Building the 1 ATF Contact Database

Robert HallAbout3 Comments

By Bob Hall and Andrew Ross

This website would not be possible were it not for the building of a massive database of 1st Australian Task Force battles and contacts: ‘The 1ATF Contact Database 1966-1971’. Constructed over many years by Dr Andrew Ross, his efforts were initially aimed at developing a database for the Operations Research analysis of Australian combat operations in Vietnam, but as that study was completed, other uses for the database emerged. Together with other researchers in our team, we saw the opportunity of using the database to create this unique website, to inform Australians and New Zealanders about the service of their troops in Vietnam. The collection and organisation of combat data has long fascinated operations analysts because, as war becomes more complex and the costs of waging it in terms of treasure, lives and national prestige continue to climb, analysts have searched for better ways of grappling with its complexities.

Contact Information in South Vietnam

In attempting to better understand the seemingly chaotic counterinsurgency battlefield of Vietnam, US Secretary of Defense Robert McNamara, drawing on his background in statistical analysis of complex problems, advised MACV that ‘everything that was measurable should in fact be measured’.[1]  This admonition resulted in a massive data collection effort. But too little effort was directed at assessing the relevance of the data to identifying trends and determining their cause. Yet many insightful Operations Research studies were completed, analysing the pattern of the war in terms of military operations, pacification programs, civilian casualties, village and hamlet security, economic development, and much more. Thomas C. Thayer did much of this work in his series of Southeast Asia Analysis Reports[2] produced for the US Assistant Secretary of Defense (Systems Analysis), Alain Enthoven, and later summarised in Thayer’s book War without Fronts: The American Experience in Vietnam.[3] Informative as his publications are in many aspects, Thayer is not definitive in regard to the progress of US Army combat operations. He and other analysts knew that the quality of their research was undermined by the sometimes poor quality of their data.

A case in point was the data pertaining to enemy body count, which, in the absence of alternatives such as an advancing front line, ‘quickly became the criterion for measuring success in Vietnam’.[4] This created a situation in which US commanders found themselves under pressure to exaggerate their enemy body count figures. Many succumbed. According to Lewy, the US Department of Defense estimated that USMACV body count figures were overstated by about 30%.[5] Other credible analysts say that the US body count was exaggerated by more than 100%.[6] More recent research by Professor Bernd Greiner, and Nick Turse, has used the US Army’s own records to show, amongst other things, that the US enemy body count was also inflated by tens of thousands of civilians who were largely innocent victims of reckless and indiscriminate use of firepower, so-called collateral damage, where little or no care was taken to protect the status of civilians.[7]

This corruption of the raw data now makes it difficult to make reliable judgements about the effectiveness of US military forces against the Viet Cong and People’s Army of Viet Nam (VC/PAVN), particularly in small unit contacts.

The Australian Task Force played a relatively small role in the Vietnam War and it tended to conduct operations differently to the US Army. The major focus of US Army and US Marine Corps ground combat operations were the provinces of I CTZ, the two northernmost provinces of II CTZ and the western provinces of III CTZ adjacent to the Vietnam/Cambodia border. These were the areas of greatest strategic importance and of South Vietnam’s 44 provinces, the five located in these areas accounted for 33% of allied, mainly American and South Vietnamese, combat deaths.[8] Thayer observes that ‘the war in these five provinces was almost four times as intense as it was in the other thirty-nine provinces’.[9] The Australian AO was well removed from these hot-spots of combat action. Phuoc Tuy Province where 1ATF was based, was, relatively speaking, a backwater of the war, and one much more suited to the conduct of classic counterinsurgency than the five provinces mentioned above. But, in addition to the more permissive combat environment, 1ATF tended to adopt a different tactical approach guided by the Australian Army’s previous service in the Malayan Emergency and Confrontation, and by its doctrine for the conduct of counter-revolutionary warfare (CRW).[10]

Drawing on its earlier counterinsurgency experience, 1ATF adopted a standard process for recording relevant data about every encounter with the enemy. Each time a 1ATF patrol encountered an enemy force and an exchange of small arms fire took place, a ‘contact’ was deemed to have occurred. As soon as the contact had finished, the patrol commander was required to send a report by radio to his higher headquarters. This report recorded the basic facts of the incident – the identity of the unit making contact; the date, time and place of contact; estimated enemy strength; a description of the enemy (including their dress, weapons and battle drills); casualties to both sides, weapons and equipment captured; direction of enemy movement, and a short description of the event, including whether the incident was initiated by friendly forces or the enemy.[11]

Later, when the unit returned to Nui Dat after the operation, a full written report, complying with the format laid down in 1ATF SOPs and based on the initial radio report, was completed by the patrol commander. He also drew on notes about the incident recorded in his field notebook and on discussions with others involved in the incident. In many cases these written reports recorded additional information such as the number of rounds fired according to weapon type, an assessment of the types of weapons that had caused the casualties, the weapons and other items captured, the use, if any, of indirect fire support from artillery, mortars or air power and other details.  Sometimes they included a sketch map showing salient features of the contact, often including the locations of individual soldiers and weapons such as the patrol’s machineguns and claymore mines, the location and direction of movement of the enemy and other details. The written reports also included a brief critique of the contact and the patrol commander’s performance by more senior officers. The patrol commander, the platoon commander, and the company commander signed the written reports before sending them to Battalion Headquarters for inclusion in the Commanding Officer’s report to the Task Force commander.[12] Other types of combat incidents such as mine incidents, hostile mortar fire, or the finding of unoccupied bunker systems or caches, had their own similarly detailed reporting requirements.

Over the course of the war there were some minor changes to the prompts in the report proforma. These changes were mainly aimed at removing ambiguity in the questions, thereby improving the quality of the data captured. But despite some small changes to the wording of particular questions, the information sought by the contact report remained remarkably stable throughout the war.[13]

The combat data captured in these 1ATF reports, although, no doubt, containing errors and omissions, tends to be of a much higher accuracy than is found in US Army reports. The answers to many of the 1ATF report’s prompts were simple statements of fact. For example, the date and time of the contact, the unit involved and its strength, can be assumed to be accurate. The 6-figure grid reference describing the location of the contact, must be assumed to be accurate within the limits of the navigation skills of the patrol commander. But some other questions required the exercise of judgment and these could theoretically introduce opportunities for exaggeration or error. For example, soldiers were required to provide an estimate of the number of rounds of ammunition they had fired in contact. It is unlikely that these estimates were exact. Soldiers in contact had other things to think about than remembering to count the shots they had fired. And in any case, counting the shots fired by an automatic weapon was not easy at the best of times. But most soldiers could recall their approximate ammunition expenditure, usually by ‘magazine’. That is, they could recall that they had fired about ‘half a magazine’ (about 10 rounds) or ‘a mag and three-quarters’ (about 35 rounds), so estimates of ammunition expenditure could be expected to be accurate to within relatively small tolerances. Examination of the ammunition expenditure estimates recorded in contact reports shows that there is a high tendency for the estimates to end in zeroes, indicating that the figures have been rounded out. Getting the estimates approximately right had benefits. If the platoon sergeant redistributed ammunition or called for an ammunition resupply, overestimation of ammunition expenditure could result in the soldier being issued with more replacement ammunition than he could fit into his magazines and an unwanted extra load of loose ammunition to carry. Furthermore, as they became more experienced, most soldiers examining enemy dead or wounded could identify with reasonable accuracy the weapons that had caused the wounds.[14] The exercise of judgement may raise questions about the accuracy of the data in some cases, but more accurate data about ammunition expenditure or the weapons causing enemy casualties is unlikely to be found.

The most potentially contentious issue is the counting of enemy dead and wounded. But, as in other issues, 1ATF took a different approach here than did its American allies. In 1ATF practice, enemy casualties were only to be counted if there remained on the battlefield a body, a significant body part without which life could not be sustained, or, in the case of wounded, a blood trail or minor body part. This method tended to underestimate enemy losses rather than exaggerate them. In some cases, intelligence gathered days or weeks later confirmed that more enemy soldiers had been killed or wounded than had been recorded at the time in the contact report.

Within 1ATF there were several factors that tended to militate against inflation of the body count figures. First, 1ATF was not responsible for the attrition strategy adopted by USMACV and, in fact, most Australian commanders thought the use of body count as a means of measuring ‘progress’ was misguided. They felt little incentive to promote the strategic goals by over-stating enemy casualty figures. Second, there was never any attempt by the higher command to evaluate the performance of Australian or New Zealand units according to their kill ratio. Third, individual NCOs and officers trained with their units before deploying to Vietnam and generally stayed with their units throughout their one-year tour. High kill ratios were not rewarded with promotion or posting to more salubrious positions. Nor were low kill ratios rewarded with relieving officers from command.[15] For 1ATF officers and NCOs – the people who compiled the contact reports – there was little incentive to exaggerate the body count.

About 2000 Combat After Action Reports (CAAR), or ‘contact reports’, were completed and filed. They can now be found in the official records collection of the Australian War Memorial, Canberra. However, individual battalions and other major combat units, particularly those deployed early in 1ATF’s tenure in Phuoc Tuy province, sometimes did not fill out CAARs for months at a time, provoking reminders to do so from Army Headquarters in Canberra. But, by the end of 1967, enough CAARs had been collected to allow some simple analysis by the Australian Army Operational Research Group (AAORG) in Australia. The first AAORG report analyzed about 100 CAARs for the period May 1966 to February 1967.[16] Other reports followed.[17]

In the 1960s, computer data processing was in its infancy and although it was applied to the analysis of at least one of the AAORG reports, it was unable to add significantly to the quality of the analysis. The resulting analysis was rather simple.[18] These early computer-aided reports produced statements of fact such as that ‘54% of contacts lasted 5 minutes or less’, or that ‘grenades were not used very often’. Such information could have been obtained simply by asking soldiers returning from operations. However, the statistical analysis had the advantage of quantifying soldiers’ subjective views and even this simple analysis produced insights deemed by Army to be ‘useful to the military planner in the formulation of policies on tactics, training, and design of weapons’.[19]

AAORG continued to add CAARs to its computer database, called ‘Tapesort’, throughout 1970, but during 1971, as 1ATF was withdrawn from South Vietnam, work ceased on the project with about 1100 contact reports collected and recorded on tape.[20] The computer tape was filed and work on analysing the combat data ceased. The focus of AAORG’s analytical effort shifted to meet the requirements of the post-war Army.

The 1 ATF Contact Database

In 1986, Dr Ross obtained a copy of the ‘Tapesort’ computer tape and, using the official records created by 1ATF, then held in the Australian War Memorial, located and added all the missing CAARs to the ‘Tapesort’ database (about 900). Ross identified a further 2000 contacts for which no CAAR had been raised. Data for these was captured from other sources such as unit war diaries and the 1ATF daily Intelligence Summary (INTSUM). Further detail was added from the official history series, unit histories and other accounts written by veterans. By 2001, the database had expanded to cover most contacts, and we now estimate that the database contains over 95% of all 1ATF recorded contacts.[21] It now records the details of 3909 contacts plus numerous mine incidents, friendly fire incidents, and other combat related events. The data captured in relation to contacts is the same as that recorded in the paper-based CAARs. Through this process, the original ‘Tapesort’ data was left far behind in terms of the number of contacts covered, the detail recorded and the accuracy of the data. The new database which emerged from this process was called the ‘1ATF Contact Database 1966-1971’ to distinguish it from its predecessor.[22]

As greater computing power became available, the automatic generation of reports such as casualty rates and ammunition usage rates became possible. New data sources were also added such as weapons performance, mine warfare, friendly fire, and civilian casualties caused by 1ATF. Although every attempt was made to collect as much information as possible for every contact, it was not possible to record every detail of the contacts for which no CAAR was created. Nevertheless, useful data such as the details of casualties, unit identification, and a brief description of the incident was collected for these.

The primary purpose of the ‘1ATF Contact Database 1966-1971’ remains the statistical analysis of 1ATF combat operations. When used for that purpose the database often requires the rearrangement or addition of information to suit the analytical purpose of the moment. But this flexibility is incompatible with the use of the database as a body of information to enable this website. Therefore, we ‘froze’ a version of the database to drive this website while a second version is retained and developed for ongoing statistical analysis. From time to time we may update the database driving the website to add more capability, but we expect such updates will be infrequent.

The Phuoc Tuy Incident Database

During the tenure of 1ATF in Phuoc Tuy province, the Detachment, 1 Divisional Intelligence Unit (Det 1 Div Int Unit) produced a daily intelligence summary (INTSUM). These recorded 1ATF combat and other incidents, but also included information relating to allied operations such as Army of the Republic of Vietnam (ARVN) or US Army contacts, airstrikes, and what was known of enemy activities such as roadblocks, mine incidents, assassinations, abductions, etc. During 1970/71, an attempt was made (by Det 1 Div Int Unit) to place all the data contained in previous INTSUMs onto a computer with the expectation that computer-aided intelligence analysis would be possible. The data was recorded as one incident per record, and this usually amounted to one or two brief sentences.

With the withdrawal of 1ATF from Vietnam, this computerized Intelligence Database, like the Contact Database before it, was filed and forgotten. It too was rescued by Dr Ross because it was a valuable research tool providing an account of every event thought by 1ATF intelligence agencies to be of intelligence interest in Phuoc Tuy Province on any given day. The database was found to be missing many months and whole days of activity, and many particular incidents were unaccountably missing when compared to the relevant paper-based daily INTSUM. Once again, these errors and omissions were painstakingly corrected by Dr Ross. The Phuoc Tuy Incident Database, as the revised and updated version is now called, provides a detailed day-by-day account of every significant event to 1ATF throughout Phuoc Tuy Province and surrounds. There are over 36,000 individual records in this database.

Because the database is now significantly different in terms of the quality and extent of the data it contains, it too was re-named to distinguish it from the database from which it was derived. The new database is now called the ‘Phuoc Tuy Incident Database 1966-1971’. It cannot be claimed that this new database records every incident in Phuoc Tuy, because the 1ATF INTSUMs did not consistently do this before 1969. But it does record every incident that 1ATF intelligence officers thought were important (i.e. the incidents they recorded in the INTSUM). But we believe it is more complete and more accurate than any other database that might attempt to cover the same period and location. It is a powerful tool for statistical and other forms of quantitative analysis. This database has been our major research tool for analyzing combat trends for the whole of Phuoc Tuy for the period during which 1ATF conducted operations in the province.

Hard at work. Staff of the Detachment,1 Division Intelligence Unit enter INTSUM data into an early model computer at Nui Dat. Although this early attempt at automated intelligence analysis seems to have produced very limited results at the time, the data collected formed the basis of the ‘Phuoc Tuy Incident Database 1966-1971’ which now supports much of our analysis.

Hard at work. Staff of the Detachment,1 Division Intelligence Unit enter INTSUM data into an early model computer at Nui Dat. Although this early attempt at automated intelligence analysis seems to have produced very limited results at the time, the data collected formed the basis of the ‘Phuoc Tuy Incident Database 1966-1971’ which now supports much of our analysis.

The Databases and the Website

The databases now allow the website to locate every 1ATF combat incident on Vietnam War-era topographical maps, Google Earth or Google maps images. For the first time it is now possible to see every one of about 4500 contacts and incidents displayed. The text describing each combat incident is drawn from the ‘1ATF Contact Database 1966-1971’ and so, reflects the official record of the incident created through the process outlined above.

The campaign in Phuoc Tuy Province was characterized by the sixteen ‘landmark’ battles (such as the battles of Long Tan, Operation Bribie, Coral, Balmoral, Binh Ba and Nui Le) set against a flat terrain of thousands of small contacts.  Historians have found the landmark battles relatively easy to deal with; large forces are involved, the ebb and flow of battle can be easily understood and often, a clear winner seems to emerge. But historians and others have found it difficult to deal with the thousands of minor contacts. By contrast they seem chaotic. They appear to conform to no pattern, they involve very small forces, and often, they seem inconsequential. Even those soldiers who fought this myriad of contacts, seeing only those they were personally involved in, often fail to detect their broader purpose. However, once displayed on this website, these smaller contacts can be better understood.


[1] Gregory A. Daddis, No Sure Victory: measuring US army effectiveness and progress in the Vietnam war, Oxford University Press, New York, 2011, p. 10.

[2] Thayer’s reports are collected in Thomas C. Thayer (ed.), A Systems Analysis View of the Vietnam War 1965-1972, OASD (SA), The Pentagon, Washington, 1975, vols. 1 – 12, For a description of the influence of these reports on the US conduct of the war see Andrew F. Krepinevich, Jr., The Army and Vietnam, The Johns Hopkins University Press, Baltimore, 1986, pp.188 – 189.

[3] Thomas C. Thayer, War Without Fronts: The American Experience in Vietnam, Westview Special Studies in Military Affairs, USA 1985.

[4] Andrew F. Krepinevich, Jr., The Army and Vietnam, Johns Hopkins University Press, Baltimore, 1986, p. 202.

[5] Guenter Lewy, America in Vietnam, Oxford University Press, New York, 1978, p. 450.

[6] M. MacLear, Vietnam: The Ten Thousand Day War, Thames/Methuen, Britain 1981, Chapter 10; James Gibson, The Perfect War: Technowar in Vietnam, Atlantic Monthly Press, USA, 1986, Chapter 5; Krepinevich, The Army and Vietnam, pp. 202 – 205; Milam, Not a Gentleman’s War, University of North Carolina Press, Chapel Hill, pp. 104-108.

[7] Bernd Greiner, War Without Fronts: The USA in Vietnam; Bodley Head London 2009; Nick Turse, Kill Anything that Moves: The Real American War in Vietnam; Picador, USA 2013.

[8] The five provinces were Quang Tri, Quang Nam, Binh Dinh and Quang Ngai, (all in I CTZ) and Tay Ninh (in III CTZ).

[9] Thayer, War Without Fronts, p. 14.

[10] The Division in Battle, Pamphlet no. 11, Counter Revolutionary Warfare, 1965, Military Board, Army Headquarters, Canberra, 1966.

[11] 1st Australian Task Force, Vietnam, Standing Operating Procedures for Operations in Vietnam, revised January 1971, Army Headquarters, Canberra, 1971, appendix 2 to annex D, Contact/incident report. All patrol commanders carried ‘field SOPs’ in a waterproof booklet. A typical example is 4RAR/NZ Field SOPs (revised November 1970) issued by LTCOL J.C. Hughes, CO 4RAR, 18 November 1970, in the authors’ possession.

[12] 1st Australian Task Force, Standing Operating Procedures, appendix 3 to annex D, Contact/incident after action report.

[13]  Australian Army Operational Research Group Interim Memorandum 2/69: Contact After-Action Reports Review of Format, August 1969. (Copy in possession of authors). For example, in response to the question ‘who saw the enemy first?’ some patrol commanders inserted the name of the individual whereas what was required was the individual’s role within the sub-unit (e.g., forward scout, machine gunner, etc). Such ambiguities were quickly eliminated, but the occasional insertion of an individual’s name suggests an earnest attention to detail on the part of the patrol commander.

[14] For example, fragmentation weapons usually caused quite different wounds to bullets. The entry wounds of 7.62mm rounds were larger than those of 5.56mm rounds. The M60 tended to produce multiple hits while the SLR produced single hits. These differences could result in a reasonably accurate estimation of the cause of wounds. Bodies with multiple wounds from different weapons and blood trails left by the wounded would require the exercise of judgment. But many soldiers knew where they had been firing and were keen, after the contact, to see the effects of their fire. They could often confirm that they had hit specific targets with their weapon even if the only evidence remaining in the killing ground was a blood trail.

[15] Krepinevich argues that these factors tended to inflate the US Army figures. See Krepinevich, The Army and Vietnam, pp.203 – 205.

[16] D. A. McCallum, Analysis of Australian Army Contacts in South Vietnam During 1966-67, AAORG Memo M29 of February 1968.

[17] R. G. Henderson, Analysis of Australian Army Contacts in South Vietnam During 1967-68, AAORG Memo M36 of July 1969 (analysing 260 CAAR) and M. Heaney, Analysis of Australian Army Contacts in South Vietnam During 1967-1969, AAORG Memo M38 of March 1970 (analysing a further 394 CAARs).

[18] Summary of AAORG Memo 29 by Major E S Holt to GSO2 ops 11 November 1968, AWM109, R723/1/1/2 part1; HQ 1ATF Reports General.

[19] AWM109, R723/1/1/2 part1; HQ 1ATF Reports General, Colonel M Bradbury (Chief of Staff) to 1 ATF, 12 November 1968.

[20] The tape (Tapesort) has 1300 contacts recorded, but about 200 of these were an error by the typist who re entered the same data. This data entry error was apparently not noticed by the scientists concerned.

[21] Some minor combat events such as long-range potshots at 1ATF patrols or other combat events that did not cause a tactical reaction from 1ATF may not have triggered the creation of a CAAR and therefore are not included in the database. However, Dr Ross estimates these to be very few in number. During, many hundreds of hours of research in the 1ATF official records no previously unrecorded contacts have been discovered. In addition to the 1ATF Contact Database, Ross also built a similar but smaller database of Commonwealth Forces contacts during Confrontation with Indonesia in North Borneo for comparative purposes. This database contains data on 850 Commonwealth forces contacts.

[22] The ‘1ATF Contact Database 1966-1971’ is derived from Infantry Battalion and other major unit war diaries and thousands of contact reports contained in scores of CAARs held by the AWM mainly in its record series AWM98 (records of Headquarters Australian Force Vietnam – Army component) and AWM103 (records of Headquarters 1st Australian Task Force). The database is a research tool and is regularly upgraded with new information as it is located.

3 Comments on “Building the 1 ATF Contact Database”

  1. Profile photo of Brian Donnellan

    Hello Derrill and the rest of the crew.
    I have been unable to find a contact that occurred on October 20th 1968 at YS558802 in which 3791207 Pte J R Tinkham was killed.
    The location information was extracted from the Commanders Diaries of the time.
    John Tinkham volunteered to carry out a task that carried a very high level of risk which sadly was realised with his death.
    As his Section Commander of that time I am determined that his story should be recorded in details of that contact.
    The unit was 12pl Delta Coy 4 RAR.

Leave a Reply