It was one of those gorgeous Salvadoran mornings: sunshine like soda water, an earlfj mist burnt away, the rising heat sharp and dry. There was a slight breeze.

Following a 36-hourmarch, much of it under sniper fire, the Lenca infantry battalion of the Salvadoran Army was bivouacked atop a hill called Ocotepeque in the northeastern province of Morazan. There were five journalists along for the trip-two Mexicans, an Argentine, an American and me-an army sweep north toward the Honduran border, through territory controlled by the People’s Revolutionary Army, one of five Salvadoran guerrilla forces. This morning we were simply hanging around, waiting until the army got itself reorganized so we could head out, aiming for a guerrilla-controlled town called Corinto, nestled in a valley to the north.

Presently, a white minibus bounced up along the long rock trail that scales the southern wall of Ocotepeque. Inside was a Canadian Broadcasting Corporation television crew doing a documentary for The Journal. Lieutenant Colonel jorge Adalberto Cruz-sort of a Latin version of Peter Lorre, decked out in battle fatigues-led us out along the northern edge of the hill to a rocky cliff overlooking northern Morazan. In the distance, mainly obscured by brown, sun-burnt hills, we could just make out the white chalk and red clay smudges of Corin to.

Cruz had one of those little retractable pointers that normally he wore tucked in the breast pocket of his fatigues, like a ball-point pen. Now he produced it, played it out to its full length and used it to indicate points of interest on a plastic-laminated map. He stood on a ledge of rock above the journalists and periodically waved off to the north as he explained how the rest of the counterinsurgency sweep would be conducted. Below him, the CBC cameraman was huddled over his camera, almost at Cruz’s feet, shooting up at the officer to get a dramatic effect. The soundman held the boom microphone in the air a yard or so in front of Cruz, who declaimed his imminent tactical plans as though he were running for office.

Eventually, however, the cameraman and the soundman, neither of whom understood Spanish, appeared to lose interest. The microphone swung away and the camera stopped rolling and was shifted into a neutral position. Cruz, a battle-hardened counterinsurgency officer wi th a pin t or two of blood on his hands-and who would later be behind at least one rumored coup plot-promptly faltered and then halted in mid-sentence. He gazed down at the TV crew, who at this point were chatting between themselves in English, oblivious to the seasoned Salvadoran warlord looming above them. Cruz’s grizzled face seemed to drop. His hand, clutching his pointer, fell to his side. He seemed utterly lost. “Should I go on?” he asked in Spanish-uncertainly, just as though he were requesting permission to leave the room. “Should I go on?”

That’s television. I don’t believe that the print media are able to instill quite so much respect in their interlocutors (nor, on every occasion, is TV). Still, that morning in Morazan almost three years ago reminds me now of the press’s inherent ability to direct what it ostensibly is trying to trace: the real world. During almost five years as a Globe and Mail correspondent based in Latin America, I periodically had examples of the press’s real or assumed power thrown back at me; they were always a surprise and often a shock.

On occasional trips home to Canada, I would sometimes hear things I had written-judgments I had made or conclusions I had drawn-being quoted in conversation. Because they had appeared in print (and presumably because they tended to corroborate what the person quoting them already believed), my stories were trotted out as persuasive proof that, say, the Salvadoran guerrillas were militarily superior to the army, or that the Guatemalan refugees in Mexico had no direct links to their country’s guerrillas, or that the Sandinistas in Nicaragua are good and God- fearing men.

I was always taken aback, because I knew how much simplification, hearsay and conjecture can find their way into a 1,200-word article on an excruciatingly complicated theme. Of course, we do the best we can. We weigh our sources and try to second-guess their information and their opinions. But the craziest things can happen.

Once, during the 1982 Falklands Malvinas war between Britain and Argentina, I put together an article assessing the political interplay of Argentina’s three armed forces and its impact on the way the ruling junta came to its decisions. At one point, I observed that the air force commander, General Basilio Lami Dozo, was considered politically weak and did not seem to play an important role in essentially political issues. One afternoon some weeks later, I happened to be at the Canadian embassy in Buenos Aires and I bumped into the Canadian ambassador. I asked him his opinion about some question of the moment the war was still on-and in the course of his reply he mentioned, just by the way, that General Lami Dozowas not a powerful player politically. Then the ambassador hesitated, as though recalling a dim memory. He glanced at me. He said: “But it was you who wrote that, wasn’t it?” So it was. Not only that, but the unnamed source I had been quoting when I wrote it was none other than the Canadian embassy’s political counsellor, who, of course, had merely been expressing his own opinion at the time.

The most mind-wrenching story I had to cover in Latin America, however, was unquestionably Nicaragua. Looking back now on my repeated and often lengthy visits there from 1981 to 1985, I think I can see a little more clearly than perhaps I did at the time just why that small, beleaguered land was such agony to analyse and write about. There were probably countless reasons, but a lot of them boil down to the plain fact that most of the patterns I consciously or unconsciously used when shaping stories about other Latin American countries simply ceased to make sense in revolutionary Nicaragua. One common journalistic pattern is to divide the story (or the world) into good guys and bad guys, taking care to tarnish the good guys slightly and to redeem the bad guys just a little, for verisimilitude. That worked fine in Guatemala, where the good guys were barefoot, dirt-poor, helpless and desperate, and where the bad guys carried Galil automatic rifles, wore camouflage khaki and waded through the Indian hamlets like a Praetorian guard. For a time, at least, Guatemala was as close as a story gets to being black and white.

But in Nicaragua such easy conventions did not seem to help. Nor could I simply fall back on another standard journalistic format: hewing a middle line. In Nicaragua there really isn’t one. The nine Sandinista commanders are either angels of a new dawn-or hounds of a Stalinist-inspired hell. The U.S.-backed contras are either heroic freedom-fighters-or blood-thirsty mercenaries. For a Canadian journalist steeped in that cozy western liberal convention that the truth lies somewhere slightly to the left or to the right of the midpoint between the two extremes, it was more than merely jolting to be faced suddenly with a story in which there really is no middle ground, and where I found myself being tossed relentlessly from one side to the other.

Another common journalistic habit-a way of positioning ourselves on a story-is to work out where and what the power structures are and then to shape our stories in reaction to them. We do it all the time. We bounce our questions against the people in power-from the government in Ottawa to the majority on city council. We examine what they are doing and why, and we hold them to account. That can be difficult enough, particularly in countries where the power structures are either very weak or extremely strong and not well disposed to answering questions. But in Nicaragua, it was especially complex.

Since 1979, the local source of power in the land of Sandino has unquestionably been the Sandinistas. But the real power in the region, the power that can and no doubt will determine Nicaragua’s future, resides in Washingtonand in this case it is not united with but diametrically opposed to the local power. That constant tension pervades all elements of Nicaraguan life and it makes covering the country a neverending scramble of confusions, contradictions, conflicting allegiances and confounded sentiments.

It was always a kind of agony to try to chop all those competing elements down into a 1,200-word feature, news story or analysis piece that made internal sense and somehow got Nicaragua “right .” Sooner or later, though, I had to haul out my laptop word processor, hold my nose and plunge in. When I was off the mark-as inevitably I was-1 simply tried to compensate the next time out. And that, I suppose, is what makes journalism the rough, unfinished craft it is. We never really do get it right. We just keep trying.

Oakland Ross was The Globe and Mail’s Latin American correspondent from 1981 to 1985. He is now the Globe’s Montreal bureau chief.