How we research

“Oral histories are tape recorded and transcribed interviews with individuals that document living memory.”

An oral history is more than a chat: it is recorded, then written out.

“Two criteria were applied to all selections: participants must have had a first-hand knowledge of events in Soap Creek Valley and they must have lived or worked in the area for a significant period of time.”

The thesis used two tests for choosing people: they saw it themselves, and they lived or worked there for a long time.

“Older participants were given priority and interviewed first whenever possible. This precedence, or bias, was considered worthwhile because the oldest members of potential interviewees held the earliest and most fragile memories of Soap Creek Valley.”

The thesis interviewed the oldest people first whenever possible, and named that choice a bias it considered worthwhile.

“Specific artifacts, including maps (Rowley 1994), aerial photographs (Glender 1994), sketches (Rawie 1994), prehistoric tools (Hanish 1994), and even broken glass, old bricks, and nails (Hindes 1996) were used to obtain informant interpretations and stimulate memories regarding specific events, times, and/or places.”

In the Soap Creek interviews, maps, air photos, sketches, old tools and even old bricks and nails were used to help people remember.

“Subjects were informed of their right to stop interviews at any time, to decline answering uncomfortable questions, and to have their own questions answered regarding interviewer’s motives, qualifications, and/or interests in the study.”

Consent was part of the method: people were told they could stop, skip questions, and ask why the interviewer wanted to know.

“Copies of transcriptions were read and amended while listening to the recordings (“audited”) by people present during the interview (Baum 1985). The interviewees were often the best people for this step because they were most aware of who and what was talked about, general accuracy, how names were spelled, what they had intended to say, and what had been left unsaid.”

The typed transcript is read against the recording by people who were at the interview, often the speaker: names, spellings and what they meant.

“Criteria is often established through methods which corroborates the information obtained from the interviewee (primary source) with information derived from alternate or multiple credible sources, such as signed documents, photographs, news reports, and scientific research.”

Memories are checked against other sources: signed papers, photos, newspapers and research.

“Aerial photographs are, in effect, highly detailed and time-specific maps of landscape patterns. Before aerial photography came into widespread use in the 1930s, people had to use maps or visit scenic vistas to consider landscape scale patterns of vegetation and human constructs.”

The dissertation treats an air photo as a map with a date on it, not just a picture.

“Aerial photography was first implemented in Oregon in the 1920s by such organizations as the USDA Forest Service, the Oregon State Highway Commission, and the US Soil Conservation Service.”

The oldest Oregon air photos date from the 1920s and were made by federal and state agencies.

“The University of Oregon has the most complete collection of aerial photographs available in Oregon, but businesses such as W.A.C. (Western Aerial Contractors) in Eugene and organizations such as the Forest Service maintain extensive collections that allow for the purchase and/or use of individual "site-and time-specific" sequences.”

As of 2003, the University of Oregon held the most complete set; the directory shows how copies are ordered today.

“General Land Office (GLO) subdivision and Donation Land Claim (DLC) boundary lines dating to the early 1850s are clearly apparent a century later as the straight courses of roads, fence lines, crop borders, and logging boundaries.”

In a 1948 photo of Alsea, survey and claim lines from the early 1850s still show as straight roads, fences and field edges. A dead-straight fence on your land may be following an original survey or claim line too.

“Fenced and plowed bottomland prairies, grassy meadows, ridgeline trails, hillside scatterings, and other pre-contact landscape patterns can also be easily discerned.”

Older patterns showed through in that photo as well: bottomland prairies, meadows, ridgeline trails and scattered trees on hillsides.

“[Compare the persistent patterns of trails and vegetation shown on this aerial photograph and on Map D.04, with GLO surveys of the same area, performed from 1856-1897.]”

A note under one 1966 photograph in the dissertation asks the reader to compare it with the original surveys of the same ground.

“GLO land survey records have been long accepted as useful measures of past environmental conditions (Whitney and DeCant 2001: 147). These records have specific advantages over other early historical sources of information in being a definite sampling of vegetation, written on location, and in accordance to a predetermined plan (Bourdo 1956: 755).”

Survey notes were written on the spot, by a set plan, which gives them advantages over other early sources such as travelers' diaries.

“Before 1851, these lines were absent from the environment. Since that time they have served as an exacting index to specific trees, points, locations and cultural artifacts on specific days in time.”

Because later property and section lines grew from those survey lines, a note about a fir at a corner can be put back on the ground, on a known day.

“Surveyors regularly noted specific types and species of plants and whether forest fires had entered the area, or not.”

The notes tell what grew along the lines, and whether it had burned.

“Bearing trees were carefully marked, identified as to species and general condition, measured for diameter, and located by distance and compass bearing from each corner.”

Bearing trees are listed with species, width, and distance and direction from the corner, so you know what grew right there.

“Because DLCs often had six, eight, or more corners, and because they included trees well-within the boundaries of section lines, these surveys provide us with far more exacting information--particularly near the center of the sections in which they are located--than surveys of sections that do not include DLCs (see maps 3.03 and 3.04).”

Where a Donation Land Claim crossed a section, its survey added corners and trees inside the section, so there is more detail.

“Subdivision surveys of unclaimed lands--typically steep, forested, and/or burned tracts that remained in the public domain because they were unsuitable for most agriculture or grazing purposes--were often delayed for a period of twenty to fifty years after initial township boundaries had been established.”

Steep or timbered ground was often surveyed twenty to fifty years after the township lines, so survey dates can differ within one township.

“This project is based on traditional archival research methods, followed by physical transcription of data to maps, and then location and documentation in the field.”

In that project the order was archives first, then maps, then the field.

“Fig. 5 shows another type of map used in this research: a geological map from the late 1890s, first printed in 1903 and apparently showing pre-automobile roads and trails to a very accurate degree at that time.”

Beyond the obvious maps: an 1890s geology map appeared to record the roads and trails of its day accurately.

“Note the differences in Port Orford between the maps, but also the similarity in road and trail alignments between 1875 and 1941.”

Maps of the same township from 1875 and 1941 show the town changing while road and trail lines stayed much the same.

“The thin red lines with yellow high-lighter were transcribed as possible foot-trail routes, with solid lines depicting existing road and trail surfaces, and dotted lines showing predicted (but undocumented) or documented (but no longer depicted) trail locations.”

The field maps drew what was known differently from what was predicted: solid for routes still on the ground, dotted for predicted or no-longer-mapped ones.

“One of the more helpful conditions during field work on this project was the combination of “clear skies and clearcuts” along the ridgeline roads we drove. These allowed for wonderful perspectives of the surrounding landscape, including likely, possible, and improbable trail locations.”

Clear skies and clearcuts along the ridgeline roads gave long views, and possible routes were sorted into likely, possible and improbable.

“Following field work, the “pinkish” highlighter in Fig. 7 was used to show the final editorial decisions made by the authors following much discussion and consideration of the possible alternative routes and the documentation that supported each alternative.”

Where the evidence pointed more than one way, the authors discussed the alternatives and the documents behind each, then marked their final choices in a different color.

“Each time a specific reference was located, it was scanned or photographed and entered into an Excel file database; then printed, annotated with the name of the surveyor and date the observations were written, and filed in a folder labeled with the individual township in which the notes were made.”

Each find was copied, labeled with the surveyor's name and the date, and filed in a folder for its township.

“Research for this project was conducted in four basic steps: archival research and literature review, predictive field map construction, “ground-truthing” the predictive map (field research), and synthesis. These are not discrete steps in any sense, and usually two or three were being undertaken at the same time throughout the course of this project.”

Four steps, begun in that order (read, predict, check on the ground, put it together), but two or three were usually under way at once.

“Research for this project was conducted using the “method of multiple hypotheses,” first described by Chamberlin in 1890 (Chamberlin 1965) as: "the effort is to bring up into view every rational explanation of new phenomena, and to develop every tangible hypothesis respecting their cause and history." This method involves approaching a specific research question with all information available from relevant disciplines; making a reasoned answer to the question based on the weight of the gathered evidence; and then posing more specific questions that can then be asked in order to better consider the original hypothesis (see Chapter VIII).”

The method kept every reasonable explanation on the table, answered from the weight of the evidence, then asked the next, sharper question.

“In oral history research, practitioners often use “triangulation” to establish the credibility of informants, and the reliability of their information (Hoffman 1996; Zybach 1999): if three (or more) sources of information can be shown to be in agreement, then the resulting conclusion can generally be assumed a “fact.””

The Tiller report described triangulation from oral history research: when three or more sources agree, the conclusion can generally be assumed a fact.

“Following (but not discontinuing) background investigations accomplished by literature review, discussions with knowledgeable individuals, and archival research, predictive maps were constructed to guide subsequent field work activities.”

Before the field work, maps were drawn of where the records said evidence should be.

“The combination of these three elements -- GLO field survey notes (and plats), Osborne panoramic photographs, and aerial photographs – before the post-WW II advent of major road building and logging activities, provides a highly detailed and accurate depiction of forest conditions in transition from precontact time to now.”

Three records from before the big postwar road-building and logging years (survey notes, fire-lookout panoramas and air photos) gave the Tiller study a detailed picture together.

“There is an anomaly along the North-South Trail and the ca. 1800 vegetation patterns that are shown, however, which relied on field observations and measurements in order to be better understood.”

Where the map layers disagreed along one trail, field observations and measurements were needed to understand the difference.

“Brief answers are also supplied for each question, based on the current findings of this study, but with the idea that entirely different answers might be forthcoming with additional data or alternative perspectives.”

The study gave answers and held them open: new data might change them.