Military and naval war gaming has had an old and fairly honorable history. For one thing, it was always fun. The British, for example, had enough ships to divide their fleet in two and have one pretend to fight the other. The ships would put to sea, neither fleet knowing the exact position of the other, and would send out scouts to find each other. There was a ship with umpires who knew everyone's position, and they could re-direct the fleets if they should go off in opposite directions.

Once the action was "joined," the ships might fire specially modified shells of the high explosive variety (not designed to pierce armor) with the explosive removed. Such shells could still do damage, and would kill or remove a limb from anyone they hit solidly. But each ship would fire single shells instead of salvos, and, at a range of eight miles or so, the odds were slim. Officers could therefore stand gallantly in the open on their bridges as shells screamed toward them. The rare hits had to be reported, and the umpires would calculate the damage that would have resulted from a full salvo of armor-piercing shells with explosives.

At other times, no shots were fired at all. The captains and admirals had to be content to maneuver their ships and imagine the rest. In that case, the umpires were provided with tables of probability which told them how likely it would be for a salvo fired at a given distance to straddle its target, and how likely a torpedo fired from a given position would be to hit its target. The umpires would then throw dice and determine hits. They had other tables for probable damage resulting from the various sorts of hits, and would throw dice for that, too. The results having been determined, they would be signalled to the various ships and admirals. Some ships would have to slow down with engine room or hull damage, or stop firing certain guns. Others would be "sunk," and would have to slink away in ignomy.

There would be a winner and a loser, and perhaps even a promotion for the winner. At the turn of the twentieth century, a British admiral with the improbable title of Prince Louis of Battenburg was the undisputed champion.

Army war games weren't quite so much fun, particulary for the private soldier. By 1957, an infantryman might have a five pound bag of flour dropped on his head from a helicopter, and be told to lie down because he was dead. The games had gotten quite rough, and people were killed on occasion

Whether fun or not, the games were probably good training for officers of middle or high rank, perhaps even for junior officers. But their predictive value was almost nil. It might take days to play out the game, and the outcome reflected, more than anything, accidental factors, chief among them the throws of the umpires' dice.

Computerized war games, by contrast, could be played hundreds or thousands of times with any given set of starting assumptions. That took out the accidental factor. They could also be gradually adjusted so that they would predict, with a high probability, events that had actually taken place in the past. Since some of those events might themselves have been "accidental" relative to the known background factors, this wasn't a foolproof check. But it was much better than nothing, and the game that accounted for most of the past could be projected into the future.

One day in the library, Tom Williams was looking through the section on war games and simulations. There, being, as yet, no journals devoted exclusively to such things, there was only a stack of mimeographed papers. Most of them concerned computerized war games of various sorts, but there was one title that particularly took the eye: "Notes toward a Simulation of the Functional Dynamics of a Religious Congregation." The author was one Oswaldo Entrecote at the Episcopal Theological Seminary, and Tom's curiosity got the better of him.

The author began with a discussion of the ordinary naval war simulations, but he then suggested,

"There are within the typical religious congregation opponents as bitter as any to be found on the high seas. Their goals are often in direct conflict with one another, and, while their methods of reaching their goals and denying them to the enemy are not as well-known and standardized, they may be just as effective."

In the Episcopelian congregation being studied, one clique consisted of the older, more conservative, people who provided most of the money for running the church. The other group consisted of more liberal people, mostly in their thirties and forties, who were particularly concerned to make the church attractive to teen-agers. These teen-agers were, in many cases, their own offspring. The minister, instinctively one of the liberals, nevertheless had to pay attention to the older people who, among other things, determined his own salary.

On one occasion, the minister was persuaded by the liberals to allow the teen-agers to fix the church basement up for Saturday night dances, either with records or a local amateur band. These dances were quite successful, and, being properly chaperoned, there seemed to be no posssible objection. Entrecote commented,

"The Saturday night parties represented a successful move by the liberal group in the direction of extending its control over a part of the church property for a fixed period of time. It could only be expected that there would be a riposte from the enemy."

It wasn't long in coming. The conservatives took up a collection among themselves and raised a considerable sum of money for the renovation of the church basement. It was presented to the minister with the words,

"This will create a much more pleasant atmosphere for the Bible Study groups. Of course, we have to take care of the new furniture, and we can't have any rowdy parties in there on Saturday nights."

Entrecote commented,

"Such suggestions from the pillars of the church could hardly be ignored by this minister, or any minister. One familiar with everyday internecine religious warfare can divide these various moves and counter-moves into categories in much the way that the student of naval warfare can categorize shell-fire in terms of caliber, range, type of shell, and so on. That having been done, one can prepare tables giving the probability of the success of a religious initiative in much the way that one can give tables of probability for a straddle by an eight-shot salvo from various kinds of guns at various ranges."

It seemed that the Reverend Mr. Entrecote didn't himself have access to a computer. He ended with the pious wish that someone would program his outlined simulation, and thus make possible a better understanding of religious conflict. Tom was laughing when he finished the article, but it did seem that Entrecote's methods might have applications well outside of religion.

Tom stopped by the lounge in the main building for coffee, and, as often, he came upon Mac Hollins. Tom described the paper he had just read, and concluded,

"So it looks as if Khruschev is in the position of the minister trying to push the positions of some of his factions against those of the others."

Hollins seemed quite struck with the idea and asked,

"You say this man has a simulation of his religious group?"

"He doesn't have a computer, but he's doing what he can. If he can simulate the Episcopelians, there's no reason why we can't simulate the conflicts within the Soviet government with respect to their perceptions of our willingness to take risks."

"I bet their hawks and conservatives will turn out to be the people who think we're unwilling to take risks. Their doves and liberals will be the people who take Mr. Dulles at his word."

"And, of course, we could simulate our own government while we're at it. Ideally, we'd have a simulation that could be applied to any government."

"You know, Tom, this introduces a new element into the situation. We before were speaking of ways of determining when a rising country has reached the point where it becomes rational for it to take on and attempt to knock off the top dog."

"And, in the case of the Soviet Union, we were aiming at predicting a date plus or minus a few months."

"Yes. But, when the decision to attack is finally taken, it'll probably be made by what amounts to a committee. And that'll probably be because a few fence-sitters have decided to join the hawks. It won't just be a matter of a majority vote, but of the hawk party gaining so much momentum that the doves think it would be dangerous to oppose them."

"If the attack succeeds, no one wants to be remembered as the one who opposed it."

"Exactly. And if the attack is made and fails, nothing matters anyway."

"So we need one simulation for the Soviet military-industrial complex, and another for their top-level decision-making process."

"Yes, to the extent that one or both those simulations are the kinds of things that we can reasonably undertake, given the limitations of our computer."

Hollins' secretary called him away to the telephone, but his enthusiasm was unmistakeable. With luck, there would be at least one new project on the computer, one run by the director himself. With more luck, Tom would be the member of Complab who would be working with him. Among other things, that would bring the two separate components of his work-day back together.

Returning to the library, Tom tried to make a beginning. He assumed that, in both the Soviet and American governments, there would be a group of a dozen or so men who would make the critical decisions concerning war and peace. They would naturally divide themselves into cliques, sometimes with overlapping membership. As with the religious group, it should be possible to predict in advance the general position that each clique would take in response to challenges, in this case, international crises.

The next step was to sketch out on paper the various possible roles with their associated grand strategies. Entrecote had had conservatives, liberals, and the minister- conciliator. While the president might well occupy the position of conciliator, the other roles wouldn't just be for liberals and conservatives in any ordinary political or social sense.

It seemed likely that no one could reach that level unless he had demonstrated his commitment to making his country the dominant one in the world at the expense of its only real competitor. But, of course, there would be fundamental differences as to means. For a start, there were those who:

(1) thought they could win a nuclear exchange, and wanted to attack.

(2) thought the other side would probably strike sooner or later, and wanted to strike first.

(3) thought the other side wouldn't push its nuclear button except in response to nuclear attack, and wanted to take advantage of that fact.

The people favoring the first strategy, which Tom christened "Strike to Win (StW)," thought that their side had an overwhelming nuclear advantage, perhaps enough to eliminate any return strike at all. They would typically favor striking quickly before the perceived advantage was lost.

The second strategy, "Beat to the Draw (BttD)", didn't necessarily assume that a nuclear exchange was winnable. But, still, the country which struck first would end up much better off than the one that did not. This position was consistent with the more radical StW, but not with the third strategy. The BttD man wanted to do nothing to provoke the enemy into striking first.

The third strategy, "Old-Fashioned Adventurism (OFA)," believed in taking advantage of non-nuclear military strength and threats to gain territory and various other geo-political goals in the confidence that the other side wouldn't resort to nukes unless nuked.

All three of these kinds of people were dangerous. The first kind wanted to initiate nuclear war, the second wanted to set up a nuclear hair trigger, and the third wanted to do things short of nuking that might well provoke nuclear attack.

As far as Tom's project went, there was an important difference between the kinds of hawks with respect to their estimates of enemy intentions. StW hawks might claim that the enemy was on the point of attacking in order to justify their own position, but that would be, at most, window dressing. They really wanted to attack regardless.

The Bttd hawk, on the other hand, was much more concerned about enemy intentions. He thought that the enemy was never far from attacking, and was anxious not to provoke an attack. But he was also quick to interpret the available data as being suggestive of imminent attack.

By contrast, the Adventurist was the one who thought that he could get away with almost anything without provoking a nuclear attack. Correspondingly, he tended to downplay the same data that led the Bttd hawk to expect an attack.

It struck Tom as noteworthy that hawks who, in practice, sounded almost equally bellicose could be operating on almost opposite assumptions about enemy intentions.

If these people were all hawks, where were the doves? The Doves, it seemed, were those who thought that nuclear war was neither winnable nor inevitable. They believed that the other side might press its button if provoked, and that it was prudent not to press Adventurism too far. This position seemed only sensible to Tom, but, he remembered, a Dove government could stay in power only if it had a plausible strategy for winning the Top Nation competition.

Given the Dove premises, the competition could be won only with economic power, and it had to be part of that position that the country would prevail in the ecomonic race while not losing the military race.

In either government, the doves would be surrounded by the three kinds of hawks, with Eisenhower and Khruschev trying to mediate. Entrecote had characterized the various kinds of religious moves, and the factors that would advance or frustrate them, but Tom would have to categorize the basically political moves that the four strategists might make, and the factors that would determine their success or failure in the critical committee meetings.

When Tom returned to Complab, he had hardly come past the guard's station when he saw Goldstein. He was about to tell him what had happened when Goldstein said,

"I've been hearing all these rumors about you being closeted with the director. What are you getting up to over there?"

There was really a rather unpleasant tinge to the question, with even a hint of jealousy. Suddenly, Tom wondered if he should tell Goldstein about the project. Most of what he was doing with Hollins was hardly secret at all. It was the sort of thing that might end up by being published in a journal like Foreign Affairs. On the other hand, Goldstein had an extraordinary appetite for secrets, almost any secrets. His obsessive curiosity seemed to extend to everything. As he himself had once said,

"I start by discovering how much money people have. After that, I find out exactly what their work consists in, and who they like and don't like. I end by discovering what their marriage is like. People sometimes refuse to tell me some of these things, but it does no good. I find out in the end."

It was an appetite so unbalanced, and so unhealthy, that one was tempted to deny him the fix, even if it happened to be information that wasn't secret at all. And then, too, in an organization which did have such secrets as the total size and capability of the United States nuclear arsenal and its probable targetting, the presence of someone like Goldstein was a worry.

Even if one had Top Secret clearance, one didn't have a right to know all the secrets. There was compartmentalization between projects, and one had to be careful not to give or seek information across the boundaries. Goldstein's position already gave him access to all the most important secrets, and it was rather unnerving to watch him try to get the few remaining ones out of the people he believed to possess them. Tom tried to be vague, but, of course, that did no good. He eventually told Goldstein pretty much what he had intended to tell him in the first place. The little talk ended when one of Goldstein's assistants called to him that they were on the machine.

Tom was acutely aware that the two halves of his job hadn't yet come together, and he sat down with his latest program. It was meant to simulate naval warfare about 1880, when the ships were steam-driven but there were no aircraft or submarines. It used random numbers to substitute for the tossing of dice, but, unaccountably, the same supposedly random numbers were generated each time the program ran. After a half hour or so, Tom took his problem in to Bruce.

It usually took Bruce only ten or twenty seconds to solve one of Tom's problems, but this one was harder. They were seated at his desk with the door open to catch the air- conditioning of the machine, when they suddenly heard a scream. It seemed to come from the other side of the machine, somewhere near the control position. It wasn't a woman's scream, nor was it one of fear or the ordinary kind of hysteria. It was a scream of anger, and, after a moment in which he sat transfixed, Tom realized that it had been produced by Goldstein. Bruce gave a little groan and raised his hand to smooth back the thinning hair on his forehead. He said only,

"Goldstein and Jacky must be at it again."

By this time, Tom was aware of the system that had brought Goldstein and Jacky together at the control position of the computer. Whenever anyone at DRI had an approved project that made use of the computer, he or she would be paired with a member of Complab who would be primarily responsible for writing the program and getting it executued. Most of the senior investigators outside Complab who were likely to have such a project themselves had some programming skill. However, Complab didn't wish to let non-members loose on the machine without supervision. In all reasonableness, it was so hard to get the bugs out of a program that almost anyone required some assistance. The system worked well enough on the whole, and, as Sam Harris often said,

"We can save machine time by spotting problems in a program before it's run at all."

The only difficulty was that Goldstein was perhaps the second best programmer at DRI, right after Bruce. He didn't often need help from anyone, and, on the rare occasions on which he got stuck, he came to Bruce. However, the rule said that he had to be permanently paired with someone, and, since Goldstein was the main developer of the Red v. Blue model, it was an all-consuming task. Bruce, on the other hand, was primarily responsible for the programs which ran the computer itself. He was also at work developing an alternative to FORTRAN, and, not content even with those things, he had developed a number of simulations entirely on his own. It had therefore been necessary to assign someone else to Goldstein.

The person should have been Janet. Goldstein liked and respected her, and she was close to his own level. Sam Harris, however, wasn't a very imaginative man, at least as far as people were concerned. He assigned Goldstein, not his best available person, but his most experienced person. That was Jacky.

At that very moment, knowing that Goldstein's anger was most probably directed at Jacky, Tom remembered Sid's saying,

"It's a terrible combination. Jacky does have something of a mystique that he thinks he has to maintain. Goldstein's sensitive enough to realize it, but not sympathetic enough to allow for it."

There were times when Sid would engage in her own brand of diplomatic understatement. In fact, it was hard to imagine anyone better than Goldstein at puncturing mystiques or anyone more likely to exploit the opportunity.

There was an interval of something like five seconds between the scream and the pounding of Goldstein's fist on Sam Harris' door. Sam wasn't in, and, since Bruce was next in command, Goldstein, closely followed by Jacky, came charging into Bruce's office. To Tom's consternation, Goldstein looked as much at him as at Bruce as he pointed at Jacky and said,

"He's useless. He keeps screwing up my programs. Get rid of him."

Jacky was breathing hard, his face red, and it looked for a moment as if he might strike Goldstein. But he didn't. He just stood there steaming without saying anything.

Tom was sure that Goldstein would have said the same thing to Sam Harris. He was beyond tempering his words to his audience. Sam was a man with a full sense of himself who could be rather forbidding. He would certainly have strongly resisted, and indeed resented, a suggestion from a member of a different group that he "get rid" of one of his own people. It seemed unlikely that he would have made allowances for the highly charged atmosphere and the passions of the participants.

Bruce was a quite different sort of man, much milder in manner, and much slower to anger. He immediately realized what had happened, as did Tom. It was within Jacky's prerogatives to modify a program in order to improve it. Moreover, Jacky liked to make those modifications even after a program was already loaded into memory. It helped cultivate the image of a programmer so skilled that he could see what was happening even while a program was running and eliminate the faults.

It might have worked on very simple programs, but surely not on anything Goldstein would have written. Evidently, Jacky's modifications had produced chaos to the extent that he had nothing to say for himself. Bruce said quietly,

"Let's see the program."

Goldstein was holding a sheaf of coding forms in his hand, and, after a moment's hesitation, he handed them to Bruce, pointed out a place, and, with great contempt, wrote in the margin the change Jacky had made on the machine. What happened next was quite extraordinary. Bruce managed to see at a glance what was going on in the middle of a complex program, and he wrote, also in the margin, a series of instructions that saved two steps. He said,

"I think this is what Jacky was trying to do. It'll run this way."

It wasn't, of course, what Jacky had been trying to do. Jacky wouldn't have been capable of anything of the sort, and Goldstein knew it. Bruce, really, must have known it too. But Goldstein, for the first time in Tom's experience, had nothing to say. There was nothing to do but go back to the machine and try it. Jacky, whatever his emotions might have been, followed him.

After they had left, Bruce gave Tom a wry look and said,

"People get temperamental sometimes."

Tom was still concerned because he was aware that, in the look Goldstein had given him, he expected Tom to be his ally against Jacky, a member of Tom's own group. Jacky had noticed it, too. Tom was sure of it. It was absurd, of course. Goldstein had been asking both Bruce and himself to get rid of Jacky, as if such things were decided by vote. Tom wanted to somehow discuss this with Bruce, and the best he could think to say was,

"Jacky didn't say anything at all."

"He's not a terribly articulate man. I hope he isn't making plans to murder Goldstein. Although I suppose that would solve some of our problems."

Later, when Tom passed by Jacky's office, he heard Ted Blitz say,

"If that kike gives me any shit, I'll strangle him."

In normal circumstances, Tom would have been delighted to watch the repercussions of this incident unfold, and to gossip with anyone who seemed so inclined about it. Unfortunately, it was essential for him to keep out of the way. He knew that Goldstein would want to complain to him about Jacky, and Tom suspected that he might do it rather publicly. It would endanger Tom's relations with Complab to even be seen talking with him, much less responding sympathetically. In the circumstances, he felt that he could talk openly only with Sid. He was then happy, a few minutes later, to find her alone in the tape punching room, a place where the programmers weren't likely to go. Closing the door behind him, he asked, more in the tone of one who is worried than in that of one who wants to gossip,

"What do you think of all this?"

She, of course, knew what he was talking about.

"Well, this is the last, and most serious, in a series of incidents."

"In my work with Hollins, we try to see which series of incidents between countries will lead to war. I'm not exactly sure what war within DRI would be like."

"We may already have it. It's hard for you and me because we've been friendly with Goldstein, and there's going to be pressure within our group to side against him."

"Yes. Jacky's bunch already hates him, and I'm sure Bruce is tired of the fighting. What will Sam do?"

"Sam knows that Jacky and Goldstein don't get along, but he said he doesn't want to take Jacky off the project. He's been on it for a full year, and that work would be wasted. Plus, someone else would have a lot of catch-up to do. Sam also can't discipline Goldstein because he isn't in Complab. I imagine he'll go to General Edwards."

Lieutenant General Albert Edwards USA (ret.) was in charge of the main group which included both Goldstein's unit and Complab. Tom had met him only briefly, and had no idea what he might do. Before he could ask Sid, she said,

"There's a simple solution for you, Tom. All you have to do is spend the next few days working for Hollins over in the main building, or in the library over there. It won't be so easy for me, and still less so for Janet."

"That doesn't sound very honorable, but I'll do it. Will you tell me all the gossip after the dust settles and I come back?"

Sid looked at him searchingly and said,

"But you're such a serious young man, Tom. If something unseemly happens, I'm sure you won't want to know about it."

With that, she got up and peeked out the door. She then said,

"The coast is clear, get moving."

Tom did have to go past the computer to check out with the guard, but he wasn't intercepted. He then waited for the shuttle around the corner. There was no shade, and the pavement radiated tremendous heat, but it was better than being button-holed by any of the principals.

Tom was just going into the main building when he met Hollins coming out. Hollins said,

"Hi Tom, I bet you're taking refuge here from the battle going on in Bethesda."

"I guess I am. I didn't realize things got to you so quickly."

"General Edwards called me. There's no doubt that Goldstein can be extremely obnoxious and difficult, not to say arrogant. But we need him. That's all there is to it."

Tom was happy to agree to that proposition as he watched Hollins move jauntily off to his car. It appeared that the director might actually enjoy little crises in his organization.