficial.

Finally, much is made of the low ratio of officers to enlisted men in the MI. But is this really the benefit that Rico thinks it is? He mentions officers having to take on multiple command roles as a matter of course; this is clearly not to anyone’s benefit, as an overworked officer is more likely to overlook things. Granting his complaints about the rest of the problems of having too many officers, in combat, command and control is important, and the lack of officers will be highly detrimental.

It is therefore my opinion that, on many levels, this book lacks a realistic picture of tactics and training.

Nonetheless, I love this book. It is unapologetic in portraying the spirit of camaraderie among the MI. It is also unapologetic in the notion that humans have a right to fight for their existence, even if that means destroying another species. Of course, cooperation might be the better option. But in this instance, it was not so far possible to even communicate with the Bugs.

In addition, even where I disagree with it, it raises very interesting questions to think about. Is there any advantage to ensuring that only those who understand the cost of freedom should be allowed to vote? The downside, of course, is that the book presents the advantages of such a system, but does not really engage in the debate over the disadvantages, including the very real likelihood, in my opinion, of a rebellion, as well as the problem that these oligarchs are likely to in reality vote for themselves to be made far better off than the other citizens. We know in our own system, thanks to the Princeton study, that even in our own system, with the voting franchise being widely distributed, that a small group of oligarchs have managed to usurp practical control, for their own gain. Is it really unreasonable to think that the Federal Service veterans would probably very quickly establish themselves as an upper caste, with all others relegated to lower castes?

Yet, it is nonetheless an instructive question to consider. Another interesting question is the role and form of punishment. Heinlein appears to be empirically wrong about humans not inherently having a ‘better nature,’ but correct that, unless circumstances are right, that nature will not be brought to the fore. We have an instinct for fairness and to some extent for pro-social behavior, but his criticism of avoiding punishing badly behaving children resulting in badly behaved adults has some validity, though he also ignores the role of positive reinforcement. Another valid question is why we reject the use of painful punishments. This is something which we have, in the US at least, been brought up to reject (at least when administered by the government), but as the book mentions, pain is our evolved way of being told that something is wrong. Would such punishment be effective? Could it ever be morally acceptable? We might conclude that the answer to both of these questions is ‘no,’ yet the role of good fiction is (apart from its primary role, to entertain), to encourage us to think about other perspectives.

Nonetheless, one feels that this book, as Jordan Vogt-Roberts said of his own movie, that it is ‘way better than it had any right to be.’ The unrealistic aspects of it should damn the book, yet they somehow fail to do so (and in fact, the book succeeds so much in being entertaining and thought-provoking, that I did not notice them very much until a second reading).