So, the essential question for improving our restoration decisions is: which of these many identities represents the car most accurately? Which one representation best tells its dynamic and changing story composed over so many owners, so many miles, and so many years? How do we give proper weight to these successive developments? Everything we might do creates, at best, a partial representation of a brief period, a single moment, amidst the car’s extended and evolving life story. Of course, the biggest fiction of all is the actual restoration itself. Our restored car’s bright, shiny simulacrum of newness misleads all who encounter it about its most essential property: its relationship to time. The restored car is not only a thing from the past but is a thing of the past. Necessarily, the car is full of evidence of passing time in the form of marks, damage, wear, and deterioration that speaks to the machine’s experience in the world.

When we then proceed to restore an automobile, the very nature of our restoration necessarily privileges the chosen point in the automobile’s existence that is being portrayed over its many other and, perhaps, equally valid identities. Identities most commonly selected for representation in a restoration are: the automobile as new; the automobile at a specific time in its existence (provided documentation of that moment is available); and finally, the automobile as a chimera made of its combined configurations at various times in its existence.

One such example of this latter restoration might be an important factory “test” automobile that received several color changes over its life as well as the replacement of its original engine with a special prototype engine having less weight and more power. Thirty years later, the restoration involves returning the car to its original color, restoring the interior and so on to “as new” configuration, but retaining the later incorrect engine because of its performance benefits and historical importance. This is a restoration that mixes attributes of the same car from different periods. While it is technically “incorrect,” it fairly combines two significant points in the car’s prior life, thereby giving a more “complete” picture of its trajectory through time than a chronologically pure approach might. And, of course, the restoration is wholly fictional, though perhaps more representative. Knowledgeable people will differ.

Analogous to writing history, which is the selected recounting of the past, restoration involves selectively recreating the physical past of an artifact. In both cases, the past whether historiography, the academic recounting of the past based on texts, or restoration, the recreation of the past based on physical matter, is lost, unknowable, and inaccessible. For example, we might know the scratch on the door came from parking at the movies to see a specific film (though the date may be lost to us, and wholly unrecoverable), but the origin of the dent in the rear quarter is a forgotten mystery.

If we are writing history, we can only recreate what we think happened based on accepted historiological procedures, likewise, if we are undertaking a restoration, we can only recreate how we think the artifact looked and operated in the past based on archaeological practices: the examination of other similar examples, or by consulting historical images of the actual car, and so forth. In both cases, perfectly comprehensive and complete evidence is never available. And so, both processes are, by definition, evidence-based acts of imagination that result in conjecture and hypothesis. And that means fiction. The point is not that restoration’s fictional property is a bad thing, just that it is an unavoidable property of working with things of the past.

Unavoidable fiction may appear as daunting and distressing stuff calling into question our best restoration research. But take heart, there are some things we can think about and use in working on that meaningful old car. The fascinating thing about relics from the past is that they often contain evidence about their time in the world from making to the present day. When these data are detectable, they cause us to propose narratives, which, if nothing else, bring us up against important human questions of the time.

For example, our restoration research of the 1919 Indianapolis Ballot 1003 shows very different remedies for common problems among the team’s four cars. The mechanic’s grab handle behind the driver’s right shoulder on the two cars for which we have clear photographs shows unrelated design and fabrication methods. The first is a blacksmith’s solution made from round bar stock shaped with multiple right-angle bends, the second is a leather wrapped, semi-circular handle made from pipe. Based on images showing all the riding mechanics with their right arms behind the drivers, it appears that security grips were needed on each car, a common problem. However, instead of addressing the issue collectively and making four handles to a common design that could be efficiently “production-lined,” each mechanic made his own in a design “vacuum.” What might we conjecture about the function of the team from our viewpoint one hundred years later?

Well, we know the Ballot Indy initiative was conducted under the most obsessive secrecy. Consequently, for security reasons, the drivers were only hired two weeks before the cars were to leave Paris for Indianapolis. It seems logical that the contracted drivers would be asked to hire their own riding mechanics as the partnership chemistry would be important. Indeed, we might hazard those professional drivers had customary mechanics with whom they worked, much like today’s pro-tour golfers and their caddies. Consequently, the driver/mechanic partnership was infinitely more important than the group dynamics of the hastily formed Ballot team. Essentially, each mechanic supported his driver to the exclusion of the team’s other pairings. The variations in handles, windshields, oil pump locations, tachometer configuration, even visible finishes on the cars on race day morning speaks to four pairs of partners, not to one team of four driver/mechanics.

Of course, we can’t know this beyond doubt. History is silent. No texts treating this issue have survived from the 1919 Indy race. Instead, we rely on conjecture and hypothesis. But such ruminations open the door to a very compelling and very different picture of the Ballot team at Indy.

Top photo: IMS Photo Institute.