Talk:Project Azorian

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject United States / Government (Rated C-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject United States, a collaborative effort to improve the coverage of topics relating to the United States of America on Wikipedia. If you would like to participate, please visit the project page, where you can join the ongoing discussions.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.
Taskforce icon
This article is supported by WikiProject U.S. Government (marked as Low-importance).
WikiProject Military history (Rated C-Class)
MILHIST This article is within the scope of the Military history WikiProject. If you would like to participate, please visit the project page, where you can join the project and see a list of open tasks. To use this banner, please see the full instructions.
C This article has been rated as C-Class on the quality assessment scale.

Accident during lift[edit]

I've always considered the official story of "the important half broke off for some mysterious reason and we didn't go back to get it because everyone was tired" a little fishy. I mean, they would have tracked the broken piece as it fell, which means they wouldn't even have to search for it, and its doubtful that hitting the bottom a second time did much more damage than the first time. Thats what they were there to get anyways, so why would they leave after one try? 01:36, 8 September 2007 (UTC)

The Michael White documentary now cited in this article suggests that a large portion of the piece of K-129 being recovered was lost because the "claws" that secured it suffered a catastrophic mechanical failure. (The White video shows the broken Capture Vehicle in a few places. Given the fact that AZORIAN didn't stay secret long after the first attempt, I think the CIA et al probably decided that a second attempt was infeasible from a cost and security standpoint.
Also, in the White documentary, several Lockheed and GLOMAR people said the wreck was very fragile and unstable, so it's not much of a stretch to think it could/would have crumbled after a second plunge to the depths. The Capture Vehicle very much depended on the fact that it was lifting a single, intact piece of wreckage. I doubt trying to lift thousands of little pieces would be nearly as successful or cost-effective and, if the wreck had been totally fragmented after its initial sinking, I doubt the recovery would have been attempted. Warmfuzzygrrl | Talk 18:55, 5 April 2011 (UTC)
It does seem like a fairly obvious attempt at disinformation (why tell the Soviets just how much you did learn about their technology when you can lie about it?), but on the other hand it could be true. I doubt many people really know which is the truth. Mark Grant 06:10, 13 November 2007 (UTC)

Why is it stated that maraging steel is brittle compared to other steels? Maraging steels are fantastically tough, especially for their strength range. Maraging 250 is used in fencing foils specifically for its toughness and crack-resistance; many alloy steels (like 4340) in the same strength range are brittle as glass in comparison. Maraging 250 can have a Kic fracture toughness of 75-100 ksi-in, while even Maraging 350 has a decent 40 ksi-in Kic value. You can find tougher steels, but not at that strength level. — Preceding unsigned comment added by (talk) 00:40, 11 February 2012 (UTC)

Conspiracy Theory[edit]

The article currently says: "Sewell makes the case that the particular Soviet submarine was chosen . . . ." I think "makes the case" is too strong here, because it implies that Sewell's claim is widely acknowledged and/or considered proven. I would suggest changing it to "Sewell argues . . . ." instead. Comments? Richwales 22:02, 2 November 2006 (UTC)

The Conspiracy Theory section should be spun off to a new entry on the Red Star Rogue book, with only a link from this entry. This material is non-factual and does not belong in this entry.

No sources?[edit]

How is the reader supposed to know parts of this are not just made up? -FASXFA

Very pertinent question. And then there is the citing of "sources" that are themselves among those blind guys trying to describe an elephant. Take a look at THE PARKA I EXPERIMENT (Pacific Acoustic Research Kaneohe-Alaska + Experiment), a part of LRAPP, with "preparations for the implantation of SEA SPIDER for PARKA II" as part of a temporary RDT&E effort. See if the "American SOSUS (Sea Spider) hydrophone network" actually fits together with that. Some of these authors grab project and code names out of dark little bags, string them together, mix operational with experimental and weave a tale. Hey, it sometimes sells articles or books. Palmeira (talk) 05:29, 22 December 2009 (UTC)


I am making a change to the lead-in paragraph that posits the Sewell theory as fact. I have been conversing with another contributor concerning the Golf II Soviet Submarine article trying to scrub off the Sewell theory as factual representations. Sewell's book is difficult because it cleaves into areas in which the general public is totally non-conversant, and areas where security operations are in existence to minimize the proliferation of factual data. Sewell jumps into this void and rampages through ignorance and bad information to claim discovery of a heretofore unsuspected event in our history. To Mr. Sewell: extraordinary hypotheses require extraordinary proof. Red Star Rogue is a terrible example of circular reasoning and bad logic -- at best. At worst it is merely fiction masquerading as fact.

Here is part of the discussion concerning Red Star Rogue which is taking place at the Golf II article:

Your discussion of Project Jennifer is better but still utilizes Red Star Rogue's statement that almost all of K-129 was retrieved and most of the bodies were recovered. If you read carefully how Sewell comes to this conclusion, you will realize how specious his reasoning is. He utilizes the average number of feet that the claw could be moved per hour up and down, and divides that into the total time recorded by the Glomar Explorer at the wreck site. Seems reasonable so far, but he does not provide time for repositioning the ship over the next section; no time for prep of the first down movement after arrival; no time for docking the claw and prep for sea after the last lift; no time for problems and mechanical breakdowns; no time for errors and inefficiencies (and fatigue of the crew); and most importantly, provides no time to disassemble and move-out of the moon-pool area, the section or sections that were successfully retrieved. You may remember that he discussed delays in that endeavor due to the necessity to work on a section contaminated by radioactivity. Factoring in such time requirements puts paid to the idea that the Glomar Explorer made a successful lift of five or six sections.

You mentioned the discrepancy between a normal crew of 83 and the list of 94 members on this last cruise of the K-129. While I have no experience with Soviet naval practices, I do know that U.S. national strategic resources (aircraft carriers and SSBNs) always attract extra personnel when they are about to deploy. Special observers, hardware technicians, systems designers, training specialists, and officers riding for qualifications and check-out are only a few of the reasons that extra personnel appear on board prior to and during a combat deployment. I assume that the Soviets responded in a similar manner to the unique environment of a submarine on a combat patrol to study personnel and equipment under real world conditions.

While there are always inconsistencies between information provided verbally from many sources, some of whom are often unqualified to provide such information but do so to appear knowledgeable and important, Sewell appears to grab every such inconsistency as a proof of conspiracy. Such is the danger of utilizing uncorroborated verbal data. It is a reporter's job to independently verify verbal claims; and research, where possible, into the written record. Sewell does the opposite. He bends facts to fit his theory; and apparently concocts data wherever such information is lacking. Two of the most glaring concoctions Sewell wrote to bolster his theory involve the idea that the missile launch attempt was conducted on the surface rather than submerged. Open sources on the internet document that the G-II submarine with the SS-N-5 SERB missile was a SUBMERGED launch system requiring the missile tubes to be flooded prior to launch. I suspect, but do not know for sure, that attempts to launch from a dry tube would damage the submarine to the point of endangering the safety of the ship and crew. Launching with a flooded tube on the surface is similarly unpalatable due to the high center of gravity and the ensuing unstable platform thus available for launch. Sewell apparently requires a surface launch scenario in order to bolster his theory with similarly specious claims that a U.S. infrared satellite detected two explosions at the time and near the location he proposes as the site of K-129's sinking. On pages 134, 158, 180 and elsewhere, Sewell asserts that U.S. intelligence satellite assets (specifically a NORAD satellite) provided information on the K-129 sinking by detecting the explosion of the missile propellant in tubes 1 and 2 on 7 Mar 1968. Page 158 states: "The Navy was notified by the North American Aerospace Defense Command that one of its satellites had identified and recorded an event in the Northern Pacific as having some of the characteristics of a Soviet Missile launch. A satellite sensor had recorded two massive surges of radiant energy when the missile fuel in the K-129's launch tubes one and two exploded. The camera's sensors were tuned to record the light spectrum created by the burning of specific chemicals known to be used in Soviet rocket fuel." This is an unqualified statement by Sewell, and demonstrateably false, misleading and ridiculous. A quick internet key word search will show that the NORAD launched its first missile detection satellites (IMEWS-1) in 6 Nov 1970, 20 months after Sewell claims NORAD resources recorded his theoretical missile explosion aboard the K-129. As a possible defense for Sewell, he might not have known that the infrared detection satellites of the time were under the control of the NRO (National Reconnaissance Office), NOT NORAD.

Later in the book, on page 184, Sewell conducts a discussion of intelligence satellite resources circa 1968. He discusses the CORONA PhotInt (photo intelligence) satellite as if it could have been the system that reported an explosion in the Northern Pacific in March 1968. However, this system returned its photos to earth within re-entry capsules, so photos were carefully rationed to high-interest targets. The probability is it using film on an empty northern Pacific water area is nil. He then discusses the TIROS weather satellite as a possible sensor of such an explosion (a hail-Mary pass requiring the ideas that this system 1) had such a capability; 2) was secretly connect to CIA and/or USAF customers; and 3) that this subterfuge has never come to light in the intervening 40 years). Finally he discussed the SAMOS and MIDAS systems. SAMOS satellites took pictures on film, developed the film in orbit, and transmitted TV scans of the pictures to Earth. Because the TV pictures were much blurrier than the film, SAMOS had low resolution even for its day (5–20 feet), and some authorities have claimed that SAMOS never produced useful data. Again, because each photo utilized irreplaceable satellite resources, the idea of a SAMOS satellite expending film on open ocean night photography is untenable.

MIDAS satellites transmitted launch indication via radio links; if two MIDAS satellite picked up the same indication, launch location could be calculated. Each MIDAS satellite was stationed at a much higher altitude then CORONA or SAMOS (e.g., 2170 miles vs 200 miles), from which it could see most or all of the Soviet Union at any moment. The MIDAS satellites were designed to observe Earth in the infrared band of the electromagnetic spectrum. The goal was to detect the heat radiation (infrared light) given off by missile and rocket launches; specifically ICBM launches from Russia or China. Twelve attempts to orbit MIDAS satellites were made between February, 1960, and October, 1966. Most failed, but experience with MIDAS made possible its successor, IMEWS, and the later third generation DSP satellites. The MIDAS system was rather primitive and detection of short duration signals was never attempted; nor desired (short duration signal processing would overload the downlink bandwidth and was specifically programmed out as noise). The long-duration boost phase of an ICBM (which may exceed 10 minutes), was part of the signature of an attack profile used by MIDAS signal analysis. Attempts to detect and analyze theater range missile launches (say 400 nm or less) with their much shorter boost phase became a goal of satellite IR detection only with the third generation DSP satellites in the 1980s. The idea that the explosion of two missiles on the surface of the waters of the northern Pacific in 1968 was detected by a MIDAS and forwarded via downlink as a signal of interest, overestimates the state of the art of satellite IR detection by over 20 years.

In the cases of MIDAS, IMEWS and DSP satellites, the IR signature being sought is generated by the heat of the missile exhaust plume. The IR signature is a function of heat and chemistry, so the satellites were provided with detectors which covered a band of IR frequencies centered on 2.7 microns. There is nothing in open source literature to indicate that spectrographic data was part of MIDAS' signal analysis criteria for reporting targets; thus nothing to indicate that an explosion bearing such spectrography would be processed and reported by a MIDAS. Finally, an explosion confined by missile launch tubes would have little infrared similarity to a boost phase missile exhaust plume and thus would be missed by MIDAS detectors directed specifically toward detecting plume characteristics.

It is difficult to escape the conclusion that Sewell has populated his book with spurious data and purposefully presented his "facts" without regard to normal and traditional independent verification testing. It is tempting to conclude that Sewell's efforts were never intended to represent historical reality, but rather are the most recent example of that most troublesome form of imaginative writing: fiction masquerading as fact. Gwyncann 21:38, 11 July 2007 (UTC)

The theory has additional problems. For example, United States intelligence knew China had not developed a missile for it's Golf class submarines in 1968[1]. Furthermore, China did not have a thermonuclear weapon small enough to be launched from a submarine [2]. --Work permit 06:39, 30 August 2007 (UTC)

U.S. Government Continuing Security[edit]

I think this section adds no value to the article. The section Public disclosure is more detailed and free from speculations on motive--Work permit 19:37, 13 September 2007 (UTC)

I removed the section--Work permit 18:34, 16 September 2007 (UTC)

Citation from Moscow Times is not reliable and should be removed[edit]

In this citation in the Conspiracy section, the Moscow Times writer, Gary E. Weir, appears to be manipulating information from The Silent War. If we agree that is the case, then this reference should be removed.

In discussing Sewell's book Red Star Rogue, Weir says:

"The author mentions John Pina Craven, chief scientist for the American Polaris Ballistic Missile Project in the 1960s, noting the possibility of a rogue launch. However, an examination of Craven's book "The Silent War" shows that he immediately declared it unlikely and never qualified his conclusion."

This appears to be a deliberate misrepresentation. Here are quotes from The Silent War (ISBN 0-684-87213-7, Simon & Schuster, 1st edition, page 217):

"A highly probable scenario is that the submarine was a rogue and that it was probably a Golf."
"To put it succinctly, there existed a possibility, small though it might be, that the skipper of this rogue submarine was attempting to launch or had actually launched a ballistic missile with a live warhead in the direction of Hawaii. There is also a small possibility that this launch attempt doomed the sub."

These aren't passing remarks on Craven's part, the whole of Chapter 15 discusses the technical and political situation. To read that Craven "immediately declared it unlikely" and "never qualified his conclusion", is a flat out lie (unless, of course, Weir never read The Silent War, as he claims).

My opinion, having read The Silent War, Blind Man's Bluff and Red Star Rogue, is that a lot of people inside and outside the intelligence community are having just a tremendous amount of fun, story-telling and obscuring the facts (or making them up). This confusion, however, does not mean that Wikipedia should quote a questionable source such as Weir.

Alpha Ralpha Boulevard (talk) 21:26, 4 July 2008 (UTC)

Technical Aspects[edit]

I would be interested in clarifications concerning the technical aspects: Why was it necessary to stabilize a platform above the site of the accident? Why wasn't the boat hauled to the surface with cables, instead of lifting it with a claw attached to pipes? Why weren't lifting bodies hooked to the wreck to float it to the surface? The solutions chosen seem more complex than necessary - clarifying why things had to be so complicated would be of interest.-- (talk) 21:48, 24 January 2010 (UTC)

Change title of article?[edit]

In light of the recently declassified CIA article on Operation AZORIAN, where it is stated "[t]his article describes how the Glomar project -- code-named AZORIAN, not "JENNIFER" as stated in the press -- came about, how it was managed and conducted, and to what extent it met its goal," should the title of this article be changed? WikiJohnS (talk) 20:02, 13 February 2010 (UTC)

A redirect from AZORIAN might be in order. For now, few people will know it by the "new" name? Alpha Ralpha Boulevard (talk) 23:29, 13 February 2010 (UTC)
I think it will eventually be the other way around... Going forward, we will no doubt use the CIA's real name of AZORIAN, not the cover name. I would propose sometime in the next few months to make this change, but the article will need a fresh rewrite to discuss this. Note well that the words "Project AZORIAN" appear prominently in the CIA released document and this should be the title of our article. This is the MOST RELIABLE source we now have on the subject. I like to saw logs! (talk) 23:46, 13 February 2010 (UTC)
I changed the title, added that the code name "Jennifer" assigned by the press was wrong (including a reference), and did a search-and-replace from Jennifer to Azorian in the article. I hope that's fine with everyone. --Enemenemu (talk) 00:36, 14 February 2010 (UTC)
Well, that's faster and less encyclopedic than I would have done. :) There are sections which cover what the press did in the 70s... all of that should use the "Project Jennifer" name of contemporary reporting in the press. It needs to be explained, it needs to be discussed, and it needs to specify Project Jennifer a lot... for search engine friendliness and for the fact that 35 years of usage makes Jennifer more recognizable. The article title is one thing, but the body of the article should use familiar terms. AZORIAN hasn't appeared much until the last year or two publicly. Also, JENNIFER was a code name for a security system associated with AZORIAN, so its nature should be discussed at any rate!

I like to saw logs! (talk) 06:44, 14 February 2010 (UTC)

The confusion between the security compartment code name JENNIFER and the project name AZORIAN is now mentioned in section Burglary and press reports on Project Jennifer. Do you think that this is sufficient, or should one go into more details (like mention other examples - CORONA and KEYHOLE come to mind)? --Enemenemu (talk) 20:58, 18 February 2010 (UTC)

Project MATADOR[edit]

Anyone notice that there is a Project MATADOR mentioned on the CIA document? Since the only time I see it mentioned it is slashed with AZORIAN, I would assume it was the phase which came after AZORIAN, so Operation Matador might have been the actual recovery of intelligence information from the AZORIAN project. Obviously, the raised part of the ship went somewhere for extracting that information and to decontaminate and analyze the hulk. I like to saw logs! (talk) 07:59, 14 February 2010 (UTC)

Operation Matador (1975) suggests that this was a plan to return for a 2nd recovery attempt. -- Enemenemu (talk) 22:15, 17 April 2010 (UTC)

Project Matador - several factors probably added up to it's being cancelled. Multiple (if inaccurate) press reports about Jennifer (aka Project Azorian), Russion diplomatic messages and statements about going public, unofficial statements from a Russian diplomat "if you go back to that site we go to war", memorandum to the president from Kissinger, the report about the Soviet tug MB-11 in the vicinity of where the K-129 sank in March 1975 -- all added up to its cancellation. [1]

Project Matador started the day after Project Azorian ended. Matador was supposed to lift the remaining portion of the K-129, however, due to the veil of secrecy covering Azorian being lifted, Matador was ultimately scrapped. There was a break-in in LA that resulted in documents linking the HGO to the CIA which blew the HGO's cover. Coupled with the facts that the New York Times broke the story and that the Soviets were aware of the recovery, made any attempt of going back to the site impossible without being an open act of war. — Preceding unsigned comment added by (talk) 08:51, 5 December 2012 (UTC)

Level of Success[edit]

As to the level of success, I think the following redacted statement can only be filled in something like this:

On 8 August 1974 [redacted] that submarine was brought to the surface in [redacted] a recovery system designed and developed specifically

for that mission.

On 8 August 1974, a portion of that submarine was brought to the surface in the Lockheed Capture Vehicle, a recovery system designed and developed specifically for that mission.

The original Secret document likely contained a qualifier to state which portion of the submarine was recovered. All of this is just from reading the first page. I like to saw logs! (talk) 07:59, 14 February 2010 (UTC)

Distance to Hawaii and the International Date Line story[edit]

The released document shows 1,560 miles from Hawaii. I just did a test from a beach on Oahu to 40N/180W and the distance I got was within 10 miles of that figure. I tried the coordinates of Waimea Bay, Hawaii which is near the NW side of Oahu, and got 1570 nautical miles. Note: statute miles may have been implied, but not stated, but nautical miles seems appropriate to me and fits the data I just tried. I like to saw logs! (talk) 13:19, 15 February 2010 (UTC)

The document also states that the distance from Long Beach, CA to the recovery side is 3,008 nm (page 37, top, 1,888 nm covered, 1,120 nm to go on June 27). The intersection of the two circles should yield a pretty accurate position of the recovery site. For starters, I entered 3008nm@LAX, 1560nm@HNL on (Great Circle Mapper) --Enemenemu (talk) 01:38, 17 February 2010 (UTC)
I'm impressed by your first estimate. Using the coordinates of Long Beach, CA, and Pearl Harbor, HI, the intersection of the two circles with radius 3,008nm and 1,560nm, respectively, is at 38°6'N and 178°58'E, i.e. just beyond the international date line. Of course we don't know if HGE sailed "straight" on a great circle from Long Beach to the location of K-129, hence the true location of the recovery site might be even somewhat closer to the international date line. --Enemenemu (talk) 22:18, 17 February 2010 (UTC)
I changed it to nmi (and added a conversion to km). The document doesn't say what kind of miles but it's pretty clear (I think) that it's nautical miles, just based on the 40N/180W position. Rees11 (talk) 18:42, 18 February 2010 (UTC)
The International Atomic Energy Agency (IAEA) quotes the location as 40° 06’ N and 179° 57’ (W or E not indicated, though that close to the date line and at 40°N, this only corresponds to a positional uncertainty of +-4 km in E-W direction), and "1230 miles from Kamchatka". In principle, they should know, though they also quote the depth as 6000m, which seems to be beyond the depth reachable by HGE: (see Appendix I.3) --Enemenemu (talk) 21:19, 18 February 2010 (UTC)
By assuming that 1230 miles from Kamchatka refers to K-129's home port Petropavlovsk, the 1,560 nm distance NW of Pearl Harbor, HI, yields 38°4'N and 178°56'E for the accident (recovery) site. This is in good agreement with the value of 38°6'N and 178°58'E derived by computing the intersection of the two circles with radius 3,008nm and 1,560nm, centered on Long Beach, CA, and Pearl Harbor, HI, respectively. Computing the intersection of the circles for Petropavlovsk and Long Beach, CA, yields quite similar coordinates. This strongly supports Rees11's conclusion that all distances are in nautical miles. The coordinates of the recovery site are then 38°5'N and 178°57'E with an uncertainty of ±1' (±1.9 km in NW, ±1.4 km in EW direction) --Enemenemu (talk) 22:26, 18 February 2010 (UTC)

Fictional Accounts[edit]

Recently I came into possession of a 1977 novel entitled Fireplay. The authors name is William Wingate. The novel is clearly based on the 1975 press reports and covers the successful attempt to raise a Soviet submarine from the ocean bottom by the Pacific Klondike (eg the Hughes Glomar Explorer). In Wingates' novel the entire submarine is raised using a barge mounted grab guided by men in a bathyscaphe. Once the submarine is bought to the surface it is found to be an empty shell, part of an elaborate scheme to discredit the CIA.

I've located another fictional reference this one from 1978. In the novel Ice by James Follett the Soviet Union dispatches a Delta II class submarine into the South Atlantic to search for a missing British submarine. The Delta II is destroyed when it collides with a submerged iceberg while transmitting a position report.

The Soviet Navy is then ordered to mobilize portions of the Black Sea fleet to sortie on a salvage mission designed to forestall any American attempt to recover the submarine. A footnote specifically refers to Project Jennifer (Azorian). It would seem that for a few years after the event it was believed that the US recovered the entire submarine.

I've posted a summary of these two novels to the appropriate section of the article, but it does raise the question. How many other fictional accounts are there between 1975 and 2006 and what changes in the view of just what was recovered do they reflect. Given what seems to be emerging, this may be a fruitful line of inquiry. Anyone willing to join me in finding out?Graham1973 (talk) 15:48, 25 December 2010 (UTC)

You should read other published non-fiction accounts to understand how complicated this situation is. Given that everyone on the sub died, and that now many of the people in USSR who were involved are also dead, particularly those that planned the operation, and that it's still regarded as highly secret, the chances of you, yourself, being able to run down anything new are remote to nil.
Something that's not commonly, fully appreciated by readers is that a novelist's primary job is to be entertaining. Readers assume sci-fi novelists are experts working from esoteric knowledge. In my experience from the expert and the writer side, barring very few people such as Tom Clancy (who has a huge writing staff), writers gather a basic knowledge, then make up anything that doesn't fit in the story line. (talk) 07:36, 9 March 2011 (UTC)
That has been appreciated. I just think that the media interaction in this case (including novels) to be particularly interesting. Fireplay is a case in point, the author has the CV towed to the salvage site inside a barge and then lowered from the barge to the target. Obviously this is a reaction to media reports about the Lockheed barge.Graham1973 (talk) 22:54, 10 April 2011 (UTC)

Use of air?[edit]

Look, this is probably not the right place to ask this question, but lacking any other, here goes anyway. After watching a doco on the project, a question that comes to mind is, why was the project designed from the beginning as simply a dead-weight lift? Why not use air - in some kind of air bags, or rigid or semi-rigid containers - to take some of the weight off that oh-so fragile 5 km long pipe string? (talk) 11:44, 11 October 2011 (UTC)

The lifting pipe used was made to look like normal drill pipe but was specially made out of extremely strong steel by the Hughes Tool Company. Source Empire: The Life, Legend, and Madness of Howard Hughes by Donald Bartlett and James Steel. As to what page id have to go back and look but the Hughes company was contracted for the special pipe. Also i think using lift bags would be an enormous undertaking since i have heard of them being used at depths of ~600ft in free diving competitions ( for the surface returns) but no more then a few hundred in actual work environments.(CaptianNemo (talk) 07:38, 9 November 2011 (UTC))

Documentary[edit] — Preceding unsigned comment added by (talk) 17:23, 24 December 2012 (UTC)

  1. ^ Polmar, Norman; White, Michael (2010). Project Azorian : the CIA and the raising of the K-129 (null ed.). Annapolis, Md.: Naval Institute Press. ISBN 978-1-59114-690-2.